System Administration Guide:

Volume 1

Adaptive Server Enterprise

15.7

DOCUMENT ID: DC31654-01-1570-01

LAST REVISED: September 2011
Copyright 2011 by Sybase, Inc. All rights reserved.
This publication pertains to Sybase software and to any subsequent release until otherwise indicated in new editions or technical notes.
Information in this document is subject to change without notice. The software described herein is furnished under a license agreement,
and it may be used or copied only in accordance with the terms of that agreement.
To order additional documents, U.S. and Canadian customers should call Customer Fulfillment at (800) 685-8225, fax (617) 229-9845.
Customers in other countries with a U.S. license agreement may contact Customer Fulfillment via the above fax number. All other
international customers should contact their Sybase subsidiary or local distributor. Upgrades are provided only at regularly scheduled
software release dates. No part of this publication may be reproduced, transmitted, or translated in any form or by any means, electronic,
mechanical, manual, optical, or otherwise, without the prior written permission of Sybase, Inc.
Sybase trademarks can be viewed at the Sybase trademarks page at http://www.sybase.com/detail?id=1011207. Sybase and the marks listed
are trademarks of Sybase, Inc. indicates registration in the United States of America.
SAP and other SAP products and services mentioned herein as well as their respective logos are trademarks or registered trademarks of
SAP AG in Germany and in several other countries all over the world.
Java and all Java-based marks are trademarks or registered trademarks of Sun Microsystems, Inc. in the U.S. and other countries.
Unicode and the Unicode Logo are registered trademarks of Unicode, Inc.
All other company and product names mentioned may be trademarks of the respective companies with which they are associated.
Use, duplication, or disclosure by the government is subject to the restrictions set forth in subparagraph (c)(1)(ii) of DFARS 52.227-7013
for the DOD and as set forth in FAR 52.227-19(a)-(d) for civilian agencies.
Sybase, Inc., One Sybase Drive, Dublin, CA 94568.

Contents

CHAPTER 1

Overview of System Administration.............................................. 1

Adaptive Server administration tasks ............................................... 1
Roles required for system administration tasks......................... 2
Using isql to perform system administration tasks .................... 5
Using Sybase Central for system administration tasks ............. 6
System tables................................................................................... 7
Querying the system tables ....................................................... 8
Keys in system tables................................................................ 9
Updating system tables ............................................................. 9
System procedures ........................................................................ 10
Using system procedures ........................................................ 10
System procedure tables......................................................... 11
Creating system procedures ................................................... 12
System extended stored procedures ............................................. 12
Creating system ESPs ............................................................ 13
Logging error messages ................................................................ 13
Connecting to Adaptive Server ...................................................... 14
The interfaces file .................................................................... 14
Directory services .................................................................... 15
LDAP as a directory service .................................................... 16
Security features available in Adaptive Server............................... 18

CHAPTER 2

System and Optional Databases .................................................

Overview of system databases ......................................................
master database ............................................................................
Controlling object creation in master .......................................
Backing up master and keeping copies of system tables........
model database..............................................................................
sybsystemprocs database..............................................................
tempdb database ...........................................................................
Creating temporary tables .......................................................
sybsecurity database .....................................................................
sybsystemdb database ..................................................................
sybmgmtdb database .....................................................................

System Administration Guide: Volume 1

21
21
23
24
24
25
26
26
27
28
28
28

iii

Contents

pubs2 and pubs3 sample databases .............................................

Maintaining the sample databases..........................................
pubs2 image data....................................................................
dbccdb database............................................................................
sybdiag database ...........................................................................
Determining the version of the installation scripts..........................

29
29
29
30
30
30

CHAPTER 3

System Administration for Beginners ......................................... 33

Logical page sizes.......................................................................... 33
Using test servers ........................................................................ 34
Planning resources.................................................................. 34
Achieving performance goals .................................................. 35
Considerations when installing Sybase products........................... 35
Check product compatibility .................................................... 35
Install or upgrade Adaptive Server .......................................... 35
Install additional third-party software....................................... 36
Configure and test client connections ..................................... 36
Allocating physical resources ......................................................... 37
Dedicated versus shared servers............................................ 37
Decision-support and OLTP applications ................................ 38
Advance resource planning..................................................... 38
Operating system configuration............................................... 39
Backup and recovery ..................................................................... 39
Keep up-to-date backups of master ........................................ 40
Automate backup procedures ................................................. 41
Verify data consistency before backing up a database ........... 41
Monitor the log size ................................................................. 42
Ongoing maintenance and troubleshooting ................................... 42
Starting and stopping Adaptive Server.................................... 43
Viewing and pruning the error log ........................................... 43
Keeping records ............................................................................. 43
Contact information ................................................................. 44
Configuration information ........................................................ 44
Maintenance schedules........................................................... 44
System information.................................................................. 45
Disaster recovery plan............................................................. 45
Additional resources....................................................................... 46

CHAPTER 4

Managing and Monitoring Adaptive Server................................. 47

Sybase Control Center for Adaptive Server ................................... 47
Adaptive Server Sybase Central Plug-in........................................ 48
Using the Adaptive Server Plug-in .......................................... 49
Starting and stopping Sybase Central..................................... 50

iv

Adaptive Server Enterprise

Contents

Registering Adaptive Server Plug-in ....................................... 50

Performing common tasks....................................................... 51
Using Interactive SQL ............................................................. 58

CHAPTER 5

Setting Configuration Parameters............................................... 61

Overview ........................................................................................ 61
The Adaptive Server configuration file .................................... 62
Modifying configuration parameters ........................................ 62
Required roles for modifying configuration parameters........... 62
Unit specification using sp_configure ...................................... 64
Global versus session settings................................................ 64
Getting help information on configuration parameters ............ 65
Using sp_configure ........................................................................ 66
Syntax elements...................................................................... 67
Using sp_configure with a configuration file ............................ 68
The parameter hierarchy ......................................................... 72
User-defined subsets of the parameter hierarchy: display levels 74
Performance tuning with sp_configure and sp_sysmon .......... 76
Using configuration parameters in a clustered environment ... 76
sp_configure output........................................................................ 77
Named cache configuration parameters ........................................ 79
sysconfigures and syscurconfigs tables ......................................... 79
Querying syscurconfigs and sysconfigures: an example ........ 80
Configuration parameters............................................................... 80
Alphabetical listing of configuration parameters...................... 81

CHAPTER 6

Overview of Disk Resource Issues ...........................................

Device allocation and object placement.......................................
Commands for managing disk resources.....................................
Considerations in storage management decisions ......................
Recovery ...............................................................................
Performance ..........................................................................
Status and defaults at installation time.........................................
System tables that manage storage.............................................
The sysdevices table .............................................................
The sysusages table .............................................................
The syssegments table .........................................................
The sysindexes table.............................................................
The syspartitions table ..........................................................

CHAPTER 7

Managing Remote Servers ......................................................... 293

Overview ...................................................................................... 293

System Administration Guide: Volume 1

283
283
284
286
286
287
287
288
289
290
291
291
291

Contents

Managing remote servers ............................................................

Adding a remote server .........................................................
Managing remote server names............................................
Setting server connection options .........................................
Getting information about servers .........................................
Dropping remote servers.......................................................
Adding remote logins ...................................................................
Mapping users server IDs.....................................................
Mapping remote logins to particular local names ..................
Mapping all remote logins to one local name ........................
Keeping remote login names for local servers ......................
Example of remote user login mapping.................................
Password checking for remote users ...........................................
Effects of using the untrusted mode......................................
Getting information about remote logins ......................................
Configuration parameters for remote logins.................................

294
295
296
297
299
299
300
300
301
301
302
302
304
304
305
305

CHAPTER 8

Initializing Database Devices...................................................... 307

Database devices......................................................................... 307
Using the disk init command ........................................................ 308
disk init syntax.............................................................................. 308
Specifying a logical device name .......................................... 308
Specifying a physical device name ....................................... 309
Choosing a device number.................................................... 309
Specifying the device size ..................................................... 310
Specifying the dsync setting (optional).................................. 311
Using directio to bypass operating system buffer.................. 313
Other optional parameters for disk init .................................. 314
Getting information about devices................................................ 314
Dropping devices ......................................................................... 316
Designating default devices ......................................................... 316
Choosing default and nondefault devices ............................. 317
Increasing the size of devices with disk resize............................. 318
Insufficient disk space ........................................................... 319

CHAPTER 9

Setting Database Options ...........................................................

Using the sp_dboption procedure ................................................
Database option descriptions.......................................................
Viewing the options on a database ..............................................
Displaying currently set switches with sysoptions........................

CHAPTER 10

vi

321
321
322
323
324

Configuring Character Sets, Sort Orders, and Languages...... 327

Adaptive Server Enterprise

Contents

Understanding internationalization and localization ..................... 327

Advantages of internationalized systems..................................... 328
A sample internationalized system............................................... 329
Elements of an internationalized system...................................... 331
Selecting the character set for your server .................................. 331
Unicode ................................................................................. 334
Selecting the server default character set ............................. 338
Selecting the sort order ................................................................ 341
Using sort orders ................................................................... 342
Different types of sort orders ................................................. 342
Selecting the default sort order ............................................. 343
Selecting a language for system messages................................. 349
Setting up your server: examples................................................. 350
A Spanish-version server ...................................................... 350
A U.S.-based company in Japan........................................... 351
A Japan-based company with multinational clients............... 352
Changing the character set, sort order, or message language .... 352
Changing the default character set ....................................... 353
Changing the sort order with a resources file ........................ 354
Changing the default sort order............................................. 355
Reconfiguring the character set, sort order, or message language
355
Unicode examples................................................................. 356
Preliminary steps................................................................... 357
Setting the users default language ....................................... 359
Recovery after reconfiguration .............................................. 359
Handling suspect partitions ................................................... 362
Installing date strings for unsupported languages........................ 363
Server versus client date interpretation................................. 364
Internationalization and localization files...................................... 365
Types of internationalization files .......................................... 365
Character sets directory structure ......................................... 365
Types of localization files ...................................................... 366
Software messages directory structure ................................. 367
Message languages and global variables ............................. 368

CHAPTER 11

Configuring Client/Server Character Set Conversions ...........

Character set conversion .............................................................
Supported character set conversions...........................................
Conversion for native character sets.....................................
Conversion in a Unicode system...........................................
Adaptive Server direct conversions.......................................
Unicode conversions .............................................................
Choosing a conversion type .........................................................

System Administration Guide: Volume 1

369
369
370
370
371
372
372
374
vii

Contents

CHAPTER 12

Non-Unicode client/server systems.......................................

Unicode client/server systems...............................................
Configuring the server ...........................................................
Enabling and disabling character set conversion.........................
Characters that cannot be converted ....................................
Error handling in character set conversion...................................
Conversions and changes to data lengths ...................................
Configuring your system and application ..............................
Specifying the character set for utility programs ..........................
Display and file character set command line options ............

374
374
375
376
377
377
378
378
379
379

Diagnosing System Problems ....................................................

How Adaptive Server uses error messages.................................
Error log format .....................................................................
Error messages and message numbers ...............................
Variables in error message text.............................................
Adaptive Server error logging ......................................................
Severity levels .......................................................................
Severity levels 10 18 ..........................................................
Severity levels 19 26 ..........................................................
Reporting errors ....................................................................
Backup Server error logging ........................................................
Killing processes ..........................................................................
Using kill with statusonly .......................................................
Using sp_lock to examine blocking processes......................
Housekeeper functionality............................................................
Housekeeper wash................................................................
Housekeeper chores .............................................................
Housekeeper garbage collection...........................................
Configuring enable housekeeper GC ....................................
Shutting down servers..................................................................
Shutting down Adaptive Server .............................................
Shutting down a Backup Server ............................................
Learning about known problems ..................................................

381
382
384
385
385
386
387
390
392
393
394
397
398
398
399
399
399
400
401
402
402
403

381

Index ........................................................................................................................................... 405

viii

Adaptive Server Enterprise

CH A PTE R

Overview of System
Administration

This chapter introduces the basic topics of Adaptive Server system

administration.
Topic
Adaptive Server administration tasks
System tables

Page
1
7

System procedures
System extended stored procedures

10
12

Logging error messages

Connecting to Adaptive Server

13
14

Security features available in Adaptive Server

18

Adaptive Server administration tasks

Adaptive Server administration tasks include:

Installing Adaptive Server and Backup Server

Creating and managing Adaptive Server login accounts

Granting roles and permissions to Adaptive Server users

Managing and monitoring the use of disk space, memory, and

connections

Backing up and restoring databases

Diagnosing system problems

Configuring Adaptive Server to achieve the best performance

In addition, system administrators may assist with certain database design

tasks that overlap with the work of application designers, such as
enforcing integrity standards.

System Administration Guide: Volume 1

Adaptive Server administration tasks

Although a system administrator generally concentrates on tasks that are

independent of the applications running on Adaptive Server, he or she is likely
to have the best overview of all applications. For this reason, a system
administrator can advise application designers about the data that already
exists on Adaptive Server, make recommendations about standardizing data
definitions across applications, and so on.
However, the distinction between what is specific to an application is
sometimes unclear. Owners of user databases may consult certain sections of
this book. Similarly, system administrators and database owners will use the
Transact-SQL Users Guide (especially the chapters on data definition, stored
procedures, and triggers). Both system administrators and application
designers will use the Performance and Tuning Series.

Roles required for system administration tasks

Many of the commands and procedures discussed in this manual require the
system administrator or system security officer role. Other sections in this
manual are relevant to database owners.
Various security-related, administrative, and operational tasks are grouped into
the following user roles:

system administrator by default, the system administrator (sa) is

assigned these roles:

sa_role

sso_role

oper_role

sybase_ts_role

The system administrators tasks include:

Managing disk storage

Monitoring the Adaptive Server automatic recovery procedure

Fine-tuning Adaptive Server by changing configurable system

parameters

Diagnosing and reporting system problems

Backing up and loading databases

Modifying and dropping server login accounts

Adaptive Server Enterprise

CHAPTER 1

Overview of System Administration

Granting and revoking the system administrator role

Granting permissions to Adaptive Server users

Creating user databases and granting ownership of them

Setting up groups, which can be used for granting and revoking

permissions

System security officer performs security-related tasks, such as:

Creating server login accounts, which includes assigning initial

passwords

Changing the password of any account

Granting and revoking the system security officer and operator roles

Creating, granting, and revoking user-defined roles

Granting the capability to impersonate another user throughout the

server

Setting the password expiration interval

Setting up Adaptive Server to use network-based security services

Managing the audit system

Operator backs up and loads databases on a server-wide basis. The

operator role allows a single user to use the dump database, dump
transaction, load database, and load transaction commands to back up and
restore all databases on a server without having to be the owner of each
one. These operations can be performed for an individual database by the
database owner or by a system administrator. However, an operator can
perform them for any database.

These roles provide individual accountability for users performing operational

and administrative tasks. Their actions can be audited and attributed to them.
A system administrator operates outside the discretionary access control
(DAC) protection system; that is, when a system administrator accesses
objects, Adaptive Server does not check the DAC permissions.
In addition, two kinds of object owners have special status because of the
objects they own. These ownership types are:

Database owner

Database object owner

System Administration Guide: Volume 1

Adaptive Server administration tasks

Database owner
The database owner is the creator of a database or someone to whom database
ownership has been transferred. A system administrator can use the grant
command to grant users the authority to create databases.
A database owner logs in to Adaptive Server using his or her assigned login
name and password, and has the dbo account. When this user logs in to
databases they did not create, this user is known by his or her regular user
name.
A database owner can:

Run the system procedure sp_adduser to allow other Adaptive Server

users access to the database

Use the grant command to give other users permission to create objects
and execute commands within the database

Adding users to databases is discussed in Chapter 3, Managing Adaptive

Server Logins and Database Users in the Security Administration Guide.
Granting permissions to users is discussed in Chapter 6, Managing User
Permissions.
The database owner does not automatically receive permissions on objects
owned by other users. However, a database owner can temporarily assume the
permissions of other users in the database at any time by using the setuser
command. Using a combination of the setuser and grant commands, the
database owner can acquire permissions on any object in the database.
Note Because the database owner role is so powerful, the system administrator

should plan carefully who should own databases in the server. The system
security officer should consider auditing the database activity of all database
owners.

Database object owner

A database object owner is a user who creates a database object. Database
objects include tables, indexes, views, defaults, triggers, rules, constraints, and
procedures. Before a user can create a database object, the database owner
must grant the user permission to create objects of a particular type. There is
no special login name or password for a database object owner.
The database object owner creates an object using the appropriate create
statement, and then grants permission to other users.

Adaptive Server Enterprise

CHAPTER 1

Overview of System Administration

The creator of a database object is automatically granted all permissions on that

object. The system administrator also has all permissions on the object. The
owner of an object must explicitly grant permissions to other users before they
can access the object. Even the database owner cannot use an object directly
unless the object owner grants him or her the appropriate permission. However,
the database owner can always use the setuser command to impersonate any
other user in the database, including the object owner.
Note When a database object is owned by someone other than the database

owner, the user (including a system administrator) must qualify the name of
that object with the object owners nameownername.objectnameto access
the object. If an object or a procedure must be accessed by a large number of
users, particularly in ad hoc queries, having these objects owned by dbo
greatly simplifies access.

Using isql to perform system administration tasks

This book assumes that you use the commmand line utility isql to perform the
system administration tasks described in this guide. This section provides some
basic information about using isql. For complete information, see the Utility
Guide.
You can also use the graphic tool Sybase Central to perform many of the
tasks described in this book, as described in Using Sybase Central for system
administration tasks on page 6.

Starting isql
To start isql on most platforms, type this command at an operating system
prompt, where username is the system administrator:
isql -Uusername

Adaptive Server prompts you for your password.

Note Do not use the -P option of isql to specify your password; another user

might then see your password.

You can use isql in command line mode to enter many of the Transact-SQL
examples in this manual.

System Administration Guide: Volume 1

Adaptive Server administration tasks

Entering statements
The statements that you enter in isql can span several lines. isql does not process
statements until you type go on a separate line. For example:
1>
2>
3>
4>

select *
from sysobjects
where type = "TR"
go

The examples in this manual do not include the go command between

statements. If you are typing the examples, you must enter the go command to
see the sample output.

Saving and reusing statements

This manual frequently suggests that you save the Transact-SQL statements
you use to create or modify user databases and database objects. The easiest
way to do this is to create or copy the statements to an ASCII-formatted file.
You can then use the file to supply statements to isql to re-create databases or
database objects later.
The syntax for using isql with an ASCII-formatted file is the following, where
filename is the full path and file name of the file that contains Transact-SQL
statements:
isql -Uusername -ifilename

On UNIX and other platforms, use the less than symbol (<) to redirect the
file.
The Transact-SQL statements in the ASCII file must use valid syntax and the
go command.

When reading commands from a file, you must:

Supply the -Ppassword option at the command line, or,

Include the named users password on the first line of the input file.

Using Sybase Central for system administration tasks

You can perform many system administration tasks using Sybase Control
Center and Sybase Central, a graphic tool that comes with Adaptive Server:

Initializing database devices

Adaptive Server Enterprise

CHAPTER 1

Overview of System Administration

Setting configuration parameters

Viewing the amount of free log space in a database

Generating data definition language (DDL)

Creating logins

Adding remote servers

Creating databases

Creating stored procedures

Defining roles

Adding data caches

Setting database options

Backing up and restoring databases

Use the Monitor Viewer feature of Sybase Central to access Adaptive Server
Monitor. Sybase Central includes extensive online help.
Use the Sybase Central DDL-generation feature to record your work to
Transact-SQL scripts. The DDL-generation feature lets you save to a script the
actions you perform in an entire server or within a specific database.

System tables
The master database contains system tables that keep track of information
about Adaptive Server. In addition, each database (including the master
database) contains system tables that keep track of information specific to that
database.
All the Adaptive Server-supplied tables in the master database (the Adaptive
Server controlling database) are considered system tables. Each user database
is created with a subset of these system tables. The system tables may also be
called the data dictionary or the system catalogs.
A master database and its tables are automatically created when Adaptive
Server is installed. The system tables in a user database are created when the
create database command is issued. The names of all system tables start with
sys. You cannot create tables in user databases that have the same names as
system tables. See Reference Manual: Tables for detailed descriptions of
system tables and their columns.

System Administration Guide: Volume 1

System tables

Querying the system tables

Query system tables in the same manner as any other tables. For example, the
following statement returns the names of all triggers in the database:
select name
from sysobjects
where type = "TR"

In addition, Adaptive Server supplies stored procedures (called system

procedures), many of which provide shortcuts for querying the system tables.
These system procedures provide information from the system tables:
sp_commonkey
sp_configure

sp_helpremotelogin
sp_help_resource_limit

sp_countmedatada
sp_dboption

sp_helprotect
sp_helpsegment

sp_estspace
sp_help

sp_helpserver
sp_helpsort

sp_helpartition
sp_helpcache

sp_helptext
sp_helpthreshold

sp_helpconfig
sp_helpconstraint

sp_helpuser
sp_lock

sp_helpdb
sp_helpdevice

sp_monitor
sp_monitorconfig

sp_helpgroup
sp_helpindex

sp_showcontrolinfo
sp_showexeclass

sp_helpjava
sp_helpjoins

sp_showplan
sp_spaceused

sp_helpkey
sp_helplanguage

sp_who
sp_help_resource_limit

sp_helplog

For complete information about the system procedures, see the Reference
Manual: Procedures.

Adaptive Server Enterprise

CHAPTER 1

Overview of System Administration

Keys in system tables

Primary, foreign, and common keys for system tables are defined in the master
and model databases. You can generate a report on defined keys by executing
sp_helpkey. For a report on columns in two system tables that are likely join
candidates, execute sp_helpjoins.
The Adaptive Server System Tables Diagram shows the relationships between
columns in the system tables.

Updating system tables

The Adaptive Server system tables contain information that is critical to the
operation of your databases. Under ordinary circumstances, you need not
perform direct data modifications to system tables.
Update system tables only when you are instructed to do so by Sybase
Technical Support, by an instruction in the Error Messaging and
Troubleshooting Guide, or in this manual.
Before you update system tables, you must issue an sp_configure command that
enables system table updates. While this command is in effect, any user with
appropriate permission can modify a system table. Other requirements for
direct changes to system tables are:

Modify system tables only inside a transaction. Issue a begin transaction

command before you issue the data modification command.

Verify that only the rows you wanted changed have been affected by the
command, and that the data has been changed correctly.

If the command was incorrect, issue a rollback transaction command. If the

command was correct, issue a commit transaction command.
Warning! Some system tables should not be altered by any user under any
circumstances. Some system tables are dynamically built by system
processes, contain encoded information, or display only a portion of their
data when queried. Imprudent, ad hoc updates to certain system tables can
prevent Adaptive Server from running, make database objects
inaccessible, scramble permissions on objects, or terminate a user session.
Moreover, never attempt to alter the definition of the system tables in any
way. For example, do not alter system tables to include constraints.
Triggers, defaults, and rules are not allowed in system tables. If you create
a trigger, bind a rule, or default to a system table, you see an error message.

System Administration Guide: Volume 1

System procedures

System procedures
The names of all system procedures begin with sp_. They are located in the
sybsystemprocs database, but you can run many of them in any database by
issuing the stored procedure from the database or by qualifying the procedure
name with the database name.
Sybase-supplied system procedures (such as sp_who) are created using the
installmaster installation script. Use sp_version to determine the version of
installmaster was most recently executed. See the Reference Manual: System
Procedures for more information about sp_version.
If you execute a system procedure in a database other than sybsystemprocs, the
procedure operates on the system tables in the database from which it was
executed. For example, if the database owner of pubs2 runs sp_adduser from
pubs2 or issues the command pubs2..sp_adduser, the new user is added to
pubs2..sysusers. However, this does not apply to system procedures that update
only tables in the master database.
Permissions on system procedures are discussed in the Reference Manual:
Procedures.

Using system procedures

A parameter is an argument to a stored or system procedure. If a parameter
value for a system procedure contains reserved words, punctuation, or
embedded blanks, you must enclose it in single or double quotes. If the
parameter is an object name, and the object name is qualified by a database
name or owner name, enclode the entire name in single or double quotes.
System procedures can be invoked during a session using either chained or
unchained transaction mode. Chained mode implicitly begins a transaction
before any data retrieval or modification statement. Unchained mode requires
explicit begin transaction statements paired with commit transaction or
rollback transaction statements to complete the transaction. See Chapter 21,
Transactions: Maintaining Data Consistency and Recovery, in the TransactSQL Users Guide.
You cannot execute the system procedures that modify data in system tables in
the master database from within a transaction, since this may compromise
recovery. You cannot run system procedures that create temporary worktables
from transactions.

10

Adaptive Server Enterprise

CHAPTER 1

Overview of System Administration

If no transaction is active when you execute a system procedure, Adaptive

Server turns off chained mode and sets transaction isolation level 1 for the
duration of the procedure. Before returning, the sessions chained mode and
isolation level are reset to their original settings. See Chapter 21,
Transactions: Maintaining Data Consistency and Recovery, in the TransactSQL Users Guide.
All system procedures report a return status. For example, the following means
that the procedure executed successfully:
return status = 0

If the system procedures do not execute successfully, the return status is a

number other than 0.

System procedure tables

The system procedures use several system procedure tables in the master and
sybsystemdb databases to convert internal system values (for example, status
bits) into human-readable format. One of these tables, spt_values, is used by a
variety of system procedures, including:
sp_configure
sp_dboption
sp_depends
sp_help
sp_helpdb
sp_helpdevice
sp_helpindex
sp_helpkey
sp_helprotect
sp_lock

The spt_values table can be updated only by an upgrade; you cannot modify it.
To see how it is used, execute sp_helptext and look at the text for one of the
system procedures that references it.
The other system procedure tables are spt_monitor, spt_committab, and tables
needed by the catalog stored procedures. (The spt_committab table is located in
the sybsystemdb database.)
In addition, several system procedures create, and then drop, temporary tables.
For example, sp_helpdb creates #spdbdesc, sp_helpdevice creates #spdevtab,
and sp_helpindex creates #spindtab.

System Administration Guide: Volume 1

11

System extended stored procedures

Creating system procedures

Many system procedures are explained in this manual, in the sections where
they are relevant. For detailed reference information, see the Reference
Manual: Procedures.
System administrators can write system procedures that can be executed in any
database. Create a stored procedure in sybsystemprocs and assign it a name that
begins with sp_. The uid of the stored procedure must be 1, the uid of the
database owner.
Most system procedures that you create query the system tables. Sybase
recommends that you do not create stored procedures that modify the system
tables.
To create a stored procedure that modifies system tables, a system security
officer must first turn on the allow updates to system tables configuration
parameter. Any stored procedure created while this parameter is set on can
always update system tables, even when allow updates to system tables is turned
off. To create a stored procedure that updates the system tables:
1

Use sp_configure to set allow updates to system tables on.

Use create procedure to create the stored procedure.

Use sp_configure to set allow updates to system tables off.

Warning! Use caution when you modify system tables. Always test the
procedures that modify system tables in development or test databases,
rather than in your production database.

System extended stored procedures

An extended stored procedure (ESP) letes you call external language functions
from within Adaptive Server. Adaptive Server includes a set or predefined
ESPs; users can also create their own. The names of all system extended stored
procedures begin with xp_, and are located in the sybsystemprocs database.
One very useful system ESP is xp_cmdshell, which executes an operating
system command on the system that is running Adaptive Server.

12

Adaptive Server Enterprise

CHAPTER 1

Overview of System Administration

Invoke a system ESP just like a system procedure. The difference is that a
system ESP executes procedural language code rather than Transact-SQL
statements. All ESPs are implemented by an Open Server application called
XP Server, which runs on the same machine as Adaptive Server. XP Server
starts automatically on the first ESP invocation.
For information about the system ESPs provided with Adaptive Server, see the
Reference Manual: Procedures.

Creating system ESPs

Use create procedure to create a system ESP in the sybsystemprocs database.
System procedures are automatically included in the sybsystemprocs database.
The name of the ESP, and its procedural language function, must begin with
xp_. The uid of the stored procedure must be 1, the uid of the database owner.
For general information about creating ESPs see Chapter 17, Using Extended
Stored Procedures, in the Transact-SQL Users Guide.

Logging error messages

Each time it starts, Adaptive Server writes start-up information to a local error
log file. The installation program automatically sets the error log location when
you configure a new Adaptive Server. See the Configuration Guide for your
platform to learn the default location and file name of the error log.
Many error messages from Adaptive Server are written only to the users
terminal. However, fatal error messages (severity levels 19 and above), kernel
error messages, and informational messages from Adaptive Server are
recorded in the error log file.
Adaptive Server keeps the error log file open until you stop the server process.
Before deleting old messages to reduce the size of the error log, stop the
Adaptive Server process.
Note On some platforms, such as Windows, Adaptive Server also records error

messages in the operating system event log. See the installation guide and
configuration guide for your platform.

System Administration Guide: Volume 1

13

Connecting to Adaptive Server

Connecting to Adaptive Server

Adaptive Server can communicate with other Adaptive Servers, Open Server
applications, and client software on the network. Clients can talk to one or
more servers, and servers can communicate with other servers using remote
procedure calls. For products to interact with one another, each must know
where the others reside on the network. This network service information is
stored in the interfaces file.

The interfaces file

The interfaces file is usually named interfaces, interface, or sql.ini, depending
on the operating system.
The interfaces file lists the name and address of every known server. When you
use a client program to connect to a server, the program looks up the server
name in the interfaces file and then connects to the server using the address, as
shown in Figure 1-1.
Figure 1-1: Connecting to Adaptive Server

Connect me
to the TEST
Adaptive Server.

interfaces
interfaces file
TEST

TEST
Adaptive Server

The name, location, and contents of the interfaces file differ between operating
systems. Also, the format of the Adaptive Server addresses in the interfaces file
differs between network protocols.

14

Adaptive Server Enterprise

CHAPTER 1

Overview of System Administration

The Adaptive Server installation program creates a simple interfaces file that
you can use for local connections to Adaptive Server over one or more network
protocols. As a system administrator, modify the interfaces file and distribute
it to users so that they can connect to Adaptive Server over the network. See
the Configuration Guide for your platform for information about the interfaces
file.
See Chapter 2, Networks and Performance in the Performance and Tuning
Series: Basics for more information about the interfaces file and network
listeners.

Directory services
A directory service manages the creation, modification, and retrieval of
network service information. Directory services are provided by platform or
third-party vendors and must be purchased and installed separately from
Adaptive Server. Two examples of directory services are Registry and
Distributed Computing Environment (DCE).
The $SYBASE/$SYBASE_OCS/config/libtcl.cfg file is a Sybase-supplied
configuration file used by servers and clients to determine:

Which directory service to use, and

The location of the specified directory service driver.

If no directory services are installed or listed in the libtcl.cfg file, Adaptive

Server defaults to the interfaces file for obtaining network service information.
The system administrator must modify the libtcl.cfg file as appropriate for the
operating environment.
Some directory services are specific to a given platform; others can be used on
several different platforms. Because of the platform-specific nature of
directory services, see the configuration documentation for your platform for
detailed information about configuring for directory services.

System Administration Guide: Volume 1

15

Connecting to Adaptive Server

LDAP as a directory service

Lightweight Directory Access Protocol (LDAP) is an industry standard for
accessing directory services. Directory services allow components to look up
information by a distinguished name (DN) from an LDAP server that stores
and manages server, user, and software information that is used throughout the
enterprise or over a network.
The LDAP server can be located on a different platform from the one on which
Adaptive Server or the clients are running. LDAP defines the communication
protocol and the contents of messages exchanged between clients and servers.
Messages are operators, such as client requests for read, write and query, and
server responses, including metadata (data about data).
The LDAP server can store and retrieve information about:

Adaptive Server, such as IP address, port number, and network protocol

Security mechanisms and filters

High availability companion server name

Authentication information for user access to Adaptive Server

You can authenticate users logging in to Adaptive Server through
information stored in the syslogins directory or through a centralized
LDAP server that enables a single login and password throughout the
enterprise. See Chapter 3, Managing Adaptive Server Logins and
Database Users, in the Security Adiminstration Guide.

You can configure the LDAP server to use these access restrictions:

16

Anonymous authentication all data is visible to any user.

User name and password authentication Adaptive Server uses the default
user name and password from the appropriate file:

UNIX, 32-bit $SYBASE/$SYBASE_OCS/config/libtcl.cfg

UNIX, 64-bit $SYBASE/$SYBASE_OCS/config/libtcl64.cfg

Windows %SYBASE%\%SYBASE_OCS%\ini\libtcl.cfg

Adaptive Server Enterprise

CHAPTER 1

Overview of System Administration

User name and password authentication properties establish and end a session
connection to an LDAP server.
Note The default user name and password stored in libtcl.cfg and passed to the

LDAP server for authentication purposes are distinct and different from those
used to access Adaptive Server. The default user name and password allow
access to the LDAP server for administrative tasks.
When an LDAP server is specified in the libtcl.cfg or libtcl64.cfg file
(collectively called libtcl*.cfg file), the server information is then accessible
only from the LDAP server; Adaptive Server ignores the interfaces file.
If multiple directory services are supported in a server, the order in which they
are searched is specified in libtcl*.cfg. You cannot use the dataserver command
line option to specify the search order.

Multiple directory services

Any type of LDAP service, whether it is an actual server or a gateway to other
LDAP services, is called an LDAP server.
You can specify multiple directory services for high-availability failover
protection in libtcl*.cfg. Not every directory service in the list must be an
LDAP server.
In the following example, if the connection to test:389 fails, the connection
fails over to the DCE driver with the specified directory information tree (DIT)
base. If this also fails, a connection to the LDAP server on huey:11389 is
attempted. Different vendors employ different DIT base formats.
[DIRECTORY]
ldap=libdldap.so ldap://test:389/dc=sybase,dc=com
dce=libddce.so ditbase=/.:/subsys/sybase/dataservers
ldap=libdldap.so ldap://huey:11389/dc=sybase,dc=com

See the Open Client Client-Library/C Programmers Guide and the Open
Client Client-Library/C Reference Manual.

LDAP directory services versus the Sybase interfaces file

The LDAP driver implements directory services for use with an LDAP server.
The LDAP infrastructure provides:

A network-based alternative to the traditional Sybase interfaces file

System Administration Guide: Volume 1

17

Security features available in Adaptive Server

A single, hierarchical view of information, including users, software,

resources, networks, files, and so on

Table 1-1 highlights the differences between the Sybase interfaces file and an
LDAP server.
Table 1-1: interfaces file versus LDAP directory services
interfaces file
Platform-specific

Directory services
Platform-independent

Specific to each Sybase installation

Contains separate master and query entries

Centralized and hierarchical

One entry for each server that is accessed by both clients and servers

Cannot store metadata about the server

Stores metadata about the server

Performance

Performance when using an LDAP server may be slower than when using an
interfaces file because the LDAP server requires time to make a network
connection and retrieve data. Since this connection is made when Adaptive
Server is started, changes in performance are seen at login time, if at all. During
normal system load, the delay should not be noticeable. During high system
load with many connections, especially repeated connections with short
duration, the overall performance difference of using an LDAP server versus
the traditional interfaces file might be noticeable.

Security features available in Adaptive Server

The major security features available for Adaptive Server are:

18

Identification and authentication controls Ensures that only authorized

users can log in to the system. In addition to password-based login
authentication, Adaptive Server supports external authentication using
Kerberos, LDAP, or pluggable authentication modules (PAM).

Discretionary access controls (DAC) Provides access controls that let

object owners restrict access to objects, usually with the grant and revoke
commands. This type of control is dependent upon an object owners
discretion. .

Adaptive Server Enterprise

CHAPTER 1

Overview of System Administration

Division of roles Allows an administrator to grant privileged roles to

specified users so only designated users can perform certain tasks.
Adaptive Server has predefined roles, called system roles, such as
system administrator and system security officer. In addition, Adaptive
Server allows system security officers to define additional roles, called
user-defined roles.

Accountability Provides the ability to audit events such as logins,

logouts, server start operations, remote procedure calls, accesses to
database objects, and all actions performed by a specific user or with a
particular role active. Adaptive Server also provides a single option to
audit a set of server-wide, security-relevant events.

Confidentiality of data Maintains a confidentiality of data using

encryption for client/server communication, available with Kerberos or
secure sockets layer (SSL). Inactive data is kept confidential with
password-protected database backup.

See Chapter 2, Getting Started with Security Administration in Adaptive

Server, in the Security Administration Guide.

System Administration Guide: Volume 1

19

Security features available in Adaptive Server

20

Adaptive Server Enterprise

CH A PTE R

System and Optional Databases

This chapter describes the system databases that reside on all Adaptive
Server systems. It also describes optional Sybase-supplied databases that
you can install, and the sybdiag database, which Sybase Technical Support
may install for diagnostic purposes.
Topic
Overview of system databases

Page
21

master database
model database

23
25

sybsystemprocs database
tempdb database

26
26

sybsecurity database
sybsystemdb database

28
28

sybmgmtdb database
pubs2 and pubs3 sample databases

28
29

dbccdb database
sybdiag database

30
30

Determining the version of the installation scripts

30

Overview of system databases

A default installation of Adaptive Server includes these system databases:

The master database

The model database

The system procedure database, sybsystemprocs

The two-phase commit transaction database, sybsystemdb

The temporary database, tempdb

Optionally, you can install:

The auditing database, sybsecurity

System Administration Guide: Volume 1

21

Overview of system databases

The sample databases, pubs2 and pubs3

The dbcc database, dbccdb

The Job Scheduler database, sybmgmtdb

For information about installing the master, model, sybsystemprocs, tempdb,

and sybmgmtdb databases, see the installation guide for your platform. For
information on installing dbccdb, Chapter 10, Checking Database
Consistency, in the System Administration Guide: Volume 2. For information
about using Job Scheduler, see the Job Scheduler Users Guide.
The master, model, sybsystemdb, and temporary databases reside on the master
device which is named during installation. The master database is contained
entirely on the master device and cannot be expanded onto any other device.
Create all other databases and user objects on other devices.
Warning! Do not store user databases on the master device; doing so makes it
difficult to recover both the system databases and any user databases stored on
the master device.

Install the sybsecurity and sybmgmtdb databases on their own devices and
segment. See the installation documentation for your platform.
Install the sybsystemprocs database on a device of your choice. You may want
to modify the installation scripts for pubs2 and pubs3 to share the device you
create for sybsystemprocs.
Use the installjsdb script (located in $SYBASE/ASE-15_0/scripts) to install the
sybmgmtdb database. installjsdb looks for a device named sybmgmtdev on
which to create the sybmgmtdb database and its accompanying tables and
stored procedures. If the sybmgmtdb database already exists, installjsdb creates

the Job Scheduler tables and stored procedures in the existing database. If
installjsdb cannot find either a sybmgmtdev device or a sybmgmtdb database, it
creates sybmgmtdb on the master device. However, Sybase strongly
recommends that you remove the sybmgmtdb database from the master device.
The installpubs2 and the installpubs3 scripts do not specify a device in their
create database statement, so they are created on the default device. During
installation, the master device is the default device. To change this, either edit
the scripts or follow the instructions in Chapter 8, Initializing Database
Devices.

22

Adaptive Server Enterprise

CHAPTER 2

System and Optional Databases

master database
The master database controls the operation of Adaptive Server and stores
information about all user databases and their associated database devices.
Table 2-1 describes the information that the master database stores.
Table 2-1: Information the master database stores
Information

System table

User accounts
Remote user accounts

syslogins

Remote servers that this server can interact with

Ongoing processes

sysservers

Configurable environment variables

sysconfigures

System error messages

Databases on Adaptive Server

sysmessages

Storage space allocated to each database

Tapes and disks mounted on the system

sysusages

Active locks
Character sets

syslocks

Languages
Users who hold server-wide roles

syslanguages

Server roles
Adaptive Server engines that are online

syssrvroles

sysremotelogins
sysprocesses

sysdatabases
sysdevices
syscharsets
sysloginroles
sysengines

Because the master database stores information about user databases and
devices, you must be in the master database to issue the create database, alter
database, disk init, disk refit, disk reinit, and disk mirroring commands.
The minimum size of your master database depends on your server's logical
page size. The master database must contain at least 6656 logical pages, so its
minimum physical size for each logical page size is:

2K page 13MB

4K page 26MB

8K page 52MB

16K page 104MB

System Administration Guide: Volume 1

23

master database

Controlling object creation in master

When you install Adaptive Server, only a system administrator can create
objects in the master database, because the system administrator implicitly
becomes dbo of any database he or she uses. Any objects created on the
master database should be used only for system administration. Set
permissions in master so that most users cannot create objects there.
Warning! Do not place user objects in master. Storing user objects in master
causes the transaction log to fill quickly. If the transaction log runs out of space
completely, you cannot use dump transaction commands to free space in
master.

You may also want to use alter login to change the default database for users (the
database to which a user is connected when he or she logs in). See Chapter 3,
Managing Adaptive Server Logins and Database Users, in the Security
Administration Guide.
Create any system procedures in the sybsystemprocs database rather than in
master.

Backing up master and keeping copies of system tables

To be prepared for hardware or software failure on Adaptive Server:

Perform frequent backups of the master database and all user databases.
See Keep up-to-date backups of master on page 40, and Chapter 13,
Restoring the System Databases, in System Administration Guide:
Volume 2.

Keep a copy (preferably offline) of these system tables: sysusages,

sysdatabases, sysdevices, sysloginroles, and syslogins. See Keep offline
copies of system tables on page 40. If you have copies of these scripts,
and a hard-disk failure or other disaster makes your database unusable,
you can use the recovery procedures described in Chapter 13, Restoring
the System Databases, in System Administration Guide: Volume 2. If you
do not have current copies of your scripts, it is much more difficult to
recover Adaptive Server when the master database is damaged.

24

Adaptive Server Enterprise

CHAPTER 2

System and Optional Databases

model database
Adaptive Server includes the model database, which provides a template, or
prototype, for new user databases. Each time a user enters the create database
command, Adaptive Server makes a copy of the model database and extends
the new database to the size specified by the create database command.
Note New databases must be at least as large as the model database.

The model database contains the required system tables for each user database.
You can modify model to customize the structure of newly created databases
everything you do to model is reflected in each new database. Some of the
changes that system administrators commonly make to model are:

Adding user-defined datatypes, rules, or defaults.

Adding users who need access to all databases on Adaptive Server.

Granting default privileges, particularly for guest accounts.

Setting database options such as select into/bulkcopy/pllsort. These settings

are reflected in all new databases. The default settings for these options in
model is off. See Chapter 9, Setting Database Options.

Typically, most users do not have permission to modify the model database.
There is not much point in granting read permission either, since Adaptive
Server copies its entire contents into each new user database.
The model database cannot be larger than tempdb. By default, the size of model
is six allocation units (an allocation unit is 256 logical pages.). You see error
message if you increase the size of model without making tempdb at least as
large.
Note Keep a backup copy of the model database, and back up model with dump
database each time you change it. In case of media failure, restore model as you

would a user database.

System Administration Guide: Volume 1

25

sybsystemprocs database

sybsystemprocs database
Sybase system procedures are stored in the database sybsystemprocs. When a
user in any database executes a system stored procedure (that is, a procedure
whose name begins with sp_), Adaptive Server first looks for that procedure in
the users current database. If there is no procedure there with that name,
Adaptive Server looks for it in sybsystemprocs. If there is no procedure in
sybsystemprocs, Adaptive Server looks for the procedure in master.
If the procedure modifies system tables (for example, sp_adduser modifies the
sysusers table), the changes are made in the database from which the procedure
was executed.
To change the default permissions on system procedures, modify those
permissions in sybsystemprocs.
Note Any time you make changes to sybsystemprocs, back up the database.

tempdb database
Adaptive Server has a temporary database, tempdb, that provides a storage
area for temporary tables and other temporary working storage needs. The
space in tempdb is shared among all users of all databases on the server.
The default size of tempdb depends on the logical page size for your server, 2,
4, 8, or 16K. Certain activities may make it necessary for you to increase the
size of tempdb:

Large temporary tables.

A lot of activity on temporary tables, which fills up the tempdb logs.

Large or many simultaneous sorts. Subqueries and aggregates with group

by also cause some tempdb activity.

Use alter database to increase the size of tempdb. tempdb is initially created on
the master device. You can add space to tempdb from the master device or from
any other database device.
If you run update index statistics against large tables, the command fails with
error number 1105 if tempdb is not large enough.

26

Adaptive Server Enterprise

CHAPTER 2

System and Optional Databases

You can create and manage multiple temporary databases in addition to the
system temporary database, tempdb. Multiple temporary databases reduce
contention on system catalogs and logs in tempdb.

Creating temporary tables

No special permissions are required to create temporary tables or to execute
commands that may require storage space in the temporary database.
Create temporary tables either by preceding the table name in a create table
statement with a pound sign (#), or by specifying the name prefix tempdb...
Temporary tables created with a pound sign are accessible only by the current
Adaptive Server session: users on other sessions cannot access them. These
nonsharable, temporary tables are destroyed at the end of each session. The
first 13 bytes of the tables name, including the pound sign (#), must be unique.
Adaptive Server assigns the names of such tables a 17-byte number suffix.
(You can see the suffix by querying tempdb..sysobjects.)
Temporary tables created with the tempdb.. prefix are stored in tempdb and
can be shared among Adaptive Server sessions. Adaptive Server does not
change the names of temporary tables created this way. The table exists either
until you restart Adaptive Server or until its owner drops it using drop table.
System procedures work on temporary tables, but only if you use them from
tempdb.

If a stored procedure creates temporary tables, the tables are dropped when the
procedure exits. You can also explicitly drop temporary tables before a session
ends.
Warning! Do not create temporary tables with the tempdb.. prefix from
inside a stored procedure unless you intend to share those tables among other
users and sessions.

Each time you restart Adaptive Server, it copies model to tempdb, which clears
the database. You cannot recover temporary tables.

System Administration Guide: Volume 1

27

sybsecurity database

sybsecurity database
The sybsecurity database, which contains the auditing system for Adaptive
Server, includes :

The system tables, sysaudits_01, sysaudits_02, ... sysaudits_08, which

contain the audit trail

The sysauditoptions table, which contains rows describing the global audit
options

All other default system tables that are derived from model

See Chapter 7, Auditing, in the Security Administration Guide.

sybsystemdb database
The sybsystemdb database stores information about distributed transactions.
Adaptive Server versions 12.0 and later can provide transaction coordination
services for transactions that are propagated to remote servers using remote
procedure calls (RPCs) or Component Integration System (CIS). Information
about remote servers participating in distributed transactions is stored in the
syscoordinations table.
The sybsystemdb database also stores information about SYB2PC transactions
that use the Sybase two-phase commit protocol. The spt_committab table,
which stores information about and tracks the completion status of each twophase commit transaction, is stored in the sybsystemdb database.
See the Configuration Guide for your platform for information about twophase commit transactions and how to create the sybsystemdb database.

sybmgmtdb database
The sybmgmtdb database stores jobs, schedules, scheduled jobs information,
and data the internal Job Scheduler task needs for processing. sybmgmtdb also
maintains the output and results from these executed tasks. See the Job
Scheduler Users Guide.

28

Adaptive Server Enterprise

CHAPTER 2

System and Optional Databases

pubs2 and pubs3 sample databases

Installing the pubs2 and pubs3 sample databases is optional. These databases
are provided as a learning tool for Adaptive Server. The pubs2 sample database
is used for most of the examples in the Adaptive Server documentation, except
for examples, where noted, that use the pubs3 database.
The sample databases are intended for training purposes only. Do not install
them in an Adaptive Server production environment.
For information about installing pubs2 and pubs3, see the installation guide for
your platform. For information about the contents of these sample databases,
see the Transact-SQL Users Guide.

Maintaining the sample databases

The sample databases include a guest user login that allows access to the
database by any authorized Adaptive Server user. The guest login has been
given a wide range of privileges in pubs2 and pubs3, including permissions to
select, insert, update, and delete user tables. See Chapter 3, Managing
Adaptive Server Logins and Database Users, in the Security Administration
Guide.
The size of the pubs2 and pubs3 databases are determined by the size of the
logical page size for your server; 2, 4, 8, and 16K. If possible, give each new
user a clean copy of pubs2 and pubs3 so that she or he is not confused by other
users changes. To place pubs2 or pubs3 on a specific database device, edit the
installation script before installing the database.
If space is a problem, instruct users to issue the begin transaction command
before updating a sample database. After the user has finished updating one of
the sample databases, he or she can issue the rollback transaction command to
undo the changes.

pubs2 image data

Adaptive Server includes a script for installing image data in the pubs2
database (pubs3 does not use the image data). The image data consists of six
pictures, two each in PICT, TIF, and Sun raster file formats. Sybase does not
provide any tools for displaying image data. You must use the appropriate
screen graphics tools to display the images after you extract them from the
database.

System Administration Guide: Volume 1

29

dbccdb database

See the installation documentation for your platform for information about
installing the image data in pubs2.

dbccdb database
dbcc checkstorage records configuration information for the target database,
operation activity, and the results of the operation in the dbccdb database.
Stored in the database are dbcc stored procedures for creating and maintaining
dbccdb and for generating reports on the results of dbcc checkstorage

operations. See Chapter 10, Checking Database Consistency, in the System

Administration Guide: Volume 2.

sybdiag database
Sybase Technical Support may create the sybdiag database on your system for
debugging purposes. This database holds diagnostic configuration data, and
should not be used by customers.

Determining the version of the installation scripts

sp_version lets you determine the current version of the scripts (installmaster,
installdbccdb, and so on) installed on Adaptive Server, whether they ran
successfully or not, and the length of time they took to complete.

The syntax for sp_version is:

sp_version [script_file [, all]]

where:

script_file is the name of the installation script (the default value is NULL).

all reports details about each script, such as the date executed, and the
length of time for execution.

If you run sp_version without any parameters, it reports on all scripts.

30

Adaptive Server Enterprise

CHAPTER 2

System and Optional Databases

This example describes what installation scripts were run, what time they were
run, and what time they finished:
sp_version null, 'all'
Script
Version
Status
-------------------------------------------------------------------------------installmaster
15.0/EBF XXXXX/B/Sun_svr4/OS 5.8/asemain/1/32-bit/OPT/Thu Sep
23 22:12:12 2004
Complete [Started=Sep 24 2004 3:39PM]-[Completed=Sep 24 2004 3:45PM

System Administration Guide: Volume 1

31

Determining the version of the installation scripts

32

Adaptive Server Enterprise

CH A PTE R

System Administration for

Beginners

This chapter:

Introduces new system administrators to important topics

Helps system administrators find information in the Sybase

documentation

Experienced administrators may also find this chapter useful for

organizing ongoing maintenance activities.
Topic
Logical page sizes
Using test servers

Page
33
34

Considerations when installing Sybase products

Allocating physical resources

35
37

Backup and recovery

39

Ongoing maintenance and troubleshooting

Keeping records

42
43

Additional resources

46

Logical page sizes

Database objects are built with logical pages. A databases and any of its
related objects must use the same logical page size. That is, you cannot
create a server that uses more than one logical page size. Adaptive Server
allows you to create master devices and databases with logical page sizes
of 2K, 4K, 8K, or 16K, but a given server installation can use only one of
these four logical page sizes.
All databases in a serverand all objects in every databaseuse the same
logical page size. For example, all the pages on a server with a logical
page size of 4K must be 4K, even though you may not use some pages
beyond the initial 2K.
System Administration Guide: Volume 1

33

Using test servers

Select the page size when you create the master device with dataserver -z.
For more information about the dataserver command, which is the command
used to create the master device, see the Utility Guide. For more information
about logical page sizes, see Chapter 3, Configuring Memory, in System
Administration Guide: Volume 2.

Using test servers

Sybase suggests that you install and use a test or development Adaptive Server,
then remove it before you create the production server. Using a test server
makes it easier to plan and test different configurations and less stressful to
recover from mistakes. It is much easier to learn how to install and administer
new features when there is no risk of having to restart a production server or
re-create a production database.
If you use a test server, Sybase suggests that you do so from the point of
installing or upgrading Adaptive Server through the process of configuring the
server. It is in these steps that you make some of the most important decisions
about your final production system. The following sections describe how using
a test server can help system administrators.

Planning resources
Using a test server helps you plan the final resource requirements for your
system and helps you discover resource deficiencies that you might not have
anticipated.
In particular, disk resources can have a dramatic effect on the final design of
the production system. For example, you may decide that, in the event of a
media failure, a particular database requires nonstop recovery. This means you
must configure one or more additional database devices to mirror the critical
database. Discovering these resource requirements in a test server allows you
to change the physical layout of databases and tables without affecting
database users.
Use a test server to benchmark both Adaptive Server and your applications
using different hardware configurations. This allows you to determine the
optimal setup for physical resources at both the Adaptive Server level and the
operating system level before bringing the entire system online for general use.

34

Adaptive Server Enterprise

CHAPTER 3

System Administration for Beginners

Achieving performance goals

Most performance objectives can be met only by carefully planning a
databases design and configuration. For example, you may discover that the
insert and I/O performance of a particular table causes a bottleneck. In this
case, the best course of action may be to re-create the table on a dedicated
segment and partition the table. Changes of this nature are disruptive to a
production system; even changing a configuration parameter may require you
to restart Adaptive Server.

Considerations when installing Sybase products

The responsibility for installing Adaptive Server and other Sybase products is
sometimes placed with the system administrator. If installation is one of your
responsibilities, use the following pointers to help you in the process.

Check product compatibility

Before installing new products or upgrading existing products, always read the
release bulletin included with the products to understand any compatibility
issues that might affect your system. Compatibility problems can occur
between hardware and software and between different release levels of the
same software. Reading the release bulletin in advance can save the time and
guesswork of troubleshooting known compatibility problems. Pay particular
attention to the lists of known problems that are included in the release bulletin.

Install or upgrade Adaptive Server

Read through the installation guide for your platform before you begin a new
installation or upgrade. You may also want to consult with the operating system
administrator to discuss operating system requirements for Adaptive Server.
These requirements can include the configuration of memory, raw devices,
asynchronous I/O, and other features, depending on the platform you use.
Many of these tasks must be completed before you begin the installation.

System Administration Guide: Volume 1

35

Considerations when installing Sybase products

If you are upgrading a server, back up all data (including the master database,
user databases, triggers, and system procedures) offline before you begin. After
upgrading, immediately create a separate, full backup of your data, especially
if there are incompatibilities between older dump files and the newer versions.

Install additional third-party software

Adaptive Server generally includes support for the network protocols that are
common to your hardware platform. If your network supports additional
protocols, install the required protocol support.
As an alternative to the Sybase interfaces file, you can use a directory service
to obtain a servers address and other network information. Directory services
are provided by platform or third-party vendors and must be purchased and
installed separately from the installation of Adaptive Server. See also
Directory services on page 15 and the Configuration Guide for your
platform for a list of the directory services that Adaptive Server currently
supports.

Configure and test client connections

A successful client connection depends on the coordination of Adaptive
Server, the client software, and network products. If you are using one of the
network protocols installed with Adaptive Server, see the Configuration Guide
for your platform for information about testing network connections. If you are
using a different network protocol, follow the instructions that are included
with the network product. You can also use ping utilities that are included
with Sybase connectivity products to test client connections with Adaptive
Server. For a general description of how clients connect to Adaptive Server, see
Connecting to Adaptive Server on page 14. For details about the name and
contents of the interfaces file, see the Configuration Guide for your platform

36

Adaptive Server Enterprise

CHAPTER 3

System Administration for Beginners

Allocating physical resources

Allocating physical resources is providing Adaptive Server the memory, disk
space, worker processes, and CPU power required to achieve your
performance and recovery goals. When installing a new server, every system
administrator must make decisions about resource utilization. If you upgrade
your platform, or if the design of your database system changes, you must also
reallocate Adaptive Server resources by adding new memory, disk controllers,
or CPUs. Early benchmarking of Adaptive Server and your applications can
help you identify hardware resource deficiencies that create performance
bottlenecks.
See Chapter 16, Overview of Disk Resources in Volume 2 of the System
Administration Guide: Volume 2 to understand the kinds of disk resources
required by Adaptive Server. See also see Chapter 3, Configuring Memory,
in System Administration Guide: Volume 2 and Chapter 5, Managing
Mulitprocessor Servers, in System Administration Guide: Volume 2 for
information about memory and CPU resources.
The following sections provide helpful pointers in determining physical
resource requirements.

Dedicated versus shared servers

The first step in planning Adaptive Server resources to understand the
resources required by other applications running on the same machine.
Generally, system administrators dedicate an entire machine for Adaptive
Server use, which means that only the operating system and network software
consume resources that might otherwise be reserved for Adaptive Server. On a
shared system, other applications, such as Adaptive Server client programs or
print servers, run on the same machine as Adaptive Server. It can be difficult
to calculate the resources available to Adaptive Server on a shared system,
because the types of applications and their pattern of use may change over
time.
It is the system administrators responsibility to take into account the resources
used by operating systems, client programs, windowing systems, and so forth
when configuring resources for Adaptive Server. Configure Adaptive Server to
use only the resources that are available to it. Otherwise, the server may
perform poorly or fail to start.

System Administration Guide: Volume 1

37

Allocating physical resources

Decision-support and OLTP applications

Adaptive Server contains many features that optimize performance for OLTP,
decision-support, and mixed-workload environments. However, to make
optimal use of these features, determine in advance the requirements of your
systems applications.
For mixed-workload systems, list the individual tables that you anticipate will
be most heavily used for each type of application; this can help you achieve
maximum performance for applications.

Advance resource planning

It is extremely important that you understand and plan resource usage in
advance. In the case of disk resources, for example, after you initialize and
allocate a device to Adaptive Server, that device cannot be used for any other
purpose (even if Adaptive Server never fills the device with data). Likewise,
Adaptive Server automatically reserves the memory for which it is configured,
and this memory cannot be used by any other application.
When planning resource usage:

38

For recovery purposes, always place a databases transaction log on a

separate physical device from its data. See Chapter 6, Creating and
Managing User Databases, in System Administration Guide: Volume 2.

Consider mirroring devices that store mission-critical data. See Chapter 2,

Mirroring Database Devices, in System Administration Guide: Volume
2. If your operating system supports these features, consider using disk
arrays and disk mirroring for Adaptive Server data.

If you are working with a test Adaptive Server, for convenience, you may
find it easier to initialize database devices as operating system files, rather
than raw devices. Adaptive Server supports either raw partitions or
certified file systems for its devices.

Changing configuration options can affect the way Adaptive Server

consumes physical resources, especially memory. See Chapter 5, Setting
Configuration Parameters, for details about the amount of memory used
by individual parameters.

Adaptive Server Enterprise

CHAPTER 3

System Administration for Beginners

Operating system configuration

Once you have determined the resources that are available to Adaptive Server
and the resources you require, configure these physical resources at the
operating system level:

If you are using raw partitions, initialize the raw devices to the sizes
required by Adaptive Server. If you initialize a raw device for Adaptive
Server, you cannot use that device for any other purpose (for example, to
store operating system files). Ask your operating system administrator for
assistance in initializing and configuring raw devices to the required sizes.

Configure the number of network connections. Make sure that the

machine on which Adaptive Server runs can actually support the number
of connections you configure. See your operating system documentation.

Additional configuration may be required for your operating system and

the applications that you use. Read the installation guide for your platform.
Also read your client software documentation or consult with your
engineers to understand the operating system requirements for your
applications.

Backup and recovery

Making regular backups of your databases is crucial to the integrity of your
database system. Although Adaptive Server automatically recovers from
system crashes (for example, power outages) or server failures, only you can
recover from data loss caused by media failure.
The following chapters, from the System Adminstration Guide: Volume 2,
describe how to develop and implement a backup and recovery plan:

Chapter 11, Developing a Backup and Recovery Plan

Chapter 12, Backing Upa and Restoring User Databases

Chapter 13, Restoring the System Databases

Chapter 16, Managing Free Space with Thresholds

System Administration Guide: Volume 1

39

Backup and recovery

Keep up-to-date backups of master

Backing up the master database is the most crucial element of any backup and
recovery plan. The master database contains details about the structure of your
entire database system. Its stores information about the Adaptive Server
databases, devices, and device fragments that make up those databases.
Because Adaptive Server needs this information for recovery, it is crucial that
you maintain an up-to-date backup copy of the master database at all times.
To ensure that your backup of master is always up to date, back up the database
after each command or procedure that affects disks, storage, databases, or
segments, including:

Creating or deleting databases

Initializing new database devices

Adding new dump devices

Using any device mirroring command

Creating or dropping system stored procedures, if they are stored in master

Creating, dropping, or modifying a segment

Adding new Adaptive Server logins

To back up master to a tape device, start isql and enter the command:
dump database master to "tape_device"

where tape_device is the name of the tape device (for example, /dev/rmt0).

Keep offline copies of system tables

In addition to backing up master regularly, keep offline copies of these system
tables: sysdatabases, sysdevices, sysusages, sysloginroles, and syslogins. Use
the bcp utility described in the Utility Guide and store a printed copy of the
contents of each system table. Create a printed copy by printing the output of:
select
select
select
select
select

*
*
*
*
*

from
from
from
from
from

sysusages order by vstart

sysdatabases
sysdevices
sysloginroles
syslogins

If you have copies of these tables, and a hard-disk failure or other disaster
makes your database unusable, you can use the recovery procedures described
in Chapter 13, Restoring the System Databases, in System Administration
Guide: Volume 2.

40

Adaptive Server Enterprise

CHAPTER 3

System Administration for Beginners

Also keep copies of all data definition language (DDL) scripts for user objects,
as described under Keeping records on page 43.

Automate backup procedures

Creating an automated backup procedure makes the process easier and quicker
to perform. Automating backups can be as simple as using an operating system
script or a utility (for example, the UNIX cron utility) to perform the necessary
backup commands. Or you can automate the procedure further by using
thresholds, which are discussed in Chapter 16, Managing Free Space with
Thresholds, in System Administration Guide: Volume 2.

Creating an automated backup procedure

Although the commands required to create an automated script vary, depending

on the operating system you use, all scripts should accomplish the same basic
steps:
1

Start isql and dump the transaction log to a holding area (for example, a
temporary file).

Rename the dump file to a name that contains the dump date, time, and
database name.

In a history file, record information about the new backup.

In a separate file, record any errors that occurred during the dump.

Automatically send mail to the system administrator for any error

conditions.

Verify data consistency before backing up a database

Your database backups must be consistent and accurate, especially for master.
If you back up a database that contains internal errors, the errors persist in a
restored version of the database.
Use the dbcc commands to check a database for errors before backing it up.
Always use dbcc commands to verify the integrity of a database before
dumping it. If dbcc detects errors, correct them before dumping the database.

System Administration Guide: Volume 1

41

Ongoing maintenance and troubleshooting

Over time, if you discover few or no errors while running dbcc, you may decide
that the risk of database corruption is small and that you need to run dbcc only
occasionally. If the consequences of losing data are too high, continue to run
dbcc commands each time you back up a database.
Note For performance considerations, many sites choose to run dbcc checks

outside of peak hours or on separate servers.

See Chapter 10, Checking Database Consistency, in the System
Administration Guide: Volume 2.

Monitor the log size

When the transaction log becomes nearly full, it may be impossible to use
standard procedures to dump transactions and reclaim space. The system
administrator should monitor the log size and perform regular transaction log
dumps (in addition to regular database dumps) to avoid this situation. Set up a
threshold stored procedure that notifies you (or dumps the log) when the log
reaches a certain capacity. See Chapter 16, Managing Free Space with
Thresholds, in System Administration Guide: Volume 2. Sybase also suggests
that, to shorten the time required to dump and load the database, dump the
transaction log immediately prior to performing a full database dump.
You can also monitor the space used in the log segment manually using
sp_helpsegment, as described under Chapter 8, Creating and Using Segments in
System Administration Guide: Volume 2.

Ongoing maintenance and troubleshooting

In addition to making regularly scheduled backups, the system administrator
performs the maintenance activities throughout the life of a server discussed in
this section.

42

Adaptive Server Enterprise

CHAPTER 3

System Administration for Beginners

Starting and stopping Adaptive Server

Most system administrators automate the procedure for starting Adaptive
Server to coincide with the start-up of the server machine. Do this by editing
operating system start-up scripts, or by using other operating system
procedures. See the configuration documentation for your platform to
determine how to start and stop Adaptive Server.

Viewing and pruning the error log

Examine the contents of the error log on a regular basis to determine whether
serious errors have occurred. You can also use operating system scripts to scan
the error log for particular messages and to automatically notify the system
administrator when specific errors occur. Checking the error log regularly may
help determine whether there are continuing problems of the same nature, or
whether a particular database device is likely to fail. See Chapter 12,
Diagnosing System Problems, for more information about error messages
and their severity.
The error log file can grow large over time, since Adaptive Server appends
informational and status messages to it each time it starts. You can periodically
prune the log file by opening the file and deleting old records. Keeping the
log file to a manageable size saves disk space and makes it easier to locate
current errors.

Keeping records
Keeping records about your Adaptive Server system is an important part of
your job as a system administrator. Accurate records of changes and problems
that you have encountered can be a valuable reference when you are contacting
Sybase Technical Support or recovering databases. They can also provide vital
information for administrators who manage the Adaptive Server system in
your absence.

System Administration Guide: Volume 1

43

Keeping records

Contact information
Maintain a list of contact information for yourself as well as the System
Security Officer, Operator, and database owners on your system. Also, record
secondary contacts for each role. Make this information available to all
Adaptive Server users so that the appropriate contacts receive enhancement
requests and problem reports.

Configuration information
Ideally, create databases and database objects, and configure Adaptive Server
using script files that you store in a safe place. Storing the script files makes it
possible to re-create your entire system in the event of a disaster. You can also
use script files to quickly re-create database systems for evaluation purposes on
new hardware platforms. If you use a third-party tool to perform system
administration, remember to generate equivalent scripts after performing
administration tasks.
Consider recording the following kinds of information:

Commands used to create databases and database objects (DDL scripts)

Commands that add new Adaptive Server logins and database users

The current Adaptive Server configuration file, as described in Using

sp_configure with a configuration file on page 68

The names, locations, and sizes of all files and raw devices initialized as
database devices

Maintain a dated log of all changes to the Adaptive Server configuration. Mark
each change with a brief description of when and why you made the change, as
well a summary of the end result.

Maintenance schedules
Keep a calendar of regularly scheduled maintenance activities; list any of the
procedures you perform at your site:

44

Using dbcc to check database consistency

Backing up user and system databases

Monitoring the space left in transaction logs (if this is not done
automatically)

Adaptive Server Enterprise

CHAPTER 3

System Administration for Beginners

Dumping the transaction log

Examining the error log contents for Adaptive Server, Backup Server, and
Adaptive Server Monitor

Running the update statistics command (see Chapter 1, Using the set
statistics Commands, in Performance and Tuning Series: Improving
Performance with Statistical Analysis)

Examining auditing information, if the auditing option is installed

Recompiling stored procedures

Monitoring resource utilization of the server machine

System information
Record information about the hardware and operating system on which you run
Adaptive Server, including:

Copies of operating system configuration files or start-up files

Copies of network configuration files (for example, the hosts and services
files)

Names and permissions for the Adaptive Server executable files and
database devices

Names and locations of the tape devices used for backups

Copies of operating system scripts or programs for automated backups,

starting Adaptive Server, or performing other administration activities

Disaster recovery plan

Consolidate the basic backup and recovery procedures, the guidelines in
Backup and recovery on page 39, and your personal experiences in
recovering data into a concise list of recovery steps tailored to your system.
This can be useful to both yourself and to other system administrators who may
need to recover a production system in the event of an emergency.

System Administration Guide: Volume 1

45

Additional resources

Additional resources
The amount of information for system administrators to learn may seem
overwhelming. There are several software tools that can help you learn and
facilitate basic administration tasks. These include Adaptive Server Monitor,
used for monitoring server performance and other activities, and Sybase
Central, which simplifies many administration tasks. There are also many
third-party software packages available designed to help system administrators
manage daily maintenance activities.

46

Adaptive Server Enterprise

CH A PTE R

Managing and Monitoring

Adaptive Server

Topic
Sybase Control Center for Adaptive Server
Adaptive Server Sybase Central Plug-in

Page
47
48

This chapter describes the GUI managing and monitoring tools for
Adaptive Server: This is meant merely as an overview to introduce you to
these tools. See the online help for a complete description.

Sybase Control Center for Adaptive Server

The Sybase Control Center for Adaptive Server is the preferred tool for
managing and monitoring the status and availability of Adaptive Server.
Sybase Control Center is a Web-based tool for monitoring and
administration. See the Installation Guide at
http://sybooks.sybase.com/nav/summary.do?prod=10680 for information
about downloading and installing Sybase Control Center.
You can use a single instance of Sybase Control Center to monitor and
manage multiple Adaptive Servers.
Sybase Control Center supports Adaptive Server Enterprise version
15.0.2 and later. See the Sybase Control Center documentation for a list of
supported features.

System Administration Guide: Volume 1

47

Adaptive Server Sybase Central Plug-in

The Sybase Control Center client/server architecture allows multiple clients to

monitor all Adaptive Servers in an enterprise through one or more Sybase
Control Center servers. Sybase Control Center for Adaptive Server provides
availability monitoring, historical performance monitoring, and administration
capabilities in a scalable Web application that is integrated with management
modules for other Sybase products. It offers shared, consolidated management
of heterogeneous resources from any location, e-mail alerts that provide stateand threshold-based notifications about availability and performance in real
time, and intelligent tools for spotting performance and usage trends, all via a
thin-client, rich Internet application (RIA) delivered through your Web
browser.
Use Sybase Control Center for Adaptive Server to gather statistics that over
time will give you powerful insight into patterns of use and the behavior of
databases, devices, caches, and processes on your servers. You can display
collected data as tables or graphs. By plotting results over any period of time
you choose, from a minute to a year, you can both see the big picture and focus
on the particulars. Detailed knowledge of how your servers have performed in
the past helps you ensure that Adaptive Server meets your needs in the future.
See the online help here for more information about using Sybase Control
Center at http://sybooks.sybase.com/nav/summary.do?prod=10680.

Adaptive Server Sybase Central Plug-in

Sybase Central is a graphical user interface (GUI) management tool. Sybase
Central accepts a variety of plug-ins that allow you to manage specific
Sybase products. The Adaptive Server plug-in allows you to manage Adaptive
Server and helps you perform complex administration tasks without the need
to remember the syntax of Transact-SQL commands or system stored
procedures. You can use the Adaptive Server plug-in to:

48

Manage multiple servers from one console You can manage all the
Adaptive Server installations from the Sybase Central main window.

Generate database definition language (DDL) You can generate DDL for
the objects in Adaptive Server.

Adaptive Server Enterprise

CHAPTER 4

Managing and Monitoring Adaptive Server

Visually represent objects You can see the databases and logins in each
Adaptive Server and the objects in each database, and windows expand
and contract to display information about databases and logins. The
Adaptive Server plug-in expands to display information about many
items, including:

Databases and tables

Disk devices

Active processes and locks

Logins and users

Data caches

ASE Replicator, Job Scheduler, and Messaging Services

Access to other utilities such as Interactive SQL (for sending queries

and displaying query results).

Navigate between related objects To get more information about a

database object related to the one whose property sheet you are displaying,
navigate directly through the displayed objects dialog box to the related
object.

Create a cluster The Adaptive Server plug-in allows you to create a

cluster if you have purchased Adaptive Server Cluster Edition. See the
online help and the User Guide to Clusters.

Using the Adaptive Server Plug-in

The Adaptive Server plug-in for Sybase Central provides you with an intuitive
and easy way to administer Adaptive Server Enterprise. Sybase Central
displays the Adaptive Server plug-in in its left-hand pane. Included in this pane
is a hierarchical list of folders that represent different objects the plug-in can
manage, including:

Viewing and changing the characteristics of the object

Creating another object:

Generating the SQL text for creating an object (which allows you to
reverse engineer Adaptive Server objects)

Deleting an object

Configuring Adaptive Server

System Administration Guide: Volume 1

49

Adaptive Server Sybase Central Plug-in

Managing:

Database devices

Proxy and temporary databases

Indexes

Partitions

Segments

Triggers

Logins and roles

Views

ASE Replicator

Configuring Adaptive Server jobs with Job Scheduler

Starting and stopping Adaptive Server

Executing queries

Logging SQL statements generated by the plug-in, based on a users

actions.

Starting and stopping Sybase Central

To start Sybase Central:

(UNIX) Move to the $SYBASE/shared/sybcentral600 directory and run

the scjview.sh script.

(Windows) Choose Programs | Sybase | Sybase Central v6 from the Start

menu, or move to the %SYBASE%\Shared\Sybase Central 6.0.0\ directory
and run the scjview.bat script.

To stop Sybase Central, select File | Exit.

Registering Adaptive Server Plug-in

The Adaptive Server plug-in is registered in Sybase Central as part of the
server installation. However, if Adaptive Server plug-in is not correctly
registered, you can manually register the Adaptive Server plug-in:

50

Adaptive Server Enterprise

CHAPTER 4

Managing and Monitoring Adaptive Server

On Unix, run $SYBASE/ASEP/bin/registerASEP.

On Windows, run %SYBASE%\ASEP\bin\registerASEP.bat

You can register the Adaptive Server plug-in manually by:

a

Select Register from Tools | Plug-ins. A registration wizard appears.

Select Register

Select Register a plug-in by specifying a plug-in registration file.

Click Browse.

Navigate to $SYBASE/ASEP/bin (%SYBASE%\ASEP\bin on

Windows) and select ASEPlugin.jpr. Follow the wizard to register the
Adaptive Server plug-in.

Performing common tasks

The following are some common tasks users perform with the Adaptive Server
plug-in.
For more information about all the following tasks, see the Adaptive Server
plug-in online help.
Starting and stopping
Adaptive Server

If the Unified Agent is monitoring Adaptive Server, you can start, stop, and
restart the server by right-clicking on the server and selecting Shutdown, Start,
or Restart.
If the Unified Agent is not monitoring Adaptive Server, you can shutdown the
server by selecting Shutdown.

Connecting to
Adaptive Server

You can connect to an Adaptive Server by any of these methods:

Select the Connect icon from the tool bar.

Right click on Adaptive Server Enterprise and select Connect from the
menu.

Right click on any server group and select Connect from the menu.
The connected server is displayed in the Default server group if the
connection is initiated from the Adaptive Server Enterprise folder or the
connect icon. The plug-in displays Connected to server in the
corresponding server group if the connection is initiated from the server
group.

System Administration Guide: Volume 1

51

Adaptive Server Sybase Central Plug-in

You can also specify a server to which you want to connect by any of the
following:

Creating a database

Specifying the servers host name and port number in the Connect dialog
box.

Selecting a pre-defined Adaptive Server from the server name dropdown

list. This drop down list is derived from the servers listed in the interfaces
file (UNIX) and sql.ini files (Windows) and LDAP servers.

Discover which Adaptive Servers are available by clicking on Find in the

Connect dialog. Before you can use this method, you must first define the
discovery servers in Server Discovery tab located in the Adaptive Server
Enterprise property page.

Before creating a database, make sure enough space is available on the

database devices you plan to use.
To create a database:

Right-click on the Add Database icon in the right-hand panel, or,

Select the Databases folder.

Choose File | New | Database or click on the Add Database option in the
Databases folder. The Create a New Database wizard opens. The Create a
New Database wizard asks for the following information:
Table 4-1: Inputs to create a new database wizard

Input
Database name

Description
Enter a name for the database

Database device
Database device
size

Specify the database device or devices on which to allocate the new database
Specify a size for each database device

Data or log
With override

Specify whether the database device will store data or the transaction log.
Specify with override if you want to store data and log on the same device.

For load

If you are creating the database so you can restore it from a backup, check the For Load check
box. This is the case only if you are recovering from media failure or if you are moving a
database from one location to another.
Specify whether to create a guest user in the database.

Guest account

If you do not enter a size, Adaptive Server allocates either the value of the
database size configuration variable or the size of the model database,
whichever is larger.

52

Adaptive Server Enterprise

CHAPTER 4

Managing and Monitoring Adaptive Server

If you have limited storage and must put the transaction log and the data on the
same logical device, specifying With Override allows Adaptive Server to
maintain the log on separate device fragments from the data.
You cannot remove or change a database device after creating the database
unless you first delete the database.
Warning! Deleting a database also deletes all its objects.
Deleting a database

Only the owner of a database can delete it.

To delete a database:
1

Select the database icon.

Choose Edit | Delete.

Confirm the deletion in the confirmation dialog box.

Note Sybase recommends that you back up the master database after you

delete a user database.

Adding a user

Database owners can add and delete users in the databases they own.
To create a user:
1

Expand the databases folder (select the + icon) and select the Users
folder.

Choose File | New | User.

The Add a New User wizard opens and asks for this information:
Table 4-2: Inputs to Add a New User wizard
Input
Name
Login name

Description
A name for the user. The name does not have to be the same as the
login.
Login to which this user is assigned.

Group

Optionally, assign a group to the user. Default: public

Note A user can be a member of one assigned group or the default public

group.

System Administration Guide: Volume 1

53

Adaptive Server Sybase Central Plug-in

You can also select the Users folder. In the right pane, double-click the Add
User icon.
Deleting a user

You cannot delete a user who owns objects. Since there is no command to
transfer ownership of objects, you must delete objects owned by a user before
you can delete the user. Also, you cannot delete a user who has granted
permissions to other users without first revoking the permissions with cascade.
If appropriate, re-grant the permissions to the other users.
Locking a login is a simple alternative to deleting a user.
To delete a user:
1

Select the user icon.

Choose Edit | Delete.

Confirm the deletion in the confirmation dialog box.

You can also select the user folder by right-clicking on the user icon and select
Delete.
Before you delete a user:

Creating a table

Revoke the users command and object permissions with cascade.

Re-grant the permissions to the other users, if appropriate.

Delete the users objects.

Only a database owner or a user with create table permission can create a table.
To create a table:
1

In a database you are working in, select the User Tables folder.

Choose File | New | Table or click on the Add Table icon in the User Tables
folder.
The Table Editor opens.

In the Name box, enter a name.

From the Owner list, choose an owner. The default is dbo.

You can also select the User Tables folder. In the right pane, double-click the
Add Table icon.

54

Adaptive Server Enterprise

CHAPTER 4

Deleting a table

Managing and Monitoring Adaptive Server

Before you delete a table, be sure that no other objects reference it. If any
objects reference it, edit those objects to avoid errors. To find out if other
objects reference a table, check its dependencies.
Note When you delete a table, Adaptive Server deletes the indexes and triggers

associated with the table and unbinds the rules or defaults that are bound to its
columns.
Only table owners can delete tables.
To delete a table:

Creating a server
group

Follow these steps:

Select the table icon.

Choose Edit | Delete.

Confirm the deletion in the confirmation dialog box, or,

You can also select the table by right-clicking on the table icon and
selecting Delete.

To create a server group:

1

Select Adaptive Server Enterprise

Choose File | New | Server Group

Follow the steps provided by the Create New Server Group wizard.

You can also add a server group by double-clicking on the Add Server Group
from the right-hand pane.
Getting server status

If the Unified Agent is monitoring Adaptive Server, check the server status by
any of the following:

Click on the server group to which the server belongs. Check the Status
column in the Details pane of the server group.

Click on the Adaptive Server Enterprise listed under Sybase Central, and
then click on Servers tab on the right hand side panel. The server status is
printed in the Status column.

System Administration Guide: Volume 1

55

Adaptive Server Sybase Central Plug-in

A green triangle on the lower right-hand side of the server icon indicates
that Adaptive Server is running. A red square indicates that Adaptive
Server is stopped.

Note By default, the Adaptive Server plug-in does not have Check Server
Status enabled. To enable Unified Agent to monitor Adaptive Server:

Getting the server log

Right click on Adaptive Server Enterprise and select Properties.

Select Preferences and check Enable Unified Agent (UA) related

features.

If the Unified Agent is monitoring Adaptive Server, retrieve the server log by
selecting the server and clicking on the Server Log tab in the right-hand pane.
The server log is retrieved based on how you have configured the filter for the
the server log. To configure the server log filtering, right-click on the server
and select Server Log Filter. By default, the Adaptive Server plug-in retrieves
the last 1000 lines from the server log. You can configure the server filter to
retrieve:

Logging SQL
statements

Executing SQL
statements

The entire log file.

The last n number of lines.

The log from the last n number of days.

The lines that match the regular expression

To log all SQL statements executed through the Adaptive Server plug-in:

Right click on a server and select Log SQL Statement.

Select whether you want SQL statements logged directly to a window or

to a file.

Execute SQL statements from within the Adaptive Server plug-in by using the
Interactive SQL query tool. To start the Interactive SQL tool, you can either:

Right-click the server on which you want to execute the SQL statements
and select Open Interactive SQL from the menu, or

Click on Adaptive Server Enterprise.

Click the Utilities tab on the right-hand pane and select Interactive SQL

Execute SQL statements simultaneously on a set of servers belonging to a

server group:
1

56

Right -click the server group and choose Execute SQL.

Adaptive Server Enterprise

CHAPTER 4

Managing and Monitoring Adaptive Server

Select the servers on which you want to execute the SQL statements

Click Execute.

The result set for each server is listed in the Result Set pane of the SQL
Execution dialog.
Viewing SQL
execution plan and
cost information

Use the Adaptive Server plug-in to view a GUI version of the SQL execution
plan for individual queries (much like a GUI version of showplan) and
execution plans for all queries in a stored procedure. This GUI display includes
nodes for each of the operators of the execution plan.
To get the GUI plan:
1

Start Interactive SQL.

Execute the query or stored procedure

Click on the plan tab in the Results pane of Interactive SQL

Select a query from the queries drop down list.

Click the Details tab to see the GUI plan of the selected query. Click on an
operator node to see the detailed statistics for that node.

Click on the XML tab to see an XML representation of the execution plan
for the selected query

Click on the Text tab to see the execution plan in a text format for the
submitted queries
For more information about Interactive SQL, see Starting Interactive
SQL on page 58.

Viewing and updating

object properties

View and modify the configuration of any object represented in the Adaptive
Server plug-in using the Property dialog.
To bring up the Property dialog:

Generate the SQL text

for creating an object

Click on the object you want to view or modify.

Right-click on the object and select Properties.

Select the appropriate tab to perform your task.

Make any modification in the Property dialog.

Click on Apply, OK, or Cancel.

Generate the SQL text required for creating an object, which allows you to
reverse engineer the object. To generate SQL text, right-click on the object and
select Generate DDL.

System Administration Guide: Volume 1

57

Adaptive Server Sybase Central Plug-in

Viewing and updating

Adaptive Server
configuration
parameters

View and update the Adaptive Server configuration parameters using the
Server Properties dialog.
1

Right click on the server and select Configuration in the menu

Select the functional group from the drop down list in the Show
Configuration Parameters

Find and select the parameter you want to view or update

Enter new valuing the value column if update is necessary

Click on Apply/OK/Cancel accordingly

Using Interactive SQL

Interactive SQL allows you to execute SQL statements, build scripts, and
display database data to the server. You can use it to:

Browse the information in a database.

Test SQL statements that you plan to include in an application.

Save query results to a file.

Edit data in result sets.

Load data into a database and carry out administrative tasks.

In addition, Interactive SQL can run command files or script files. For
example, you can build repeatable scripts to run against a database and then use
Interactive SQL to execute these scripts as batches.

Starting Interactive SQL

To start Interactive
SQL from Sybase
Central

To start Interactive SQL, either:

Select a database in Sybase Central and select File | Open Interactive SQL.
Interactive SQL connects to the database. You can also right-click on the
database and select Open Interactive SQL.
The menu item Open Interactive SQL opens a connection to a server.
However, when you select the menu item for a server, Interactive SQL
opens a connection to the default database for that server. When you select
a specific database from the Open Interactive SQL menu, Interactive SQL
opens to the selected database.

58

Adaptive Server Enterprise

CHAPTER 4

To start Interactive
SQL from the
command line

Managing and Monitoring Adaptive Server

Select Tools | Adaptive Server Enterprise | Open Interactive SQL to start

Interactive SQL without a connection to a server. The Connect dialog
appears.

How you start Interactive SQL from the command line depends on your
operating system.
If you start Interactive SQL independently, the Connect dialog appears, which
lets you connect to a database just as you would in Sybase Central.

For UNIX, change to the $SYBROOT/DBISQL/bin directory and enter:

dbisql

On Windows, change to the %SYBROOT%\DBISQL\bin directory and

enter:
dbisql.bat

In the Connection dialog, enter the information to connect to a database in

the Connect dialog box and click OK.

To open a new Interactive SQL window:

1

Choose Window | New Window. The Connect dialog appears.

In the Connect dialog, enter connection options, and click OK to connect.

The connection information (including the database name, your user ID,
and the database server) appears on the title bar above the SQL Statements
pane.

You can also connect to or disconnect from a database with the Connect and
Disconnect commands in the SQL menu, or by executing a connect or
disconnect statement in the SQL Statements pane.

System Administration Guide: Volume 1

59

Adaptive Server Sybase Central Plug-in

60

Adaptive Server Enterprise

CH A PTE R

Setting Configuration
Parameters

This chapter describes the Adaptive Server configuration parameters,

which are listed here alphabetically.
A configuration parameter is a user-definable setting that you set with
sp_configure. Configuration parameters are used for a wide range of
services, from basic to specific server operations, and for performance
tuning.
Topic
Overview
Using sp_configure

Page
61
66

sp_configure output
Named cache configuration parameters

77
79

sysconfigures and syscurconfigs tables

79

Configuration parameters

80

Overview
Configuration parameters are user-definable settings that control various
aspects of Adaptive Server behavior. Adaptive Server supplies default
values for all configuration parameters. Use configuration parameters to
tailor Adaptive Server for an installations particular needs.
Read this chapter carefully to determine which configuration parameters
you should reset to optimize server performance.
Warning! Change configuration parameters with caution. Arbitrary
changes in parameter values can adversely affect Adaptive Server
performance and other aspects of server operation.

System Administration Guide: Volume 1

61

Overview

The Adaptive Server configuration file

Adaptive Server stores the values of configuration parameters in a
configuration file, which is an ASCII text file. When you install a new
Adaptive Server, your parameters are set to the default configuration; the
default name of the file is server_name.cfg, and the default location of the file
is the Sybase Adaptive Server home directory ($SYBASE_ASE). Each time
you modify a configuration parameter, Adaptive Server creates a copy of the
outdated configuration file, using the naming convention server_name.001,
server_name.002, server_name.003...server_name.999. Adaptive Server
writes the new values to the file server_name.cfg or to a file name you specify
at start-up.

Modifying configuration parameters

Set or change configuration parameters in one of the following ways:

By executing sp_configure with the appropriate parameters and values,

By editing your configuration file and then invoking sp_configure with the
configuration file option, or

By specifying the name of a configuration file at start-up.

Configuration parameters are either dynamic or static. Dynamic parameters

take effect as soon as you execute sp_configure. Static parameters require
memory to be reallocated, so they take effect only after you have restarted
Adaptive Server. The description of each parameter in this chapter indicates
whether it is static or dynamic.
Adaptive Server writes the new value to the system table sysconfigures and to
the configuration file when you change the value. The current configuration
file and sysconfigures reflect configured values, not run values. The system
table syscurconfigs reflects current run values of configuration parameters.

Required roles for modifying configuration parameters

The roles required for using sp_configure:

Any user can execute sp_configure to display information about

parameters and their current values.
Only a system administrator or a system security officer can execute
sp_configure to modify configuration parameters.

62

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

Only a system security officer can execute sp_configure to modify values

for:

allow procedure grouping

allow remote access

allow sendmsg

allow updates to system tables

auditing

audit queue size

check password for digit

current audit table

enable ldap user auth

enable pam user auth

enable ssl

log audit logon failure

log audit logon success

maximum failed logins

minimum password length

msg confidentiality reqd

msg integrity reqd

secure default login

select on syscomments.text

SQL Perfmon Integration

syb_sendmsg port number

suspended audit when device full

systemwide password expiration

unified login required

use security services

System Administration Guide: Volume 1

63

Overview

Unit specification using sp_configure

sp_configure allows you to specify the value for configuration parameters in
unit specifiers. The unit specifiers are p or P for pages, m or M for megabytes,
and g or G for gigabytes. If you do not specify a unit, and you are configuring

a parameter that controls memory, Adaptive Server uses the logical page size
for the basic unit.
Note When you are configuring memory-related parameters, use only the P
(page size) parameter for your unit specification. If you use any other
parameter to configure memory related parameters, Adaptive Server may issue
an arithmetic overflow error message.

The syntax to indicate a particular unit specification is:

sp_configure "parameter name", 0, "p|P|k|K|m|M|g|G"

You must include the 0 as a placeholder.

You can use this unit specification to configure any parameter. For example,
when setting number of locks to 1024 you can enter:
sp_configure "number of locks", 1024

or:
sp_configure "number of locks", 0, "1K"

This functionality does not change the way in which Adaptive Server reports
sp_configure output.

Global versus session settings

Some Adaptive Server global settings are similar to session-level settings. For
example, both sp_configure and sp_passwordpolicy include a minimum
password length parameter.
If a global and session-level parameters configure the same setting, only the
session level setting applies to the login (login level settings override global
settings).
For example, if you set this global configuration configuration parameter:
sp_configure 'minimum password length' 12

And you attempt to add user joe to the server with a insufficiently long
password:

64

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

sp_adduser joe, joejoe

The command fails:

Msg 10317, Level 14, State 1:
Procedure 'sp_password', Line 118:
The specified password is too short. Passwords must be
at least 12 character(s) long.
Msg 17720, Level 16, State 1:
Procedure 'sp_password', Line 128:
Error: Unable to set the Password.

However, if you set sp_passwordpolicy to a shorter length, Adaptive Server

requires a shorter password length for this session:
sp_passwordpolicy 'set', 'minimum password length', 6

The command succeeds:

sp_addlogin joe, joejoe
Password correctly set.
Account unlocked.
New login created.
(return status = 0)

Adaptive Server tests all configuration limits independently, and the login
attempt must pass all the applicable limits to succeed.

Getting help information on configuration parameters

Use either sp_helpconfig or sp_configure to display information on a particular
configuration parameter. For example:
sp_helpconfig "number of open"
Configuration option is not unique.
option_name
config_value run_value
------------------------------ ------------ ----------number of open databases
12
12
number of open indexes
500
500
number of open objects
500
500
sp_helpconfig "number of open indexes"
number of open indexes sets the maximum number of indexes that can be open at
one time on SQL Server. The default value is 500.
Minimum Value Maximum Value Default Value Current Value Memory Used

System Administration Guide: Volume 1

65

Using sp_configure

------------- ------------- ------------- ------------- ----------100

2147483647
500
500
208
sp_configure "number of open indexes"
Parameter Name
----------------------number of open indexes

Default
------500

Memory Used
----------208

Config Value
-----------500

Run Value
--------500

See Chapter 3, Configuring Memory, in System Administration Guide:

Volume 2.

Using sp_configure
sp_configure displays and resets configuration parameters. You can restrict the
number of parameters that sp_configure shows by using sp_displaylevel to set

your display level to one of:

Basic

Intermediate

Comprehensive

For information about display levels, see User-defined subsets of the

parameter hierarchy: display levels on page 74. For information about
sp_displaylevel, see the Reference Manual: Procedures.
Table 5-1 describes the syntax for sp_configure. The information in the
Effect column assumes that your display level is set to comprehensive.
Table 5-1: sp_configure syntax
Command

sp_configure parameter, value

Effect
Displays all configuration parameters by group, their current values, their
default values, the value to which they have most recently been set, and
the amount of memory used by this particular setting.
Displays current value, default value, most recently changed value, and
amount of memory used by the specified parameter.
Resets parameter to value.

sp_configure parameter, 0, default

Resets the specified parameter to its default value.

sp_configure

sp_configure parameter

66

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

Command

Effect

sp_configure group_name

Displays all configuration parameters in group_name, their current values,

their default values, the values to which they were recently set, and the
amount of memory used by each.
Sets configuration parameters from the configuration file. See Using
sp_configure with a configuration file on page 68 for descriptions of the
parameters.

sp_configure configuration file, 0,

sub_command, file_name

Syntax elements
The commands in Table 5-1 use the following variables:

parameter is any valid Adaptive Server configuration parameter or

parameter substring.

value is any integer within the valid range for that parameter. (See the
descriptions of the individual parameters for valid range information.)
Parameters that work as toggles have only two valid values: 1 (on) and 0
(off).

group_name is the name of any group in the parameter hierarchy.

Parameter parsing
sp_configure parses each parameter (and parameter name fragment) as
%parameter%. A string that does not uniquely identify a particular
parameter returns values for all parameters matching the string.

The following example returns values for all configuration parameters that
include lock, such as lock shared memory, number of locks, lock promotion
HWM, server clock tick length, print deadlock information, and deadlock retries:
sp_configure "lock"
Note If you attempt to set a parameter value with a nonunique parameter name
fragment, sp_configure returns the current values for all parameters matching
the fragment and asks you to specify a unique parameter name.

System Administration Guide: Volume 1

67

Using sp_configure

Using sp_configure with a configuration file

Configure Adaptive Server either interactively, by using sp_configure as
described above, or noninteractively, by instructing Adaptive Server to read
values from an edited or restored version of the configuration file.
By making your changes from the configuration file, you can:

Replicate a specific configuration across multiple servers by using the

same configuration file.

Use a configuration file as a baseline for testing configuration values on

your server.

Use a configuration file to perform validation checking on parameter

values before actually setting the values.

Create multiple configuration files and switch between them as your

resource needs change.

For information on editing the file, see Editing the configuration file on page
70. For information on specifying the name of the configuration file at start-up,
see Starting Adaptive Server with a configuration file on page 71.

Naming tips for the configuration file

To work with a configuration file that has a name other than the default name,
keep the server_name part of the file name, and include at least one alphabetic
character in the extension (for example my_server.A001). Alternatively, you
can change the server_name part of the file name (for example,
A_my_server.001). Doing this avoids confusion with the backup configuration
files generated by Adaptive Server when you modify a parameter.

Using sp_configure to read or write the configuration file

The syntax for using the configuration file option with sp_configure is:
sp_configure "configuration file", 0, "subcommand", "file_name"

where:

configuration file including quotes, specifies that this command uses the
configuration file.

0 requiredfor backward compatibilityafter the configuration file

parameter.

68

subcommand is one of:

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

write creates a file named file_name with the current configuration.

If file_name already exists, a message is written to the error log; the
existing file is renamed using the convention server_name.001,
server_name.002, and so on. If you have changed a static parameter,
but you have not restarted your server, write displays the currently
running value for that parameter. If you do not specify a directory with
file_name, the file is written to the directory from which Adaptive
Server was started.

read performs validation checking on values contained in file_name

and reads those values that pass validation into the server. If any
parameters are missing from file_name, the current values for those
parameters are used.

If the value of a static parameter in file_name is different from its

current running value, read fails and a message is printed. However,
validation is still performed on the values in file_name.

verify performs validation checking on the values in file_name. This

is useful if you have edited the configuration file, as it prevents you
from attempting to configure your server with invalid configuration
values.

restore creates file_name with the most recently configured values.

If you have configured static parameters to new values, this
subcommand writes the configured, not the currently running, values
to the file. This is useful if all copies of the configuration file have
been lost and you must generate a new copy. If you do not specify a
directory with file_name, the file is written to the directory from
which Adaptive Server was started.

file_name specifies the configuration file to use in conjunction with any

subcommand. If you do not specify a directory as part of the file name, the

directory where Adaptive Server was started is used.

Examples

Example 1 Performs validation checking on the values in the file srv.config

and reads the parameters that pass validation into the server. Current run values
are substituted for values that do not pass validation checking:

sp_configure "configuration file", 0, "read", "srv.config"

Example 2 Creates the file my_server.config and writes the current
configuration values the server is using to that file:
sp_configure "configuration file", 0, "write", "my_server.config"

System Administration Guide: Volume 1

69

Using sp_configure

Editing the configuration file

The configuration file is an ASCII file that you can edit with any text editor that
can save files in ASCII format. The syntax for each parameter is:
parameter_name={value | DEFAULT}

where:

Examples

parameter_name is the name of the parameter you want to specify.

value is the numeric value for set parameter_name.

DEFAULT specifies that you want to use the default value for
parameter_name.

Example 1 This example specifies that the transaction can retry its attempt to
acquire a lock one time when deadlocking occurs during an index page split or
shrink:
deadlock retries = 1
Example 2 This example specifies that the default value for the parameter
cpu accounting flush interval should be used:
cpu accounting flush interval=DEFAULT

When you edit a configuration file, your edits are not validated until you check
the file using the verify option, read the file with the read option, or restart
Adaptive Server with that configuration file.
If all your configuration files are lost or corrupted, you can re-create one from
a running server by using the restore subcommand and specifying a name for
the new file. The parameters in the new file are set to the values with which
your server is currently running.

70

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

Permissions for configuration files

Configuration files are nonencrypted ASCII text files. By default, they are
created with read and write permissions set for the file owner, and read
permission set for all other users. If you created the configuration file at the
operating system level, you are the file owner; if you created the configuration
file from Adaptive Server, using the write or restore parameter, the file owner
is the user who started Adaptive Server. Usually, this is the user sybase. To
restrict access to configuration files, use your operating systems file
permission command to set read, write, and execute permissions as
appropriate.
Note You must set permissions accordingly on each configuration file created.

Backing up configuration files

Configuration files are not automatically backed up when you back up the
master database. They are operating system filesback them up in the same
way you back up your other operating system files.
Checking the name of the configuration file currently in use

Due to space limitations, sp_configure output truncates the name of the

configuration file. To see the full name of the configuration file, use:
select s1.value2
from syscurconfigs s1, sysconfigures s2
where s1.config = s2.config
and s2.name = "configuration file"

Starting Adaptive Server with a configuration file

By default, Adaptive Server reads the configuration file server_name.cfg in the
start-up directory when it starts. If this file does not exist, it creates a new file
and uses Adaptive Server defaults for all values.
You can start Adaptive Server with a specified configuration file. For more
information, see the Utility Guide.
If the configuration file you specify does not exist, Adaptive Server prints an
error message and does not start.

System Administration Guide: Volume 1

71

Using sp_configure

If the command is successful, the file server_name.bak is created. This file

contains the configuration values stored in sysconfigures prior to the time
sysconfigures was updated with the values read in from the configuration file
you specified. This file is overwritten with each subsequent start-up.
Configuration file errors

When there are errors in the configuration file, Adaptive Server may not start,
or may use default values.
Adaptive Server uses default values if:

There are illegal values. For example, if a parameter requires a numeric

value, and the configuration file contains a character string, Adaptive
Server uses the default value.

Values are below the minimum allowable value.

The parameter hierarchy

Configuration parameters are grouped according to the area of Adaptive Server
behavior they affect. This makes it easier to identify all parameters that you
might need to tune to improve a particular area of Adaptive Server
performance.
Although each parameter has a primary group to which it belongs, many have
secondary groups to which they also belong. For example, number of remote
connections belongs primarily to the network communication group, but it also
belongs secondarily to the memory use group. This reflects the fact that some
parameters have implications for a number of areas of Adaptive Server
behavior. sp_configure displays parameters in all groups to which they belong.
Table 5-2 lists the configuration parameter groups.
Table 5-2: Configuration groups
Parameter group

Configures Adaptive Server for

Backup/Recovery
Cache manager

Backing up and recovering data

Data and procedure caches

Component Integration Services

administration

Component Integration Services

DTM administration
Diagnostics

Distributed transaction management (DTM) facilities

Diagnostic principles

Disk I/O

Disk I/O

72

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

Parameter group

Configures Adaptive Server for

Error log
Extended stored procedures

Error log, and the logging of Adaptive Server events to the Windows event log
The behavior of extended stored procedures (ESPs).

General information
Java services

Basic system administration

Memory for Java in Adaptive Server
See the Java in Adaptive Server Enterprise manual for complete information
about Java in the database.

Languages

If you use method calls to JDBC, you may need to increase the size of the
execution stack available to the user. See stack size on page 260.
Languages, sort orders, and character sets

Lock manager
Memory use

Locking
Memory consumption

Metadata caches

Setting the metadata cache size for frequently used system catalog information.
The metadata cache is a reserved area of memory used for tracking information
on databases, indexes, or objects. The greater the number of open databases,
indexes, or objects, the larger the metadata cache size. For a discussion of
metadata caches in a memory-usage context, see Chapter 3, Configuring
Memory, in System Administration Guide: Volume 2.

Monitoring

Collecting monitoring information. By default, Adaptive Server does not collect

monitoring information.

O/S resources

See Chapter 2, Monitoring Tables, in the Performance and Tuning Guide:

Monitoring and Analyzing.
Communication between Adaptive Server and remote servers, and between
Adaptive Server and client programs
Use of operating system resources

Physical memory
Processors

Your machines physical memory resources

Processors in an SMP environment

Query Tuning
RepAgent thread administration

Query optimization
Replication via Replication Server

Shared Disk Cluster

SQL Server administration

Cluster Edition.
General Adaptive Server administration.

Security related
Unicode

Security-related features
Unicode-related features

User environment

User environments

Network communication

The syntax for displaying all groups and their associated parameters, and the
current values for the parameters, is:

System Administration Guide: Volume 1

73

Using sp_configure

sp_configure
Note The number of parameters returned by sp_configure depends on the value
to which you have your display level set. See User-defined subsets of the
parameter hierarchy: display levels on page 74.

The following is the syntax for displaying a particular group and its associated
parameter:
sp_configure "group_name"

For example, to display the disk I/O group, enter:

sp_configure "Disk I/O"
Group: Disk I/O
Parameter Name
Default Memory Used Config Value Run Value
unit
type
-------------------- ----------- ------------ -------------------------allow sql server async i/o
1
0
1
1
switch
static
diable disk mirroring
1
0
1
1
switch
static
disk i/o structures
256
0
256
256
number
dynamic
number of devices
10
0
10
10
number
dynamic
number of large I/O buffers
6
12352
6
6
number
dynamic
page utilization percent
95
0
95
95
percent
dynamic
Note If the server uses a case-insensitive sort order, sp_configure with no
parameters returns a list of all configuration parameters and groups in
alphabetical order with no grouping displayed.

User-defined subsets of the parameter hierarchy: display levels

Depending on how you use Adaptive Server, you may need to adjust some
parameters more frequently than others. It may be easier to work with a subset
of parameters.

74

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

The default display level is comprehensive. When you set your display level,
the setting persists across multiple sessions. However, you can reset it at any
time.

Basic shows only the most basic parameters, and is appropriate for
general server tuning.

Intermediate includes parameters that are somewhat more complex, in

addition to the basic parameters.

Comprehensive includes all the parameters, including the most complex

ones. This level is appropriate for users doing highly detailed server
tuning.

The syntax for showing your current display level is:

sp_displaylevel

To set the display level, use the following, where user_name is your Adaptive
Server login name:
sp_displaylevel user_name[, basic | intermediate | comprehensive]

The effect of the display level on sp_configure output

If your display level is set to either basic or intermediate, sp_configure returns
only a subset of the parameters that are returned when your display level is set
to comprehensive. For instance, if your display level is set to intermediate, and
you want to see the parameters in the languages group, enter:
sp_configure "Languages"

The output looks like this:

sp_configure
Group: Languages
Parameter Name
--------------default character set
default language id
. . .

Default Memory Used Config Value Run Value

------- ----------- ------------ --------1
0
1
1
0
0
0
0

Unit
---id
id

Type
---static
dyna

This represents only a subset of the parameters in the languages group; some
language parameters appear only when your display level is comprehensive.

System Administration Guide: Volume 1

75

Using sp_configure

Performance tuning with sp_configure and sp_sysmon

sp_sysmon monitors Adaptive Server performance and generates statistical
information that describes the behavior of your Adaptive Server system. See
the Performance and Tuning Series: Monitoring Adaptive Server with
sp_sysmon.

You can run sp_sysmon before and after using sp_configure to adjust
configuration parameters. The output gives you a basis for performance tuning
and allows you to observe the results of configuration changes.

Using configuration parameters in a clustered environment

For the Cluster Edition, Sybase supports both cluster-wide and instancespecific configuration. Cluster-wide configuration parameters are applied to all
instances in the cluster. Local configuration parameters are applied only to a
specified instance.

Local configuration overrides cluster-wide configuration.

If an instance-specific configuration has not been applied, the cluster-wide

configuration applies.

Some parameters, such as default character set id, cannot be applied to a

specific instance. These parameters can only be used over an entire cluster.

The cluster configuration file includes an instance-specific configuration

block. Parameter settings in the instance-specific block override cluster-wide
settings. For example:
max online engines = DEFAULT
[Instance:ase1]
max online engines = 5
[Instance:ase2]
max online engines = 3

See the Clusters Users Guide.

76

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

sp_configure output
The sample output below shows the type of information sp_configure prints if
your display level is comprehensive, and you execute sp_configure with no
parameters. The values it prints vary, depending on your platform and on what
values you have already changed.
sp_configure
Group: Configuration Options
Group: Backup/Recovery
Parameter Name
Default Memory Used Config Value Run Value
-------------------- ----------- ------------ --------allow remote access
1
0
1
1
print recovery info
0
0
0
0
recovery interval in m
5
0
5
5
...

Unit Type
---- ---switch dyn
switch dyn
minutes dyn

Note All configuration groups and parameters appears in output if your

display level is set to comprehensive.

Where:

The Default column displays the default value. If you do not explicitly
reconfigure a parameter, it retains its default value.

Memory Used shows the amount of memory, in kilobytes, used by the

parameter at its current value. Some related parameters draw from the
same memory pool. For instance, the memory used for stack size and stack
guard size is already accounted for in the memory used for number of user
connections. If you added the memory used by each of these parameters
separately, the sume is more than the amount actually used. Parameters
that share memory with other parameters are marked with a hash mark
(#).

Config Value displays the most recent value to which the configuration
parameter has been set. When you execute sp_configure to modify a
dynamic parameter:

The configuration and run values are updated.

The configuration file is updated.

The change takes effect immediately.

When you modify a static parameter:

System Administration Guide: Volume 1

77

sp_configure output

The configuration value is updated.

The configuration file is updated.

The change takes effect only when you restart Adaptive Server.

Run Value displays the value Adaptive Server is currently using. It

changes when you modify a dynamic parameters value and, for static
parameters, after you restart Adaptive Server.

Unit displays the unit value of the configuration parameter. Adaptive

Server displays information in the following units:

Name of unit

Unit description

number
clock ticks

Number of items.
Number of clock ticks.

microseconds

Number of microseconds.

milliseconds
seconds

Number of millisecond.s
Number of seconds.

minutes
hours

Number of minutes.
Number of hours.

bytes
days

Number of bytes.
Number of days.

kilobytes
megabytes

Number of kilobytes.
Number of megabytes.

memory pages (2K)

virtual pages (2K)

Number of 2K memory pages.

Number of 2K virtual pages.

logical pages

Number of logical pages. This value depends on the logical page size your server is using: 2,
4, 8, or 16K.

percent
ratio

Value of the configured parameter as a percentage.

Value of the configured parameter as a ratio.

switch
id

Value of the parameter is either TRUE (the parameter is turned on), or FALSE.
ID of the configured parameter you are investigating.

name

Character string name assigned to the run or configure value of the parameter. For example,
binary appears under the Run Value or Config Value column for the output of
sp_configure "lock scheme".

row

Number of rows

78

Type displays whether the configuration option is static or dynamic.

Changes to static parameters require that you restart Adaptive Server for
the changes to take effect. Changes to dynamic parameters take effect
immediately without having to restart Adaptive Server.

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

Named cache configuration parameters

The Named Cache configuration parameter group provides details for named
caches:

cache size - size of the cache. By default Adaptive Server creates 8MB
caches. Change this parameter dynamically with sp_cacheconfig, or

change the value in the server configuration file to have the change take
place after the next server restart.

cache status status of the cache. One of default data cache, log on
ly, mixed, or in-memory storage. The default data cache must have a
cache status of default data cache, and cannot be changed. cache
status for named caches can be log only, mixed, or, for in-memory
databases, in-memory storage (you cannot change the cache status for

in-memory databases).
You cannot dynamically change the cache status in a clustered
environment from log only on a local cache while other instance use a
different cache status.

cache replacement describes the cache replacement policy. For named

caches and default data caches, the replacement policy is strict LRU or
relaxed LRU. Change this parameter dynamically with sp_cacheconfig,

or change the value in the server configuration file to have the change take
place after the next server restart. The cache replacement policy must be
none for in-memory databases because they do not use buffer or page
replacement.

local cache partition number number of cache partitions. You may

partition a named cache into multiple cache partitions. The acceptable
values are 0, 2, 4, 8, 16, 32, 64 or 128. You cannot change the number of
cache partitions dynamically; you must restart Adaptive Server for the
change to take effect.

sysconfigures and syscurconfigs tables

The report displayed by sp_configure is constructed mainly from the
master..sysconfigures and master..syscurconfigs system tables, with additional
information provided from sysattributes, sysdevices, and other system tables.

System Administration Guide: Volume 1

79

Configuration parameters

The value column in the sysconfigures table records the last value set from
sp_configure or the configuration file; the value column in syscurconfigs stores
the value currently in use. For dynamic parameters, the two values match; for
static parameters, which require a restart of the server to take effect, the two
values are different if the values have been changed since Adaptive Server was
last started. The values may also be different when the default values are used.
In this case, sysconfigures stores 0, and syscurconfigs stores the value that
Adaptive Server computes and uses.
sp_configure performs a join on sysconfigures and syscurconfigs to display the
values reported by sp_configure.

Querying syscurconfigs and sysconfigures: an example

You might want to query sysconfigures and syscurconfigs to get information
organized the way you want. For example, sp_configure without any arguments
lists the memory used for configuration parameters, but does not list minimum
and maximum values. Use this query to get a complete list of current memory
usage, as well as minimum, maximum, and default values:
select b.name, memory_used, minimum_value,
maximum_value, defvalue
from master.dbo.sysconfigures b,
master.dbo.syscurconfigs c
where b.config *= c.config and parent != 19
and b.config > 100

Configuration parameters
In many cases, the maximum allowable values for configuration parameters are
usually limited by available memory, rather than by sp_configure limitations.
Note To find the maximum supported values for your platform and version of
Adaptive Server, see Adaptive Server Specifications in the Installation
Guide for your platform.

80

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

Alphabetical listing of configuration parameters

The following sections include both summary and detailed information about
each configuration parameter.

abstract plan cache

Summary information
Default value

0 (off)

Range of values
Status

0 (off), 1 (on)
Dynamic

Display level
Required role

Comprehensive
System administrator

Configuration group

Query Tuning

abstract plan cache enables caching of abstract plan hash keys. See Chapter 12,

Creating and Using Abstract Plans in the Performance and Tuning Series:
Query Processing and Abstract Plans. abstract plan load must be enabled for
plan caching to take effect.

abstract plan dump

Summary information
Default value
Range of values

0 (off)
0 (off), 1 (on)

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
Query Tuning

abstract plan dump enables the saving of abstract plans to the ap_stdout abstract
plans group. See Chapter 12, Creating and Using Abstract Plans in the
Performance and Tuning Series: Query Processing and Abstract Plans.

abstract plan load

Summary information
Default value
Range of values

System Administration Guide: Volume 1

0 (off)
0 (off), 1 (on)

81

Configuration parameters

Summary information
Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
Query Tuning

abstract plan load enables association of queries with abstract plans in the
ap_stdin abstract plans group. See Chapter 12, Creating and Using Abstract

Plans in the Performance and Tuning Series: Query Processing and Abstract
Plans.

abstract plan replace

Summary information
Default value
Range of values

0 (off)
0 (off), 1 (on)

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
Query Tuning

abstract plan replace enables plan replacement for abstract plans in the
ap_stdout abstract plans group. See Chapter 12, Creating and Using Abstract

Plans in the Performance and Tuning Series: Query Processing and Abstract
Plans. abstract plan load must be enabled for replace mode to take effect.

additional network memory

Summary information

82

Default value
Range of values

0
0 2147483647

Status
Display level

Dynamic
Intermediate

Required role
Configuration groups

System administrator
Memory Use, Network Communication, Physical
Memory

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

additional network memory sets the maximum size of additional memory that

can be used for network packets that are larger than the default packet size.
Adaptive Server rounds down the value you enter to the nearest 2K value. The
default value indicates that no extra space is allocated for large packets.
When a login requests a large packet size, Adaptive Server verifies it has
sufficient memory available to satisfy the request. If it does not, Adaptive
Server finds the largest available block of memory and tries the appropriate
size (which is a multiple of default network packet size) less than the largest
memory block. If that fails, Adaptive Server decreases the value of the request
by the number of bytes equal to default network packet size, if this is available.
Adaptive Server continues for 10 iterations, or until the size equals the value of
default network packet size, whichever comes first. On the tenth iteration,
Adaptive Server uses the value of the default network packet size for the packet
size.
If you increase max network packet size, you must increase additional network
memory because all allocated network memory is reserved for users at the
default size. Adaptive Server guarantees that every user connection can log in
at the default packet size.
If you increase max network packet size but do not increase additional network
memory, Adaptive Server does not guarantee that clients who request network
packet sizes larger than the default size can login at the requested packet size.
Increasing additional network memory may improve performance for
applications that transfer large amounts of data. To determine the value for
additional network memory when your applications use larger packet sizes:
1

Estimate the number of simultaneous users who will request the large
packet sizes, and the sizes their applications will request,

Multiply this sum by three, since each connection needs three buffers,

Add two percent for overhead for 32-bit servers, or four percent for 64-bit
servers, and

Round the value to the next highest multiple of 2048.

For example, if you estimate these simultaneous needs for larger packet sizes:
Application
bcp

Packet size
8192

Client-Library
Client-Library

8192
4096

Client-Library
Total

4096
24576

System Administration Guide: Volume 1

Overhead

83

Configuration parameters

Application

Packet size

Multiply by 3 buffers/user
Compute 2% overhead

* 3=73728

Add overhead
Additional network memory

+ 1474
75202

Round up to multiple of 2048

75776

Overhead
* .02=1474

Set additional network memory to 75,776 bytes.

allocate max shared memory

Summary information
Default value

0 (off)

Range of values
Status

0 (off), 1 (on)
Dynamic

Display level
Required role

Basic
System administrator

Configuration groups

Memory Use, Physical Memory

allocate max shared memory determines whether Adaptive Server allocates all
the memory specified by max memory at start-up or only the amount of memory
the configuration parameter requires.

By setting allocate max shared memory to 0, you ensure that Adaptive Server
uses only the amount of shared memory required by the current configuration,
and allocates only the amount of memory required by the configuration
parameters at start-up, which is a smaller value than max memory.
If you set allocate max shared memory to 1, Adaptive Server allocates all the
memory specified by max memory at start-up. If you set allocate max shared
memory to 1, and if you increase max memory, Adaptive Server attempts to
allocate the memory immediately. If the memory allocation fails, Adaptive
Server writes messages to the error log. Check the error log to verify that no
errors have occurred.
A successful memory allocation means that Adaptive Server always has the
memory required for any memory configuration changes you make and there
is no performance degradation while the server readjusts for additional
memory. However, if you do not predict memory growth accurately, and max
memory is set to a large value, you may waste total physical memory.

84

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

allow backward scans

Summary information
Default value
Valid values

1 (on)
0 (off), 1 (on)

Status
Display level

Dynamic
Intermediate

Required role
Configuration group

System administrator
Query Tuning

allow backward scans controls how the optimizer performs select queries that
contain the order by...desc command:

When the value is set to 1, the optimizer can access the index or table rows
by following the page chain in descending index order.

When the value is set to 0, the optimizer selects the rows into a worktable
by following the index page pointers in ascending order, and then sorts the
worktable in descending order.

The first methodperforming backward scanscan speed access to tables

that need results ordered by descending column values. Some applications,
however, may experience deadlocks due to backward scans. In particular, look
for increased deadlocking if you have delete or update queries that scan
forward using the same index. There may also be deadlocks due to page splits
in the index.
Use print deadlock information to send messages about deadlocks to the error
log. See print deadlock information on page 234. Alternatively, you can use
sp_sysmon to check for deadlocking. See the Performance and Tuning Series:
Locking and Concurrency Control.

allow nested triggers

Summary information
Default value
Valid values

1 (on)
0 (off), 1 (on)

Status
Display level

Static
Intermediate

Required role
Configuration group

System administrator
SQL Server Administration

System Administration Guide: Volume 1

85

Configuration parameters

allow nested triggers controls the use of nested triggers. When the value is set
to 1, data modifications made by triggers can fire other triggers. Set allow
nested triggers to 0 to disable nested triggers. A set option, self_recursion,

controls whether the modifications made by a trigger can cause that trigger to
fire again.

allow procedure grouping

Summary information
Default value
Range of values

1 (on)
0 (off), 1 (on)

Status

Dynamic

Display level
Required role

Comprehensive
System security officer

Configuration group

Security Related

allow procedure grouping controls the ability to group stored procedures of the
same name so that they can be dropped with a single drop procedure statement.

allow remote access

Summary information
Default value

1 (on)

Valid values
Status

0 (off), 1 (on)
Dynamic

Display level
Required role

Intermediate
System security officer

Configuration groups

Backup/Recovery, Network Communication

allow remote access controls logins from remote Adaptive Servers. The default

value of 1 allows Adaptive Server to communicate with Backup Server.

Note Setting the value to 0 disables server-to-server RPCs. Since Adaptive

Server communicates with Backup Server via RPCs, setting this parameter to
0 makes it impossible to back up a database.
Since other system administration actions are required to enable remote servers
other than Backup Server to execute RPCs, leaving this option set to 1 does not
constitute a security risk.

86

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

allow resource limits

Summary information
Default value
Valid values

0 (off)
0 (off), 1 (on)

Status
Display level

Static
Comprehensive

Required role
Configuration groups

System administrator
Memory Use, SQL Server Administration

allow resource limits controls the use of resource limits. When the value is set to

1, the server allocates internal memory for time ranges, resource limits, and
internal server alarms. The server also internally assigns applicable ranges and
limits to user sessions. The output of showplan and statistics io displays the
optimizers cost estimate for a query. Set allow resource limits to 0 to disable all
resource limits.

allow sendmsg
Summary information
Default value
Valid values

0 (off)
0 (off), 1 (on)

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System security officer

Network Communication

allow sendmsg enables or disables sending messages from Adaptive Server to

a User Datagram Protocol (UDP) port. When allow sendmsg is set to 1, any user
can send messages using sp_sendmsg or syb_sendmsg. To set the port number

used by Adaptive Server, see syb_sendmsg port number on page 267.

Note Sending messages to UDP ports is not supported on Windows.

allow sql server async i/o

Summary information
Default value

System Administration Guide: Volume 1

1 (on)

87

Configuration parameters

Summary information
Valid values
Status

0 (off), 1 (on)
Static

Display level
Required role

Comprehensive
System administrator

Configuration group

Disk I/O

allow sql server async i/o enables Adaptive Server to run with asynchronous
disk I/O. To use asynchronous disk I/O, enable it on both Adaptive Server and
your operating system. See your operating system documentation for
information on enabling asynchronous I/O at the operating system level.

Disk I/O always runs faster asynchronously than synchronously. This is

because when Adaptive Server issues an asynchronous I/O, it does not have to
wait for a response before issuing further I/Os.

allow updates to system tables

Summary information
Default value

0 (off)

Valid values
Status

0 (off), 1 (on)
Dynamic

Display level
Required role

Comprehensive
System administrator

Configuration group

SQL Server Administration

allow updates to system tables enables users with the system administrator role
to make changes to the system tables and to create stored procedures that can
modify system tables. A database administrator can update system tables in
any tables that he or she owns if allow updates to system tables is enabled.

System tables include:

88

All Sybase-supplied tables in the master database

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

All tables in user databases that begin with sys and that have an ID value
in the sysobjects table of less than or equal to 100
Warning! Incorrect alteration of a system table can result in database
corruption and loss of data. To protect against errors that might corrupt
your databases, always use begin transaction when modifying a system
table. Immediately after finishing your modifications, disable allow
updates to system tables.

Stored procedures and triggers you create while allow updates to system tables
is set on can update the system tables, even after the parameter has been set off.
When you set allow updates to system tables to on, you create a window of
vulnerability, a period of time during which users can alter system tables or
create a stored procedure with which the system tables can be altered in the
future.
Because the system tables are so critical, Sybase suggests that you set this
parameter to on only in highly controlled situations. To guarantee that no other
users can access Adaptive Server while the system tables can be directly
updated, restart Adaptive Server in single-user mode. For details, see
startserver and dataserver in the Utility Guide.
Note The server-wide configuration option allow updates to system tables takes
precedence over the stored procedure settings for allow updates to system
tables. If you do not enable allow updates to system tables at the server level,

individual stored procedure settings determine whether you can modify system
catalogs.

average cap size

Summary information
Default value

200

Range of values
Status

100 10000
Static

Display level
Required role
Configuration group

Diagnostics

Reserved for future use.

System Administration Guide: Volume 1

89

Configuration parameters

audit queue size

Summary information
Default value
Range of values

100
1 65535

Status
Display level

Dynamic
Intermediate

Required role
Configuration groups

System security officer

Memory Use, Security Related

The in-memory audit queue holds audit records generated by user processes
until the records can be processed and written to the audit trail. To change the
size of an audit queue, a system security officer can use audit queue size. When
you configure the queue suze, there is a trade-off between performance and
risk. If the queue is too large, records can remain in it for some time. As long
as records are in the queue, they are at risk of being lost if the system fails.
However, if the queue is too small, it can repeatedly become full, which affects
overall system performance; user processes that generate audit records sleep if
the audit queue is full.
Following are some guidelines for determining how big your audit queue
should be. You must also take into account the amount of auditing to be
performed at your site.

The memory requirement for a single audit record is 424 bytes; however,
a record can be as small as 22 bytes when it is written to a data page.

The maximum number of audit records that can be lost in a system failure
is the size of the audit queue (in records), plus 20. After records leave the
audit queue, they remain on a buffer page until they are written to the
current audit table on the disk. The pages are flushed to disk every 20
records, less if the audit process is not constantly busy.

In the system audit tables, the extrainfo field and fields containing names
are of variable length, so audit records that contain complete name
information are generally larger.

The number of audit records that can fit on a page varies from 4 to as many as
80 or more. The memory requirement for the default audit queue size of 100 is
approximately 42K.

90

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

auditing
Summary information
Default value
Range of values

0 (off)
0 (off), 1 (on)

Status
Display level

Dynamic
Intermediate

Required role
Configuration group

System security officer

Security Related

auditing enables or disables auditing for Adaptive Server.

automatic cluster takeover

Summary information
Default value
Valid values

1
1 (enabled), 0 (disabled)

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
Shared Disk Cluster

Setting automatic cluster takeover to 1 allows an instance that is starting to

automatically recover from an abrupt total cluster failure. If you set automatic
cluster takeover to 0, the cluster may not be able to recover from an abrupt
cluster failover unless you include the --cluster_takeover parameter.
The Cluster Edition uses quorum heartbeats and a cluster takeover algorithm to
determine when cluster takeover should be performed. This algorithm allows
an instance that is starting to distinguish between an inability to join the cluster
because the cluster has crashed (in which case takeover is appropriate) and an
inability to join the cluster because the instance that is starting does not have
network connectivity (in which case takeover is not appropriate).
If you disable automatic cluster takeover (set it to 0), The Cluster Edition writes
the results of the algorithm to the error log as an advisory message and then
exits.
If you enable auotomatic cluster takeover (set it to 1), the Cluster Edition starts
as the cluster coordinator and recovers the databases. This is guaranteed to be
a safe operation in environments that have I/O fencing enabled.

System Administration Guide: Volume 1

91

Configuration parameters

In environments without I/O fencing, a malfunction of the algorithm could

introduce data corruption, so you can set the configuration parameter to 0 to
disable this algorithm. However, environments without I/O fencing have a risk
of data corruption, and disabling automatic cluster takeover does not mitigate
all of those risks.

builtin date strings

Summary information
Default value
Range of values

0
01

Status

Dynamic

Display level
Required role
Configuration group

Query tuning

If a string is given as an argument in place of the chronological value, the server

interprets it as a datetime value regardless of its apparent precision. This default
behavior may be changed by setting the configuration parameter builtin date
strings or the set option builtin_date_strings. When these options are set the
server will interpret strings given to chronological builtins as bigdatetimes.

caps per ccb

Summary information
Default value

50

Range of values
Status

5 50
Static

Display level
Required role
Configuration group

Diagnostics

Reserved for future use.

capture compression statistics

Summary information
Default value

92

0 (off)

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

Summary information
Range of values
Status

1 (on), 0 (off)
Dynamic

Display level
Required role

10
System administrator

Configuration group

SQL Server Administration

Enables the monTableCompression monitoring table to begin capturing

compression statistics.

check password for digit

Summary information
Default value

0 (off)

Range of values
Status

1 (on), 0 (off)
Dynamic

Display level
Required role

10
System security officer

Configuration group

Security Related

The system security officer can tell the server to check for at least one character
or digit in a password using the server-wide configuration parameter check
password for digit. If set, this parameter does not affect existing passwords.

CIPC large message pool size

Summary information
Default value

512

Valid values
Status

512 2147483647
Static

Display level
Required role

Comprehensive
System administrator

Configuration group

Shared Disk Cluster

CIPC large message pool size specifies the number of large message buffers
allocated by CIPC at start-up time.

System Administration Guide: Volume 1

93

Configuration parameters

CIPC regular message pool size

Summary information
Default value
Valid values

8192
2048 2147483647

Status
Display level

Static
Comprehensive

Required role
Configuration group

System administrator
Shared Disk Cluster

CIPC regular message pool size specifies the number of regular message buffer
allocated by CIPC at start-up time.

cis bulk insert array size

Summary information
Default value
Range of values

50
0 2147483647

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
Component Integration Services

When performing a bulk transfer of data from one Adaptive Server to another
Adaptive Server, CIS internally buffers rows, and asks the Open Client bulk
library to transfer them as a block. The size of the array is controlled by cis bulk
insert array size.

cis bulk insert batch size

94

Summary information
Default value

Range of values
Status

0 2147483647
Dynamic

Display level
Required role

Comprehensive
System administrator

Configuration group

Component Integration Services

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

cis bulk insert batch size determines how many rows from the source tables are
to be bulk copied into the target table as a single batch using select into.

If you leave cis bulk insert batch size at 0, all rows are copied as a single batch.
Otherwise, after the count of rows specified by this parameter has been copied
to the target table, the server issues a bulk commit to the target server, causing
the batch to be committed.
If a normal client-generated bulk copy operation (such as that produced by the
bcp utility) is received, the client is expected to control the size of the bulk
batch, and the server ignores the value of this configuration parameter.

cis connect timeout

Summary information
Default value

Range of values
Status

032767
Dynamic

Display level
Required role

Comprehensive
System administrator

Configuration group

Component Integration Services

cis connect timeout determines the wait time, in seconds, for a successful

Client-Library connection.

cis cursor rows

Summary information
Default value

50

Range of values
Status

1 2147483647
Dynamic

Display level
Required role

Comprehensive
System administrator

Configuration group

Component Integration Services

cis cursor rows specifies the cursor row count for cursor open and cursor fetch

operations. Increasing this value means more rows are fetched in one
operation. This increases speed but requires more memory.

System Administration Guide: Volume 1

95

Configuration parameters

cis idle connection timeout

Summary information
Default value
Range of values

0
0 2147483647

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
Component Integration Services

cis idle connection timeout configures Adaptive Server to check for CIS
connections to any remote server that have been unused longer than the
specified number of seconds. Adaptive Server deletes the unused connections
and reallocates their resources.

Although the number you specify is in seconds, the housekeeper task wakes up,
at most , once a minute, so idle connections may be idle for much longer than
the configured value. Adaptive Server does not drop idle connections if a
transaction is active on the connection, and reestablishes the connection
automatically if the user executes any command that accesses the connection.

cis packet size

Summary information
Default value
Range of values

512
51232768

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
Component Integration Services

cis packet size specifies the size of Tabular Data Stream (TDS) packets that

are exchanged between the server and a remote server when a connection is
initiated.
The default packet size on most systems is 512 bytes, and this may be adequate
for most applications. However, larger packet sizes may result in significantly
improved query performance, especially when text, unitext, and image or bulk
data is involved.
If you specify a packet size larger than the default, then the target server must
be configured to allow variable-length packet sizes, using:

96

Adaptive Server Enterprise

CHAPTER 5

additional netmem

maximum network packet size

Setting Configuration Parameters

cis rpc handling

Summary information
Default value
Valid values

0 (off), default value of 1 for the Cluster Edition

0 (off), 1 (on)

Status
Display level

Dynamic
Comprehensive

Required role

System administrator

Configuration group

Component Integration Services

cis rpc handling specifies the default method for remote procedural call (RPC)
handling. Setting cis rpc handling to 0 sets the Adaptive Server site handler as

the default RPC handling mechanism. Setting the parameter to 1 forces RPC
handling to use Component Integration Service access methods. See set cis rpc
handling in the Component Integration Services Users Guide.

cluster heartbeat interval

Summary information
Default value

10

Valid values
Status

1 127
Dynamic

Display level
Required role

Comprehensive
System administrator

Configuration group

Shared Disk Cluster

cluster heartbeat interval controls the interval that cluster instances use to send
and check the heartbeat status.

Using a lower value for cluster heartbeat interval reduces the failure detection
time but increases the risk of a false failure because of a transient problem
(such as an overloaded CPU). Tuning cluster heartbeat interval to a larger value
reduces the risk of a false failure but increases the time needed to detect a
failure.

System Administration Guide: Volume 1

97

Configuration parameters

cluster heartbeat retries

Summary information
Default value
Valid values

1
1 127

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
Shared Disk Cluster

cluster heartbeat retries controls the number of times an instance retries a failed
cluster heartbeat before entering failure mode.

Tuning cluster heartbeat retries to a lower value reduces the time to detect
failure but increases the risk of a false failure because of a transient problem
(such as an overloaded CPU). Tuning cluster heartbeat retries to a larger value
reduces the risk of a false failure but increases the time needed to detect a
failure.

cluster redundancy level

Summary information
Default value
Valid values

1
1 to n 1, where n is the maximum number of
instances specified in cluster.cfg or the quorum file.

Status
Display level

Static
Comprehensive

Required role
Configuration group

System administrator
Shared Disk Cluster

cluster redundancy level (CRL) controls the number of recoverable failed

instances in a shared-disk cluster. It is the maximum number of instances that

can fail simultaneously while allowing recovery to proceed concurrently with
other activity. If the number of failed instances exceeds the maximum number
specified by the CRL, the cluster shuts down.
For the cluster to start, the value for cluster redundancy level must be at least
one less than the value of maximum number of instances specified in
cluster.cfg or the quorum file. Thus, the cluster cannot start if you set either of
these:

98

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

The value of maximum number of instances to a value that is equal to or

less than the value of cluster redundancy level

The value of cluster redundancy level to a value equal to or greater than the
value of maximum number of instances

When the cluster redundancy level value increases, so does the messaging
traffic, as there are multiple copies of each lock in the cluster, and an increase
in overhead required to maintain this redundancy level.
Other configuration parameters, such as number of locks and cache size need
more resources for cluster redundancy level values greater than 1, which means
you must increase max memory for the same number of locks value.
Sybase recommends that you set cluster redundancy level to 1 if the failure of
multiple instances is expected to be uncommon.

cluster vote timeout

Summary information
Default value

60

Valid values
Status

1 127
Dynamic

Display level
Required role

Comprehensive
System administrator

Configuration group

Shared Disk Cluster

cluster vote timeout controls the maximum amount of time an instance waits for
other instances to vote during the voting period. An instance waits only for
those instances which it believes are running.

Tuning cluster vote timeout to a lower value can reduce failover time, but
increases the risk that an instance that is running is excluded from the new
cluster view. Tuning cluster vote timeout to a larger value reduces the risk that
an running instance is excluded from the new cluster view, but may increase
failover time.

column default cache size

Summary information
Default value

1024 pages (2MB)

Range of values

128 8192 (pages)

System Administration Guide: Volume 1

99

Configuration parameters

Summary information
Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
User Environment

column defult cache size determines the size of the cache that Adaptive Server
must keep in memory to provide defaults for nonmaterialized columns.

compression info pool size

Summary information
Default value
Range of values

4096
0 2147483647

Status
Display level

Dynamic
10

Required role
Configuration group

System administrator
SQL Server Administration

Determines the size of the memory pool used for compression.

compression memory size

Summary information
Default value
Range of values

0
0 2147483647

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
Physical Memory

Used while loading a compressed dump into an archive database. compression

memory size determines the size (in 2KB pages) of the memory pool Adaptive
Server uses to decompress a compressed dump. When you set compression
memory size to 0, no pool is created and a compressed dump cannot be loaded.
See Creating a compression memory pool in Chapter 14, Archive Database
Access, in the System Administration Guide: Volume 2.

100

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

configuration file
Summary information
Default value
Range of values

0 (off)
One of: 0, verify, read, write, or restore

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
General Information

configuration file specifies the location of the configuration file currently in use.
See Using sp_configure with a configuration file on page 68 for a complete
description of configuration files.

In sp_configure output, the Run Value column displays only 10 characters, so

the output may not display the entire path and name of your configuration file.

cost of a logical io
Summary information
Default value
Range of values

2
0 254

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
Query Tuning

cost of a logical io specifies the cost of a single logical I/O.

cost of a physical io
Summary information
Default value

25

Range of values
Status

0 254
Dynamic

Display level
Required role

Comprehensive
System administrator

Configuration group

Query Tuning

System Administration Guide: Volume 1

101

Configuration parameters

cost of a phsyical io specifies the cost of a single physical I/O.

cost of a cpu unit

Summary information
Default value
Range of values

1000
1 65534

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
Query Tuning

cost of a cpu unit specifies the cost of a single CPU operation.

The cost of a serial plan in the optimizer is described by this formula:

Cost = PIO X estimated_pio + LIO X estimated_lio + 100 X estimated_cpu / CPU

Where the default values are:

estimated_pio = 25

estimated_lio = 2

estimated_cpu = 1000

If your Adaptive Server has sufficient memory, then all tables exist in memory,
and a value of 0 for cost of a physical io is appropriate.
If your CPU is fast enough so the value for cost of a cpu unit is not a issue, use
this formula to determine the cost of CPU, which combines 2 LIO and 25 PIO
(the default values):
CPU X 100/configuration_value

The default value for configuration_value is 1000.

As you increase the value for cost of a cpu unit, this formula reduces the impact
of CPU on cost.

cpu accounting flush interval

Summary information
Default value
Range of values

102

200
12147483647

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

Summary information
Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
SQL Server Administration

cpu accounting flush interval specifies the amount of time, in machine clock
ticks (non-Adaptive Server clock ticks), that Adaptive Server waits before
flushing CPU usage statistics for each user from sysprocesses to syslogins, a
procedure used in charge-back accounting.

When a user logs in to Adaptive Server, the server begins accumulating figures
for CPU usage for that user process in sysprocesses. When a user logs off
Adaptive Server, or when the value of cpu accounting flush interval is exceeded,
the accumulated CPU usage statistics are flushed from sysprocesses to
syslogins. These statistics continue accumulating in syslogins until you clear
the totals. Display the current totals from syslogins using sp_reportstats.
The value to which you set cpu accounting flush interval depends on the type of
reporting you intend to do. If you run reports on a monthly basis, set cpu
accounting flush interval to a relatively high value. With infrequent reporting, it
is less critical that the data in syslogins be updated frequently.
However, if you perform periodic ad hoc selects on the totcpu column in
syslogins to determine CPU usage by process, set cpu accounting flush interval
to a lower value to increase the likelihood of the data in syslogins being up-todate when you execute your selects.
Setting cpu accounting flush interval to a low value may cause the lock manager
to mistakenly identify processes as potential deadlock victims. When the lock
manager detects a deadlock, it checks the amount of CPU time accumulated by
each competing processes. The process with the lesser amount is chosen as the
deadlock victim and is terminated by the lock manager. Additionally, when cpu
accounting flush interval is set to a low value, the task handlers that store CPU
usage information for processes are initialized more frequently, thus making
processes appear as if they have accumulated less CPU time than they actually
have. Because of this, the lock manager may select a process as the deadlock
victim when, in fact, that process has more accumulated CPU time than the
competing process.
If you do not intend to report on CPU usage at all, set cpu accounting flush
interval to its maximum value. This reduces the number of times syslogins is
updated, and reduces the number of times its pages must be written to disk.

System Administration Guide: Volume 1

103

Configuration parameters

cpu grace time

Summary information
Default value
Range of values

500
02147483647

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
SQL Server Administration

cpu grace time, together with time slice, specifies the maximum amount of time
that a user process can run without yielding the CPU before Adaptive Server
preempts it and terminates it with a timeslice error. The units for cpu grace time
are time ticks, as defined by sql server clock tick length. See sql server clock
tick length on page 256.

When a process exceeds cpu grace time Adaptive Server infects it by

removing the process from the internal queues. The process is killed, but
Adaptive Server is not affected. This prevents runaway processes from
monopolizing the CPU. If any of your user processes become infected, you
may be able to temporarily fix the problem by increasing the value of cpu grace
time. However, be sure that the problem really is a process that takes more than
the current value of cpu grace time to complete, rather than a runaway process.
Temporarily increasing the cpu grace time value is a workaround, not a
permanent fix, since it may cause other complications; see time slice on page
271. Also, see Chapter 4, Using Engines and CPUs in the Performance and
Tuning Series: Basics for a more detailed discussion of task scheduling.

current audit table

Summary information
Default value
Range of values

1
08

Status
Display level

Dynamic
Intermediate

Required role
Configuration group

System security officer

Security Related

current audit table establishes the table where Adaptive Server writes audit

rows. A system security officer can change the current audit table, using:

104

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

sp_configure "current audit table", n

[, "with truncate"]

where n is an integer that determines the new current audit table, as follows:

1 means sysaudits_01, 2 means sysaudits_02, and so forth, up to 8.

0 tells Adaptive Server to set the current audit table to the next table. For
example, if your installation has three audit tables, sysaudits_01,
sysaudits_02, and sysaudits_03, Adaptive Server sets the current audit
table to:

2 if the current audit table is sysaudits_01

3 if the current audit table is sysaudits_02

1 if the current audit table is sysaudits_03

"with truncate" specifies that Adaptive Server should truncate the new table if it
is not already empty. sp_configure fails if this option is not specified and the

table is not empty.

Note If Adaptive Server truncates the current audit table, and you have not
archived the data, the tables audit records are lost. Be sure that the audit data
is archived before using the with truncate option.

To execute sp_configure to change the current audit table, you must have the
sso_role active. You can write a threshold procedure to change the current audit
table automatically.

deadlock checking period

Summary information
Default value

500

Range of values
Status

02147483
Dynamic

Display level
Required role

Comprehensive
System administrator

Configuration group

Lock Manager

System Administration Guide: Volume 1

105

Configuration parameters

deadlock checking period specifies the minimum amount of time (in

milliseconds) before Adaptive Server initiates a deadlock check for a process

that is waiting on a lock to be released. Deadlock checking is time-consuming
overhead for applications that experience no or very few deadlocks, and the
overhead grows as the percentage of lock requests that must wait for a lock also
increases.
If you set deadlock checking period to a nonzero value (n), Adaptive Server
initiates a deadlock check after a process waits at least n milliseconds. For
example, you can make a process wait at least 700 milliseconds for a lock
before each deadlock check by entering:
sp_configure "deadlock checking period", 700

If you set deadlock checking period to 0, Adaptive Server initiates deadlock

checking when each process begins to wait for a lock. Any value less than the
number of milliseconds in a clock tick is treated as 0. See sql server clock tick
length on page 256.
Configuring deadlock checking period to a higher value produces longer delays
before deadlocks are detected. However, since Adaptive Server grants most
lock requests before this time elapses, the deadlock checking overhead is
avoided for those lock requests. If your applications deadlock infrequently, set
deadlock checking period to a higher value. Otherwise, the default value of 500
should suffice.
Use sp_sysmon to determine the frequency of deadlocks in your system and the
best setting for deadlock checking period. See the Performance and Tuning
Series: Monitoring Adaptive Server with sp_sysmon.

deadlock pipe active

Summary information

106

Default value
Range of values

0
01

Status
Display level

Dynamic
Comprehensive

Required role
Configuration groups

System administrator
Memory Use, Monitoring

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

deadlock pipe active controls whether Adaptive Server collects deadlock

messages. If both deadlock pipe active and deadlock pipe max messages are

enabled, Adaptive Server collects the text for each deadlock. Use
monDeadLock to retrieve these deadlock messages.

deadlock pipe max messages

Summary information
Default value
Range of values

0
0 2147483647

Status
Display level

Dynamic
Comprehensive

Required role

System administrator

Configuration group

Monitoring

deadlock pipe max messages determines the number of deadlock messages

Adaptive Server stores per engine. The total number of messages in the
monSQLText table is the value of sql text pipe max messages times the number

of engines running.

deadlock retries
Summary information
Default value

Range of values
Status

02147483647
Dynamic

Display level
Required role

Intermediate
System administrator

Configuration groups

Lock Manager, SQL Server Administration

deadlock retries specifies the number of times a transaction can attempt to

acquire a lock when deadlocking occurs during an index page split or shrink.

For example, Figure 5-1 illustrates the following scenario:

Transaction A locks page 1007 and needs to acquire a lock on page 1009
to update the page pointers for a page split.

Transaction B is also inserting an index row that causes a page split, holds
a lock on page 1009, and needs to acquire a lock on page 1007.

System Administration Guide: Volume 1

107

Configuration parameters

In this situation, rather than immediately choosing a process as a deadlock

victim, Adaptive Server relinquishes the index locks for one of the
transactions. This often allows the other transaction to complete and release its
locks.
For the transaction that surrendered its locking attempt, the index is rescanned
from the root page, and the page split operation is attempted again, up to the
number of times specified by deadlock retries.
Figure 5-1: Deadlocks during page splitting in a clustered index
Page 1001
Bennet
1007
Karsen
1009
Smith
1062

Transaction A:
Splitting index
page 1007; holds
lock on 1007;
needs to acquire a
lock on 1009 to
update its
previous-page
pointer

Page 1007
Bennet
1132
Greane
1133
Grizley
1127

Page 1007
Bennet
1132
Greane
1133
Hunter
1127
Irons
1218

Page 1009
Karsen
1315
Lemmon 1220
Perkins
1257
Quigley 1254

Page 1033
Hunter
1127
Irons
1218

Page 1009
Karsen
1315
Lemmon 1220
Mouton
1244

Transaction B:
Splitting index
page 1009; holds
lock on 1009;
needs to acquire
a lock on 1007 to
update its
next-page pointer

Page 1044
Perkins
1257
Quigley
1254

sp_sysmon reports on deadlocks and retries. See the Performance and Tuning

Series: Locking and Concurrency Control.

108

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

default character set id

Summary information
Default value
Range of values

1
0255

Status
Display level

Static
Intermediate

Required role
Configuration group

System administrator
Languages

default character set id specifies the number of the default character set used by

the server. The default is set at installation, and can be changed later with the
Sybase installation utilities. See Chapter 10, Configuring Character Sets, Sort
Orders, and Languages.

default database size

Summary information
Default value
Range of values

3MB
2a 10000
a. Minimum determined by servers logical page size.

Status

Dynamic

Display level
Required role

Intermediate
System administrator

Configuration group

SQL Server Administration

default database size sets the default number of megabytes allocated to a new
user database if create database is issued without any size parameters. A
database size given in a create database statement takes precedence over the
value set by this configuration parameter.

If most of the new databases on your Adaptive Server require more than one
logical page size, you may want to increase the default.
Note If you alter the model database, you must also increase the default
database size, because the create database command copies model to create a

new user database.

System Administration Guide: Volume 1

109

Configuration parameters

default exp_row_size percent

Summary information
Default value
Range of values

5
0100

Status
Display level

Dynamic
Intermediate

Required role
Configuration group

System administrator
SQL Server Administration

default exp_row_size percent reserves space for expanding updates in data-

only-locked tables, to reduce row forwarding. An expanding update is any

update to a data row that increases the length of the row. Data rows that allow
null values or that have variable-length columns may be subject to expanding
updates. In data-only-locked tables, expanding updates can require row
forwarding if the data row increases in size so that it no longer fits on the page.
The default value sets aside 5 percent of the available data page size for use by
expanding updates. Since 2002 bytes are available for data storage on pages in
data-only-locked tables, this leaves 100 bytes for expansion. This value is
applied only to pages for tables that have variable-length columns.
Setting default exp_row_size percent to 0 means that all pages are completely
filled and no space is left for expanding updates.
default exp_row_size percent is applied to data-only-locked tables with
variable-length columns when exp_row_size is not explicitly provided with
create table or set with sp_chgattribute. If a value is provided with create table,

that value takes precedence over the configuration parameter setting. See the
Performance and Tuning Series: Locking and Concurrency Control.

default fill factor percent

Summary information

110

Default value
Range of values

0
0100

Status
Display level

Dynamic
Intermediate

Required role
Configuration group

System administrator
SQL Server Administration

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

default fill factor percent determines how full Adaptive Server makes each index

page when it is creating a new index on existing data, unless the fill factor is
specified in the create index statement. The fillfactor percentage is relevant only
when the index is created. As data changes, pages are not maintained at any
particular level of fullness.
default fill factor percent affects:

The amount of storage space used by your data Adaptive Server

redistributes the data as it creates the clustered index.

Performance splitting up pages uses Adaptive Server resources.

There is seldom a reason to change default fill factor percent, especially since
you can override it in the create index command. See create index in the
Reference Manual: Commands.

default language id
Summary information
Default value

Range of values
Status

032767
Dynamic

Display level
Required role

Intermediate
System administrator

Configuration group

Languages

default language id is the number of the language that is used to display system
messages unless a user has chosen another language from those available on
the server. us_english always has an ID of NULL. Additional languages are
assigned unique numbers as they are added.

default network packet size

Summary information
Default value

2048

Range of values
Status

512 65024
Static

Display level
Required role

Intermediate
System administrator

System Administration Guide: Volume 1

111

Configuration parameters

Summary information
Configuration groups

Memory Use, Network Communication, User

Environment

default network packet size configures the default packet size for all Adaptive
Server users. You can set default network packet size to any multiple of 512
bytes; values that are not even multiples of 512 are rounded down.

Memory for all users who log in with the default packet size is allocated from
the Adaptive Server memory pool, as set with total logical memory. This
memory is allocated for network packets when Adaptive Server is started.
Each Adaptive Server user connection uses:

One read buffer

One buffer for messages

One write buffer

Each of these buffers requires default network packet size bytes. The total
amount of memory allocated for network packets is:
(number of user connections + number of worker processes) * 3 * default network
packet size

For example, if you set default network packet size to 1024 bytes, and you have
50 user connections and 20 worker processes, the amount of network memory
required is:
(50 + 20) * 3 * 1024 = 215040 bytes
If you increase default network packet size, you must also increase max network
packet size to at least the same size. If the value of max network packet size is
greater than the value of default network packet size, increase the value of
additional network memory. See additional network memory on page 82.
Use sp_sysmon to see how changing the default network packet size parameter
affects network I/O management and task switching. For example, try
increasing default network packet size and then checking sp_sysmon output to
see how this affects bcp for large batches. See the Performance and Tuning
Series: Monitoring Adaptive Server with sp_sysmon.
Requesting a larger packet size at login

Clients automatically use the Adaptive Server default packet size. Use the -A
flag to Adaptive Server client programs to request a large packet size. For
example:

112

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

isql -A2048

default sortorder id
Summary information
Default value
Range of values

50
0255

Status
Display level

Static
Comprehensive

Required role
Configuration group

System administrator
Languages

default sortorder id is the number of the sort order that is installed as the default
on the server. To change the default sort order, see Chapter 10, Configuring
Character Sets, Sort Orders, and Languages.

default unicode sortorder

Summary information
Default value
Range of values

binary
Not currently used

Status
Display level

Static
Comprehensive

Required role
Configuration group

System administrator
Unicode

default unicode sortorder is a string parameter that uniquely defines the default

Unicode sort order installed on the server. To change the Unicode default sort
order, see Chapter 10, Configuring Character Sets, Sort Orders, and
Languages.

default XML sortorder

Summary information
Default value

binary

Range of values
Status

(not currently used)

Static

Display level

Comprehensive

System Administration Guide: Volume 1

113

Configuration parameters

Summary information
Required role
Configuration group

System administrator
Unicode

default XML sortorder is a string parameter that defines the sort order used by
the XML engine. A string parameter is used rather than a numeric parameter to
guarantee a unique ID. See Chapter 6, XML Support for I18N in XML
Services in Adaptive Server Enterprise.

deferred name resolution

Summary information
Default value
Range of values

0 (disabled)
0 to 1

Status
Required role

dynamic
System administrator

Configuration group

Query tuning

When deferred name resolution is active (1), deferred name resolution is

applied globally to all server connections; all procedures you create in the
server are created using deferred name resolution.
Therefore, the stored procedures are created without resolving the objects
referenced inside the stored procedure, postponing object resolution
processing to the execution time. See Chapter 17, Using Stored Procedures,
in the Transact-SQL Users Guide

disable character set conversions

Summary information

114

Default value
Valid values

0 (enabled)
0 (enabled), 1 (disabled)

Status
Display level

Static
Comprehensive

Required role
Configuration group

System administrator
Languages

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

Changing disable character set conversions to 1 turns off character set

conversion for data moving between clients and Adaptive Server. By default,
Adaptive Server performs conversion on data moving to and from clients that
use character sets that are different than the servers. For example, if some
clients use Latin-1 (iso_1) and Adaptive Server uses Roman-8 (roman8) as its
default character set, data from the clients is converted to Roman-8 when being
loaded into Adaptive Server. For clients using Latin-1, the data is reconverted
when it is sent to the client; for clients using the same character set as Adaptive
Server, the data is not converted.
By setting disable character set conversions, you can request that no conversion
take place. For example, if all clients are using a given character set, and you
want Adaptive Server to store all data in that character set, set disable character
set conversions to 1, and no conversion takes place.

disable disk mirroring

Summary information
Default value

Valid values
Status

0 (off), 1 (on)
Static

Display level
Required role

Comprehensive
System administrator

Configuration group

Disk I/O

disable disk mirroring enables or disables disk mirroring for Adaptive Server.

This is a global variable; Adaptive Server does not perform any disk mirroring
after this configuration parameter is set to 1 and Adaptive Server is restarted.
Setting disable disk mirroring to 0 enables disk mirroring.
Note You must disable disk mirroring if your Adaptive Server is configured

for failover.

disable varbinary truncation

Summary information
Default value

0 (on)

Range of values
Status

0 (on), 1 (off)
Dynamic

System Administration Guide: Volume 1

115

Configuration parameters

Summary information
Display level
Required role

Comprehensive
System Administrator

Configuration group

SQL Server Administration

disable varbinary truncation controls whether Adaptive Server includes trailing

zeros at the end of varbinary or binary null data.

disk i/o structures

Summary information
Default value

256

Range of values
Status

02147483647
Dynamic

Display level
Required role

Comprehensive
System administrator

Configuration group

Disk I/O, Memory Use

disk i/o structures specifies the initial number of disk I/O control blocks
Adaptive Server allocates at start-up.

User processes require a disk I/O control block before Adaptive Server can
initiate an I/O request for the process. The memory for disk I/O control blocks
is preallocated when Adaptive Server starts. To minimize the chance of
running out of disk I/O structures, you should configure disk i/o structures to as
high a value as your operating system allows. See your operating system
documentation for information on concurrent disk I/Os.
Use sp_sysmon to determine whether to allocate more disk I/O structures. See
the Performance and Tuning Series: Monitoring Adaptive Server with
sp_sysmon. You can set the max async i/os per server configuration parameter
to the same value as disk i/o structures. See max async i/os per server on page
167.

DMA object pool size

116

Summary information
Default value

4096

Valid values
Status

2048 2147483647
Static

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

Summary information
Display level
Required role

Comprehensive
System administrator

Configuration group

Shared disk cluster

DMA object pool size specifies the number of DMA (direct memory access)

objects allocated by CIPC at start-up time.

dtm detach timeout period

Summary information
Default value

0 (minutes)

Valid values
Status

0 2147483647 (minutes)
Dynamic

Display level
Required role

10
System administrator

Configuration group

DTM Administration

dtm detach timeout period sets the amount of time, in minutes, that a distributed

transaction branch can remain in the detached state. In some X/Open XA

environments, a transaction may become detached from its thread of control
(usually to become attached to a different thread of control). Adaptive Server
permits transactions to remain in a detached state for the length of time
specified by dtm detach timeout period. After this time has passed, Adaptive
Server rolls back the detached transaction.

dtm lock timeout period

Summary information
Default value

300 (seconds)

Valid values
Status

1 2147483647 (seconds)
Dynamic

Display level
Required role

10
System administrator

Configuration group

DTM Administration

System Administration Guide: Volume 1

117

Configuration parameters

dtm lock timeout period sets the maximum amount of time, in seconds, that a

distributed transaction branch waits for lock resources to become available.

After this time has passed, Adaptive Server considers the transaction to be in a
deadlock situation, and rolls back the transaction branch that triggered the
deadlock. This ultimately rolls back the entire distributed transaction.
Distributed transactions may potentially deadlock themselves if they propagate
a transaction to a remote server, and in turn, the remote server propagates a
transaction back to the originating server. This situation is shown in Figure 52. The work of distributed transaction dxact1 is propagated to Adaptive
Server 2 via rpc1. Adaptive Server 2 then propagates the transaction back to
the coordinating server via rpc2. rpc2 and dxact1 share the same gtrid but
have different branch qualifiers, so they cannot share the same transaction
resources. If rpc2 is awaiting a lock held by dxact1, a deadlock situation
exists.
Figure 5-2: Distributed transaction deadlock

dxact1 ASE1
ASTC
rpc1

rpc2

ASE2
ASTC
Adaptive Server cannot detect interserver deadlocks. Instead, it relies on dtm
lock timeout period. In Figure 5-2, after dtm lock timeout period has expired, the
transaction created for rpc2 is aborted. This causes Adaptive Server 2 to
report a failure in its work, and dxact1 is ultimately aborted as well.
The value of dtm lock timeout period applies only to distributed transactions.
Local transactions may use a lock timeout period with the server-wide lock wait
period parameter.
Note Adaptive Server does not use dtm lock timeout period to detect deadlocks

on system tables.

118

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

dump on conditions
Summary information
Default value
Range of values

0 (off)
0 (off), 1 (on)

Status
Display level

Dynamic
Intermediate

Required role
Configuration group

System administrator
Group Diagnostics

dump on conditions determines whether Adaptive Server generates a dump of

data in shared memory when it encounters the conditions specified in maximum
dump conditions.

Note The dump on conditions parameter is included for use only by Sybase

Technical Support. Do not modify it unless you are instructed to do so by

Sybase Technical Support.

dynamic allocation on demand

Summary information
Default value
Range of values

1 (on)
0 (off), 1 (on)

Status
Display level

Dynamic
Basic

Required role
Configuration groups

System administrator
Memory Use, Physical Memory

dynamic allocation on demand determines when memory is allocated for

changes to dynamic memory configuration parameters.

If you set dynamic allocation on demand to 1, memory is allocated only as it is

needed. That is, if you change the configuration for number of user connections
from 100 to 200, the memory for each user is added only when the user
connects to the server. Adaptive Server continues to add memory until it
reaches the new maximum for user connections.

System Administration Guide: Volume 1

119

Configuration parameters

If dynamic allocation on demand is set to 0, all the memory required for any
dynamic configuration changes is allocated immediately. That is, when you
change the number of user connections from 100 to 200, the memory required
for the extra 100 user connections is immediately allocated.

enable backupserver HA
Summary information
Default value
Valid values

1
1 (enabled), 0 (disabled)

Status
Display level

Dynamic
Comprehensive

Required role

System administrator

Configuration group

Shared disk cluster

Setting enable backupserver HA to 1 starts the high availability Backup Server

for the cluster. Setting enbale backupserver HA to 0 disables the high
availability Backup Server on the cluster.

enable cis
Summary information
Default value

1 (on)

Valid values
Status

0 (off), 1 (on)
Static

Display level
Required role

Comprehensive
System administrator

Configuration group

Component Integration Services

enable cis enables or disables Component Integration Service.

enable compression

120

Summary information
Default value

0 (off)

Valid values
Status

0 (off), 1 (on)
Static

Display level

Comprehensive

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

Summary information
Required role
Configuration group

System administrator
SQL Server Administration

enable cis enables or disables data compression. See the Compression Users

Guide.

enable console logging

Summary information
Default value

0 (off)

Valid values

0 (off), 1(on)

Status
Display level

Dynamic
10

Required role
Configuration groups

System administrator
SQL Server Administration

enable console logging determines whether Adaptive Server sends messages to

the console. By default, enable console logging is disabled, and after startup,

Adaptive Server sends no messages to the console (the messages are still sent
to the errorlog). Once enabled, enable console logging allows Adaptive Server
to send messages to the console and error log at all times.

enable DTM
Summary information
Default value
Valid values

0 (off)
0 (off), 1(on)

Status
Display level

Static
10

Required role
Configuration groups

System administrator
DTM Administration, SQL Server Administration

System Administration Guide: Volume 1

121

Configuration parameters

enable DTM enables or disables the Adaptive Server distributed transaction

management (DTM) feature. When DTM is enabled, you can use Adaptive
Server as a resource manager in X/Open XA and MSDTC systems. You must
restart the server for this parameter to take effect. See the XA Interface
Integration Guide for CICS, Encina, and TUXEDO for more information about
using Adaptive Server in an X/Open XA environment. See Using Adaptive
Server Distributed Transaction Management Features for information about
transactions in MSDTC environments, and for information about Adaptive
Server native transaction coordination services.

enable encrypted columns

Summary information
Default value

0 (off)

Range of values
Status

0 (off), 1(on)
Dynamic

Display level
Required role

Comprehensive
System administrator

Configuration group

Security Related

enable encrypted columns enables encrypted columns.

You cannot set enable encrypted columns unless you have purchased, installed,
and registered the ASE_ENCRYPTION license on your server. Any attempt to
set it without such licensing results in Msg. 10834:
Configuration parameter 'enable encrypted columns'
cannot be enabled without license 'ASE_ENCRYPTION'
Note Using encrypted columns increases the logical memory used by 8198

kilobytes.

enable enterprise java beans

122

Summary information
Default value

0 (off)

Range of values
Status

0 (off), 1(on)
Static

Display level

Comprehensive

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

Summary information
Required role
Configuration group

System administrator
Java Services

enable enterprise java beans enables and disables EJB Server in the Adaptive
Server database. You cannot use EJB Server until the Adaptive Server is
enabled for EJB Server.

enable file access

Summary information
Default value

1 (on)

Valid values

0 (off), 1 (on)

Status
Display level

Static
Comprehensive

Required role
Configuration group

System administrator
Component Integration Services

enable file access enables access through proxy tables to the external file
system. Requires a license for ASE_XFS.

enable full-text search

Summary information
Default value
Valid values

1
0 (off), 1 (on)

Status
Display level

Static
Comprehensive

Required role
Configuration group

System administrator
Component Integration Services

enable full-text search enables Enhanced Full-Text Search services. Requires a

license for ASE_EFTS.

enable functionality group

Summary information
Default value

System Administration Guide: Volume 1

123

Configuration parameters

Summary information
Range of values
Status

0 (off), 1 (on)
Dynamic

Display level
Required role

Intermediate
System administrator

Configuration group

SQL Server Administration

enable functionality group enables or disables the changes available for these

features in Adaptive Server versions 15.7 and later:

Shareable inline defaults when enable functionality group is set to 0,

Adaptive Server does not create shareable inline defaults, and does not
reuse existing shareable inline defaults. The columns that share inline
defaults (before you change this configuration parameter) continue to
share the defaults until the defaults are removed.

select for update acquiring exclusive locks

Quoted identifiers

Unicode noncharacters

Monitoring cursor statements

Reducing query processing latency

Suppressing Job Scheduler max task messages

The default value for the parameters in this group depends on the value to
which enable functionality group is set. A value of DEFAULT for the individual
configuration parameters in this groupother than enable functionality group
means they are set to the same value as enable functionality group. That is, if you
set enable functionality group to 1, a value of DEFAULT for any other
configuration parameter in the group is 1.
Aside from the value for enable functionality group, you can ignore values of
DEFAULT in the output from sp_configure and sp_helpconfig for individual
configuration parameters in the Application Functionality group.

enable inline default sharing

124

Summary information
Default value

0 (off)

Valid values
Status

0 (off), 1(on)
Dynamic

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

Summary information
Display level
Required role

Comprehensive
System administrator

Configuration group

Application Functionality

enable inline default sharing enables Adaptive Server to share inline defaults.

That is, once this configuration is enabled, Adaptive Server looks for existing
shareable inline defaults having the same value in the database belonging to the
same user. If it finds an existing shareable default, Adaptive Server binds this
object to the column instead of creating a new default. However, if Adaptive
Server does not find an existing shareable inline default, it creates a new
default
See Defining Defaults and Rules for Data in the Transact-SQL Users Guide.

enable HA
Summary information
Default value

0 (off)

Range of values
Status

02
Static

Display level
Required role

Comprehensive
System administrator

Configuration group

SQL Server Administration

Set enable HA to 1 to configure Adaptive Server as a companion server in an

active-active high availability subsystem. Set enable HA to 2 to configure
Adaptive Server as a companion server in an active-passive high availability
subsystem.

System Administration Guide: Volume 1

125

Configuration parameters

Adaptive Server uses Sybase Failover to interact with the high availability
subsystem. You must set enable HA to 1 before you run the installhasvss script
(insthasv on Windows), which installs the system procedures for Sybase
Failover.
Note The license information and the run value for enable HA are independent
of each other. Whether or not you have a license for Sybase Failover, the run
value and the config value are set to 1 when you restart Adaptive Server. Until
you have a license, you cannot run Sybase Failover. If you have not installed a
valid license, Adaptive Server logs an error message and does not activate the
feature. See the installation guide for your platform for information about
installing license keys.

Setting enable HA to 1 or 2 does not mean that Adaptive Server is configured

to work in a high availability system. You must perform the steps described in
Using Sybase Failover in a High Availability System to configure Adaptive
Server to be a companion server in a high availability system.
When enable HA is set to 0, you cannot configure for Sybase Failover, and you
cannot run installhasvss (insthasv on Windows).

enable housekeeper GC
Summary information
Default value
Range of values

1 (on)
05

Status
Display level

Dynamic
Intermediate

Required role
Configuration group

System administrator
SQL Server Administration

The housekeeper garbage collection task performs space reclamation on dataonly-locked tables. When a user task deletes a row from a data-only-locked
table, a task is queued to the housekeeper to check the data and index pages for
committed deletes.
The housekeeper garbage collection task is controlled by enable housekeeper
GC. See Chapter 3, Using Engines and CPUs in the Performance and Tuning
Series: Basics.
These are valid values for enable housekeeper GC:

126

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

0 disables the housekeeper garbage collection task, but enables the delete
commands lazy garbage collection. You must use reorg reclaim_space to
deallocate empty pages. This is the cheapest option with the lowest
performance impact, but it may cause performance problems if many
empty pages accumulate. Sybase recommends that you do not use this
value.

1 enables lazy garbage collection for the housekeeper garbage collection

task and the delete command. If more empty pages accumulate than your
application allows, consider options 4 or 5. You can use the optdiag utility
to obtain statistics of empty pages.

2 reserved for future.

3 reserved for future.

4 enables aggressive garbage collection for the housekeeper garbage

collection task and the delete command. This option is the most effective,
but the delete command is expensive. This option is ideal if the deletes on
your DOL tables are in a batch.

5 enables aggressive garbage collection for the housekeeper, and lazy

garbage collection for the delete command. This option is less expensive
for deletes than option 4. This option is suitable when deletes are caused
by concurrent transactions

sp_sysmon reports on how often the housekeeper garbage collection task

performed space reclamation and how many pages were reclaimed. See the
Performance and Tuning Series: Monitoring Adaptive Server with sp_sysmon.

enable hp posix async i/o

Summary information
Default value

Valid values
Status

1 (enabled), 0 (disabled)
Static

Display level
Required role

Comprehensive
System administrator

Configuration group

Disk I/O

Setting enable hp posix async i/o to 1 enables asynchronous I/O on database

devices created on on HP-UX 11.31 and later file systems. You must first
enable Adaptive Server to use asynchronous I/O with the allow sql server async
i/o configuration parameter.

System Administration Guide: Volume 1

127

Configuration parameters

The combination of these configuration parameters determines whether

Adaptive Server uses asynchronous or synchronous I/O for raw partitions and
file systems:
enable hp posix async i/o

allow sql server async i/o = 0

allow sql server async i/o = 1

Set to 0

Synchronous I/O for file systems

Synchronous I/O for file systems

Synchronous I/O for raw devices

Uses /dev/async asynchronous I/O for

raw devices

Synchronous I/O for file systems

POSIX asynchronous I/O for file systems

Synchronous I/O for raw devices

POSIX asynchronous I/O for raw devices

Set to 1

Note enable hp posix async i/o improves performance when you allocate
database devices on file systems, but may decrease performance on database
devices that are allocated on raw devices.

enable i/o fencing

Summary information
Default value
Valid values

0
1 (enabled), 0 (disabled)

Status
Display level

Static
Comprehensive

Required role
Configuration group

System administrator
Shared Disk Cluster

Setting enable i/o fencing to 1 enables I/O fencing for each database device that
supports the SCSI-3 Persistent Group Reservation (PGR) standard.

enable java
Summary information

128

Default value
Range of values

0 (disabled)
0 (disabled), 1 (enabled)

Status
Display level

Static
Comprehensive

Required role
Configuration group

System administrator
Java Services

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

enable java enables and disables Java in the Adaptive Server database. You
cannot install Java classes or perform any Java operations until the server is
enabled for Java.

enable job scheduler

Summary information
Default value
Range of values

0 (off)
0 (off), 1 (on)

Status
Display level

Dynamic
Comprehensive

Required role

System administrator

Configuration group

SQL Server Administration

enable job scheduler determines whether Job Scheduler starts when Adaptive

Server starts.

enable ldap user auth

Summary information
Default value

0 (off)

Valid values

0 (off) allows only syslogins authentication.

1 (on) allows both LDAP and syslogins
authentication.

Status

2 (on) allows only LDAP authentication.

Dynamic

Display level
Required role

Comprehensive
System security officer

Configuration group

Security Related

When enable ldap user auth is 1, Adaptive Server searches the LDAP server to
authenticate each user. If the LDAP authentication fails, Adaptive Server
searches syslogins to authenticate the user. Use level 1 when you are migrating
users from Adaptive Server authentication to LDAP authentication.

System Administration Guide: Volume 1

129

Configuration parameters

enable literal autoparam

Summary information
Default value
Range of values

0
1 (enabled), 0 (disabled)

Status
Display level

Dynamic
Intermediate

Required role
Configuration group

System administrator
Query Tuning

enable literal autoparam enables and disables literal server-wide

parameterization.

enable logins during recovery

Summary information
Default value
Range of values

1
0 (enabled), 1 (disabled)

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
Security Related

enable logins during recovery determines whether non-system administrator

logins are allowed during database recovery. A value of 1 indicates that logins
are allowed during recovery, and a value of 0 indicates that logins are not
allowed during recovery, that is, only the system administrator can log in to
Adaptive Server.

enable merge join

Summary information
Default value

Range of values

0 disables merge joins at the server level.

1 enables merge joins at the server level.

130

Status

2 sets merge joins to their default values at the

server level.
Dynamic

Display level

Comprehensive

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

Summary information
Required role
Configuration group

System administrator
Query Tuning

enable merge join enables or disables merge join at the server level.

The default value for merge join depends on current value of the optimization
goal configuration parameter:
Value for optimization goal
allrows_mix

Default value for merge join

on
on
off

allrows_dss
allrows_oltp

enable metrics capture

Summary information
Default value
Range of values

0 (off)
0 (off), 1 (on)

Status
Display level

Dynamic
Intermediate

Required role
Configuration group

System administrator
SQL Server Administration

enable metrics capture enables Adaptive Server to capture metrics at the server

level. Metrics for ad hoc statements are captured in the system catalogs;
metrics for statements in a stored procedure are saved in the procedure cache.

enable monitoring
Summary information
Default value

0 (off)

Range of values
Status

0 (off), 1(on)
Dynamic

Display level
Required role

Comprehensive
System administrator

Configuration group

Monitoring

System Administration Guide: Volume 1

131

Configuration parameters

enable monitoring controls whether Adaptive Server collects the monitoring

table data. enable monitoring acts as a master switch that determines whether
any Monitoring configuration parameters are enabled.

Note Any data that populates the monitoring tables without enabling enable
monitoring should not be considered valid.

enable pam user auth

Summary information
Default value
Range of values

0 (off)
0 (off) allows only syslogins authentication.
1 (on) allows both PAM and syslogins authentication.

Status

2 (on) allows only PAM authentication.

Dynamic

Display level
Required role

Intermediate
System security officer

Configuration group

Security Related

enable pam user auth controls the ability to authenticate users using pluggable
authentication modules (PAM).

When enable pam user auth is set to 1, Adaptive Server uses the PAM provider
to authenticate each user. If the PAM authentication fails, Adaptive Server
searches syslogins to authenticate the user. Use level 1 when you are migrating
users from Adaptive Server authentication to PAM authentication.

enable pci

132

Summary information
Default value

0 (off)

Valid values
Status

0 (off), 1 (on), 2 (on with operating system override)

Dynamic

Display level
Required role

Intermediate
System Administrator

Configuration group

User Environment

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

enable pci enables or disables the Java PCI Bridge for Adaptive Server.

Note Do not use setting 2 (on with operating system override) unless

instructed to do so by Sybase Technical Support. This setting enables the PCI

Bridge on operating system versions that may not fully or correctly support
PCI functionality.

enable permissive unicode

Summary information
Default value
Valid values

0 (off)
0 (off), 1(on)

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
Application Functionality

enable inline default sharing allows Adaptive Server to ignore Unicode

noncharacters.
SeeAllowing Unicode noncharacters on page 372.

enable query tuning mem limit

Summary information
Default value

1 (on)

Range of values
Status

0 (off), 1 (on)
Dynamic

Display level
Required role

Intermediate
System administrator

Configuration group

Query Tuning

enable query tuning mem limit enables the query tuning memory limit.

enable query tuning time limit

Summary information
Default value

System Administration Guide: Volume 1

1 (on)

133

Configuration parameters

Summary information
Range of values
Status

0 (off), 1 (on)
Intermediate

Display level
Required role

Intermediate
System administrator

Configuration group

Query Tuning

enable query tuning time limit enables the query tuning time limit.

enable real time messaging

Summary information
Default value

1 (on)

Range of values
Status

0 (off), 1 (on)
Dynamic

Display level
Required role

Comprehensive
System administrator

Configuration group

SQL Server Administration

enable real time messaging enables the real time messaging services.

enable rep agent threads

Summary information
Default value

1 (on)

Range of values
Status

0 (off), 1 (on)
Dynamic

Display level
Required role

Basic
System administrator

Configuration groups

Memory Use, Rep Agent Thread Administration

enable rep agent threads enables the RepAgent thread within Adaptive Server.

Other steps are also required to enable replication. For more information, see
the Replication Server documentation.

134

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

enable row level access control

Summary information
Default value
Valid values

0 (off)
0 (off), 1 (on)

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System security officer

Security Related

enable row level access control enables row level access control. You must have
the security services license key enabled before you can configure enable row
level access control.

enable semantic partitioning

Summary information
Default value
Range of values

0
1 (enabled), 0 (disabled)

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
SQL Server Administration

enable semantic partitioning enables partitioning other than round-robin (for

example list, hash, and range partitioning) in Adaptive Server. Before you use
any of these partitioning schemes, you must first have the appropriate license.

enable sort-merge join and JTC

Summary information
Default value

0 (off)

Valid values
Status

0 (off), 1 (on)
Dynamic

Display level
Required role

Comprehensive
System administrator

Configuration group

Query Tuning

System Administration Guide: Volume 1

135

Configuration parameters

Used when Adaptive Server is in compatibility mode. Once enabled, when

Adaptive Server compiles a query in compatibility mode, enable sort-merge
join and JTC allows the query processor to select a sort merge or a nested loop
join. enable sort-merge join and JTC enables join transitive closure (JTC),
which allows the query processor for releases earlier than 15.0 to use additional
join clauses.
For more information about compatibility mode, see the Migration Technology
Guide.

enable sql debugger

Summary information
Default value

1 (on)

Valid values
Status

0 (off), 1 (on)
Static

Display level
Required role

Comprehensive
System administrator

Configuration group

SQL Server Administration

Enables and disables the Adaptive Server SQL debugger, which allows you to
step through your T-SQL code.

enable ssl
Summary information
Default value

0 (off)

Valid values
Status

0 (off), 1 (on)
Static

Display level
Required role

Comprehensive
System security officer

Configuration group

Security Related

enable ssl enables or disables Secure Sockets Layer session-based security.

enable stmt cache monitoring

Summary information
Default value

136

0 (off)

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

Summary information
Valid values
Status

0 (off), 1 (on)
Dynamic

Display level
Required role

Comprehensive
System administration

Configuration group

Monitoring

enable stmt cache monitoring enables or disables Adaptive Server to collect

monitoring information about the statement cache. Once enabled,

monStatementCache and monCachedStatement display valid data.

enable surrogate processing

Summary information
Default value

1 (on)

Range of values
Status

0 (off), 1 (on)
Dynamic

Display level
Required role

Comprehensive
System administrator

Configuration group

Unicode

Activates the processing and maintains the integrity of surrogate pairs in

Unicode data. If enable surrogate processing is disabled, the server ignores the
presence of surrogate pairs in the Unicode data, and all code that maintains the
integrity of surrogate pairs is skipped. This enhances performance, but restricts
the range of Unicode characters that can appear in the data.

enable unicode conversion

Summary information
Default value

Range of values

0 uses only the built-in character-set conversion.

1 uses the built-in conversion. If it cannot find a
built-in conversion, Adaptive Server uses the Unilib
character conversion
2 uses the appropriate Unilib conversion

Status
Display level

System Administration Guide: Volume 1

Dynamic
Comprehensive

137

Configuration parameters

Summary information
Required role
Configuration groups

System administrator
Languages, Unicode

enable unicode conversion activates character conversion using Unilib for the
char, varchar, and text datatypes.

enable unicode normalization

Summary information
Default value

1 (on)

Range of values

0 (off), 1 (on)

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
Unicode

Activates Unilib character normalization. The normalization process modifies

the data so there is only a single representation in the database for a given
sequence of abstract characters. Often, characters followed by combined
diacritics are replaced by precombined forms.
Set enable unicode normalization to 1 to use the built-in process that enforces
normalization on all incoming Unicode data. If this parameter is disabled (set
to 0), the normalization step is bypassed and the client code is responsible for
normalization rather than the server. If normalization is disabled, performance
is improvedbut only if all clients present Unicode data to the server using the
same representation.
Note Once disabled, you cannot turn normalization on again. This limitation

prevents non-normalized data from entering the database.

enable webservices
Summary information

138

Default value
Range of values

0
1 (enabled), 0 (disabled)

Status

Dynamic

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

Summary information
Display level
Required role

Intermediate
System administrator

Configuration group

SQL Server Administration

Enables Webservices.

enable xact coordination

Summary information
Default value
Valid values

1 (on)
0 (off), 1(on)

Status

Static

Display level
Required role

10
System administrator

Configuration group

DTM Administration

enable xact coordination enables or disables Adaptive Server transaction

coordination services. When this parameter is set to 1 (on), coordination
services are enabled, and the server can propagate transactions to other
Adaptive Servers. This may occur when a transaction executes a remote
procedure call (RPC) to update data in another server, or updates data in
another server using Component Integration Services (CIS). Transaction
coordination services ensure that updates to remote Adaptive Server data
commit or roll back with the original transaction.

If this parameter is set to 0 (off), Adaptive Server does not coordinate the work
of remote servers. Transactions can still execute RPCs and update data using
CIS, but Adaptive Server cannot ensure that remote transactions are rolled
back with the original transaction or that remote work is committed along with
an original transaction, if remote servers experience a system failure. This
corresponds to the behavior of Adaptive Server versions earlier than version
12.x.

enable xml
Summary information
Default value

Range of values
Status

1 (enabled), 0 (disabled)
Dynamic

System Administration Guide: Volume 1

139

Configuration parameters

Summary information
Display level
Required role

Intermediate
System administrator

Configuration group

SQL Server Administration

Enables XML services.

engine memory log size

Summary information
Default value
Range of values

0
0 2147483647

Status

Dynamic

Display level
Required role
Configuration group

Physical Memory

engine memory log size is for diagnostic use only and has no relevance in a
production environment. It should be left at the default setting unless otherwise
requested by Sybase Tech Support.

errorlog pipe active

Summary information
Default value

Range of values
Status

01
Dynamic

Display level
Required role

Comprehensive
System administrator

Configuration group

Monitoring

errorlog pipe active controls whether Adaptive Server collects error log
messages. If both errorlog pipe active and errorlog pipe max messages are
enabled, Adaptive Server collects all the messages sent to the error log. Use
monErrorLog to retrieve these error log messages.

140

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

errorlog pipe max messages

Summary information
Default value
Range of values

0
02147483647

Status
Display level

Dynamic
Comprehensive

Required role
Configuration groups

System administrator
Memory Use, Monitoring

errorlog pipe max messages determines the number of error log messages
Adaptive Server stores per engine. The total number of messages in the
monSQLText table is the value of sql text pipe max messages times the number
of engines running.

esp execution priority

Summary information
Default value
Range of values

8
015

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
Extended Stored Procedure

esp execution priority sets the priority of the XP Server thread for ESP

execution. Over long periods of time ESPs can be CPU-intensive. Also, since
XP Server resides on the same machine as Adaptive Server, XP Server can
impact Adaptive Server performance.
See the Open Server Server-Library/C Reference Manual for information
about scheduling Open Server threads.

System Administration Guide: Volume 1

141

Configuration parameters

esp execution stacksize

Summary information
Default value

Platform dependent:
196608 Linux AMD-64, IBM PLinux, HP IA64,
Sun x86 64
139264 SunSparc64
65536 Sun Sparc32, Windows 32, IBM AIX 64,
Linux IA 32, Windows 64

Range of values

67584 HP 64
Platform default2147483647

Status
Display level

Static
Comprehensive

Required role
Configuration group

System administrator
Extended Stored Procedure

esp execution stacksize sets the size of the stack, in bytes, to be allocated for

ESP execution.
Use this parameter if you have your own ESP functions that require a larger
stack size than the default, 34816.

esp unload dll

Summary information
Default value
Range of values

0 (off)
0 (off), 1 (on)

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
Extended Stored Procedure

esp unload dll specifies whether DLLs that support ESPs should be

automatically unloaded from XP Server memory after the ESP call has
completed.
If esp unload dll is set to 0, DLLs are not automatically unloaded. If it is set to
1, they are automatically unloaded.
If esp unload dll is set to 0, you can still unload individual DLLs explicitly at
runtime, using sp_freedll.

142

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

event buffers per engine

Summary information
Default value
Range of values

100
1 2147483647

Status
Display level

Static
Comprehensive

Required role
Configuration groups

System administrator
Memory Use, SQL Server Administration

event buffers per engine specifies the number of event buffers available for each

engine.
Events are used by Adaptive Server Monitor for observing Adaptive Server
performance; if you are not using Adaptive Server Monitor, set this parameter
to 1.
The value to which you set event buffers per engine depends on the number of
engines in your configuration, the level of activity on your Adaptive Server,
and the types of applications you are running.
Setting event buffers per engine to a low value may result in the loss of event
information. The default value is likely to be too low for most sites. Values of
2000 and greater may be more reasonable for general monitoring. However,
experiment to determine the appropriate value for your site.
In general, setting event buffers per engine to a high value may reduce the
amount of performance degradation that Adaptive Server Monitor causes
Adaptive Server.
Each event buffer uses 100 bytes of memory. To determine the total amount of
memory used by a particular value for event buffers per engine, multiply the
value by the number of Adaptive Server engines in your configuration.

event log computer name (Windows only)

Summary information
Default value
Valid values

LocalSystem
Name of an Windows machine on the network
configured to record Adaptive Server messages
LocalSystem
NULL

Status

System Administration Guide: Volume 1

Dynamic

143

Configuration parameters

Summary information
Display level
Required role

Comprehensive
System administrator

Configuration group

Error Log

event log computer name specifies the name of the Windows PC that logs

Adaptive Server messages in its Windows Event Log. This feature is available
on Windows servers only.
A value of LocalSystem or NULL specifies the default local system.
You can also use the Server Config utility to set the event log computer name
parameter by specifying the Event Log Computer Name under Event Logging.
Setting the event log computer name parameter with sp_configure or specifying
the Event Log Computer Name under Event Logging overwrites the effects of
the command line -G option, if it was specified. If Adaptive Server was started
with the -G option, you can change the destination remote machine by setting
event log computer name.
For more information about logging Adaptive Server messages to a remote
site, see the Configuration Guide for Windows.

event logging (Windows only)

Summary information
Default value

Valid values
Status

0 (off), 1 (on)
Dynamic

Display level
Required role

Comprehensive
System administrator

Configuration group

Error Log

event logging enables and disables the logging of Adaptive Server messages in
the Windows Event Log.

The default value of 1 enables Adaptive Server message logging in the

Windows Event Log; a value of 0 disables it.
Use the Server Config utility to set the event logging parameter by selecting Use
Windows Event Logging under Event Logging.
Setting the event logging parameter or selecting Use Windows Event Logging
overwrites the effects of the command line -G option, if it was specified.

144

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

executable codesize + overhead

Summary information
Default value
Range of values

0
0 2147483647

Status
Display level

Calculated
Basic

Required role
Configuration group

System administrator
Memory Use

executable codesize + overhead reports the combined size, in kilobytes, of the

Adaptive Server executable and overhead. This a calculated value that is not
user-configurable.

extended cache size

Summary information
Default value
Range of values

0
0 31457280

Status
Display level

Static
Intermediate

Required role
Configuration group

System administrator
Cache Manager

extended cache size specifies the size of the secondary cache.

FIPS login password encryption

Summary information
Default value

Range of values
Status

01
Static

Display level
Required role

Comprehensive
System security officer

Configuration group

Security related

Enabling FIPS login password encryption requires a Security and Directory

Services license. This parameter provides FIPS 140-2 cryptographic module
support for encrypting passwords in transmission, in memory, and on disk.

System Administration Guide: Volume 1

145

Configuration parameters

Adaptive Server uses the FIPS 140-2 certified Certicom security provider for
login encryption. If this configuration is not enabled, Adaptive Server uses the
OpenSSL security provider to perform login password encryption.

global async prefetch limit

Summary information
Default value
Range of values

10
0100

Status
Display level

Dynamic
Intermediate

Required role

System administrator

Configuration group

Cache Manager

global async prefetch limit specifies the percentage of a buffer pool that can hold

the pages brought in by asynchronous prefetch that have not yet been read. This
parameter sets the limit for all pools in all caches for which the limit has not
been set explicitly with sp_poolconfig.
If the limit for a pool is exceeded, asynchronous prefetch is temporarily
disabled until the percentage of unread pages falls below the limit. See Chapter
6, Tuning Asynchronous Prefetch in the Performance and Tuning Series:
Basics.

global cache partition number

Summary information
Default value

Range of values
Status

1 64, as powers of 2
Static

Display level
Required role

Intermediate
System administrator

Configuration group

Cache Manager

global cache partition number sets the default number of cache partitions for all

data caches. The number of partitions for a particular cache can be set using
sp_cacheconfig; the local value takes precedence over the global value.

146

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

Use cache partitioning to reduce cache spinlock contention; in general, if

spinlock contention exceeds 10 percent, partitioning the cache should improve
performance. Doubling the number of partitions cuts spinlock contention by
about one-half.
See Chapter 4, Configuring Data Caches, in the System Administration
Guide: Volume 2 for information on configuring cache partitions. Also see
Chapter 6, Tuning Asynchronous Prefetch in the Performance and Tuning
Series: Basics.

heap memory per user

Summary information
Default value

4K

Valid values
Status

0 2147483647 bytes
Dynamic

Display level
Required role

Comprehensive
System administrator

Configuration groups

Memory Use, Physical Memory

heap memory per user configures the amount of heap memory per user. A heap
memory pool is an internal memory created at start-up that tasks use to
dynamically allocate memory as needed. This memory pool is important if you
are running tasks that use wide columns, which require a lot of memory from
the stack. The heap memory allocates a temporary buffer that enables these
wide column tasks to finish. The heap memory the task uses is returned to the
heap memory pool when the task is finished.

The size of the memory pool depends on the number of user connections.
Sybase recommends that you set heap memory per user to three times the size
of your logical page.

histogram tuning factor

Summary information
Default value

20

Range of values
Status

1 100
Dynamic

Display level
Required role

Intermediate
System administrator

System Administration Guide: Volume 1

147

Configuration parameters

Summary information
Configuration group

SQL Server Administration

histogram tuning factor controls the number of steps Adaptive Server analyzes
per histogram for update statistics, update index statistics, update all statistics,
and create index. A value of 1 disables the parameter.

Note For Adaptive Server versions 15.0.2 ESD #2 and later, if you set
histogram tuning factor to the default value of 20 and a large number of steps are

requested for the histogram, the actual step count used for the histogram is
limited to the value that reduces the procedure cache usage:
min (max (400, requested_steps),
histogram_tuning_factor X requested_steps)

In the following example, Adaptive Server generates an intermediate 20-step

histogram with 30 values:
sp_configure 'histogram tuning factor',20
update statistics tab using 30 values

Adaptive Server analyzes the histogram and compresses it into the resulting
histogram according to the following parameters:

The first step is copied unchanged.

The high-frequency steps are copied unchanged.

The consecutive range steps are collapsed into the resulting step, so the
total weight of the collapsed step is no bigger than one-thirtieth of the
value.

The final histogram in sysstatistics:

148

Has range steps generated in a way similar for a 30-step update statistics,
and high frequency ranges are isolated as if the histogram were created
with 600 steps.

The total number of steps in the resulting histogram may differ between 30
and 600 values.

For equally distributed data, the value should be very close to 30.

More frequent values in the table means more steps in the histogram.

If a column has few different values, all those values may appear as highfrequency cells.

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

You can achieve the same result by increasing the number of histogram steps to
600, but this uses more resources in the buffer and procedure cache
histogram tuning factor minimizes the resources histograms consume, and
increases resource usage only when it is in the best interest for optimization,
for example, when there is uneven distribution of data in a column, or highly
duplicated values within a column. In this situation, up to 600 histogram steps
are used. However, in most cases, histogram tuning factor uses the default value
(30 in the example above).

housekeeper free write percent

Summary information
Default value

Range of values
Status

0100
Dynamic

Display level
Required role

Intermediate
System administrator

Configuration group

SQL Server Administration

housekeeper free write percent specifies the maximum percentage by which the

housekeeper wash task can increase database writes.

For example, to stop the housekeeper task from working when the frequency
of database writes reaches 5 percent above normal, set housekeeper free write
percent to 5.
When Adaptive Server has no user tasks to process, the housekeeper wash task
automatically begins writing changed pages from cache to disk. These writes
result in improved CPU utilization, decreased need for buffer washing during
transaction processing, and shorter checkpoints.
In applications that repeatedly update the same database page, the housekeeper
wash may initiate some unnecessary database writes. Although these writes
occur only during the servers idle cycles, they may be unacceptable on
systems with overloaded disks.
The table and index statistics that are used to optimize queries are maintained
in memory structures during query processing. When these statistics change,
the changes are not written to the systabstats table immediately, to reduce I/O
contention and improve performance. Instead, the housekeeper chores task
periodically flushes statistics to disk.

System Administration Guide: Volume 1

149

Configuration parameters

The default value allows the housekeeper wash task to increase disk I/O by a
maximum of 1 percent. This results in improved performance and recovery
speed on most systems.
To disable the housekeeper wash task, set the value of housekeeper free write
percent to 0.

Set this value to 0 only if disk contention on your system is high, and it cannot
tolerate the extra I/O generated by the housekeeper wash task.
If you disable the housekeeper tasks, keep statistics current. Commands that
write statistics to disk are:

update statistics

dbcc checkdb (for all tables in a database) or dbcc checktable (for a single

table)

sp_flushstats

Run one of these commands on any tables that have been updated since the last
time statistics were written to disk, at the following times:

Before dumping a database

Before an orderly shutdown

After restarting, following a failure or orderly shutdown; in these cases,

you cannot use sp_flushstatsyou must use update statistics or dbcc
commands

After any significant changes to a table, such as a large bulk copy

operation, altering the locking scheme, deleting or inserting large numbers
of rows, or performing a truncate table command

To allow the housekeeper wash task to work continuously, regardless of the

percentage of additional database writes, set housekeeper free write percent to
100.
Use sp_sysmon to monitor housekeeper performance. See the Performance
and Tuning Series: Monitoring Adaptive Server with sp_sysmon.
You might also want to look at the number of free checkpoints initiated by the
housekeeper task. The Performance and Tuning Series: Basics describes this
output.

150

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

i/o accounting flush interval

Summary information
Default value
Range of values

1000
12147483647

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
SQL Server Administration

i/o accounting flush interval specifies the amount of time, in machine clock ticks,

that Adaptive Server waits before flushing I/O statistics for each user from
sysprocesses to syslogins. This is used for charge-back accounting.

When a user logs in to Adaptive Server, the server begins accumulating I/O
statistics for that user process in sysprocesses. When the value of i/o accounting
statistics interval is exceeded, or a user logs off Adaptive Server, the
accumulated I/O statistics for that user are flushed from sysprocesses to
syslogins. These statistics continue accumulating in syslogins until you clear
the totals by using sp_clearstats. You can display the current totals from
syslogins by using sp_reportstats.
The value to which you set i/o accounting flush interval depends on the type of
reporting you intend to do. If you run reports on a monthly basis, set i/o
accounting flush interval to a relatively high value. With infrequent reporting, it
is less critical that the data in syslogins be updated frequently.
If you perform periodic ad hoc selects on the totio column syslogins to
determine I/O volume by process, set i/o accounting flush interval to a lower
value. Doing so increases the likelihood of the data in syslogins being current
when you execute your selects.
If you do not report on I/O statistics at all, set i/o accounting flush interval to its
maximum value. This reduces the number of times syslogins is updated and the
number of times its pages must be written to disk.

i/o batch size

Summary information
Default value
Range of values

100
12147483647

Status

Dynamic

System Administration Guide: Volume 1

151

Configuration parameters

Summary information
Display level
Required role

Comprehensive
System administrator

Configuration group

SQL Server Administration

i/o batch size sets the number of writes issued in a batch before the task goes to

sleep. Once this batch is completed, the task is woken up, and the next batch of
writes are issued, ensuring that the I/O subsystem is not flooded with many
simultaneous writes. Setting i/o batch size to the appropriate value can improve
the performance of operations like checkpoint, dump database, select into, and
so on.

i/o polling process count

Summary information
Default value

10

Range of values
Status

1 2147483647
Dynamic

Display level
Required role

Comprehensive
System administrator

Configuration group

SQL Server Administration

Note i/o polling process count functions only when you configure Adaptive

Server for process kernel mode; it is nonfunctional for threaded kernel mode.
i/o polling process count specifies the maximum number of processes that

Adaptive Server can run before the scheduler checks for disk and network I/O
completions. Tuning i/o polling process count affects both the response time and
throughput of Adaptive Server.
Adaptive Server checks for disk or network I/O completions:

If the number of tasks run since the last time Adaptive Server checked for
I/O completions equals the value for i/o polling process count, and

At every Adaptive Server clock tick.

As a general rule, increasing the value of i/o polling process count increases
throughput for applications that generate a lot of disk and network I/O.
Conversely, decreasing the value improves process response time in these
applications, possibly at the risk of lowering throughput.

152

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

If your applications create both I/O and CPU-bound tasks, tuning i/o polling
process count to a low value (1 2) ensures that I/O-bound tasks get access to
CPU cycles.
For OLTP applications (or any I/O-bound application with user connections
and short transactions), tuning i/o polling process count to a value in the range
of 20 30 may increase throughput, but may also increase response time.
When tuning i/o polling process count, consider these other parameters:

sql server closk tick length, which specifies the duration of the Adaptive
Server clock tick in microseconds. See sql server clock tick length on
page 256.

time slice, which specifies the number of clock ticks the the Adaptive
Server scheduler allows a user process to run. See time slice on page
271.

cpu grace time, which specifies the maximum amount of time, in clock

ticks, a user process can run without yielding the CPU before Adaptive
Server preempts it and terminates it with a timeslice error. See cpu grace
time on page 104.
Use sp_sysmon to determine the effect of changing i/o polling process count.
See the Performance and Tuning Series: Monitoring Adaptive Server with
sp_sysmon.

identity burning set factor

Summary information
Default value

5000

Range of values
Status

19999999
Static

Display level
Required role

Intermediate
System administrator

Configuration group

SQL Server Administration

IDENTITY columns are of type numeric and scale zero whose values are
generated by Adaptive Server. Column values can range from a low of 1 to a
high determined by the column precision.

System Administration Guide: Volume 1

153

Configuration parameters

For each table with an IDENTITY column, Adaptive Server divides the set of
possible column values into blocks of consecutive numbers, and makes one
block at a time available in memory. Each time you insert a row into a table,
Adaptive Server assigns the IDENTITY column the next available value from
the block. When all the numbers in a block have been used, the next block
becomes available.
This method of choosing IDENTITY column values improves server
performance. When Adaptive Server assigns a new column value, it reads the
current maximum value from memory and adds 1. Disk access becomes
necessary only after all values within the block have been used. Because all
remaining numbers in a block are discarded in the event of server failure (or
shutdown with nowait), this method can lead to gaps in IDENTITY column
values.
Use identity burning set factor to change the percentage of potential column
values that is made available in each block. This number should be high
enough for good performance, but not so high that gaps in column values are
unacceptably large. The default value, 5000, releases .05 percent of the
potential IDENTITY column values for use at one time.
To get the correct value for sp_configure, express the percentage in decimal
form, and then multiply it by 10 7 (10,000,000). For example, to release 15
percent (.15) of the potential IDENTITY column values at a time, specify a
value of .15 times 107 (or 1,500,000) in sp_configure.

identity grab size

Summary information
Default value

Range of values
Status

12147483647
Dynamic

Display level
Required role

Intermediate
System administrator

Configuration group

SQL Server Administration

identity grab size allows each Adaptive Server process to reserve a block of

IDENTITY column values for inserts into tables that have an IDENTITY
column.

154

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

This is useful if you are performing inserts, and you want all the inserted data
to have contiguous IDENTITY numbers. For instance, if you are entering
payroll data, and you want all records associated with a particular department
to be located within the same block of rows, set identity grab size to the number
of records for that department.
identity grab size applies to all users on Adaptive Server. Large identity grab size
values result in large gaps in the IDENTITY column when many users insert
data into tables with IDENTITY columns.

Sybase recommends that you set identity grab size to a value large enough to
accommodate the largest group of records you want to insert into contiguous
rows.

identity reservation size

Summary information
Default value

Range of values
Status

12147483647
Dynamic

Display level
Required role

Comprehensive
System administrator

Configuration group

SQL Server Administration

identity reservation size sets a limit for the number of identity values.

idle migration timeout

Summary information
Default value

60

Valid values
Status

0 32767
Dynamic

Display level
Required role

Comprehensive
System administrator

Configuration group

Shared Disk Cluster

idle migration timeout specifies the amount of time after which an idle

connection is closed without invalidating the migration request sent to the

client, allowing you to stop an instance after a specified period of time without
waiting for idle client connections to migrate.

System Administration Guide: Volume 1

155

Configuration parameters

Setting idle migration timeout to a high value slows down a graceful shutdown
because the instance must wait the specified period of time for all idle
connections that issued a migration request without the client having initiated
migration.

job scheduler interval

Summary information
Default value
Range of values

1 (in minutes)
1 600

Status
Display level

Dynamic
Comprehensive

Required role

System administrator

Configuration group

SQL Server Administration

job scheduler interval sets the interval when the Job Scheduler checks which

scheduled jobs are due to be executed.

job scheduler tasks

156

Summary information
Default value

Range of values
Status

1 640
Dynamic

Display level
Required role

Comprehensive
System administrator

Configuration group

SQL Server Administration

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

job scheduler tasks sets the maximum number of jobs that can run
simultaneously through Job Scheduler.

Note If you increase the value of job scheduler tasks to a higher value, you
must increase the value for number of user connections by at least twice the
value you incremented job scheduler tasks before starting the Job Scheduler.

However, if the Adaptive Server running the scheduled jobs is the same
Adaptive Server that is hosting the Job Scheduler, you must increase the value
for number of user connections by three times the value you incremented job
scheduler tasks before starting the Job Scheduler.
Increasing the number of user connections may require that you increase the
value for max memory.

js job output width

Summary information
Default value
Range of values

80
1 32768

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
SQL Server Administration

js job output width determines the line width the output uses for jobs stored in the
js_output table.

kernel mode
Summary information
Default value

threaded

Range of values
Status

Static

Display level
Required role

Basic
System administrator

Configuration group

SQL Server Administration

System Administration Guide: Volume 1

threaded or process

157

Configuration parameters

kernel mode determines the mode the Adaptive Server kernel uses, threaded or
process. In threaded mode, Adaptive Server uses operating system threads to
support Adaptive Server engines and specialized tasks. In this mode, Adaptive
Server is a single process running on the operating system. In process mode,
Adaptive Server uses individual processes to support Adaptive Server engines.
In this mode, each engine is a distinct process running on the operating system.
The process mode is the same kernel used by Adaptive Server versions earlier
than 15.7.

Note Adaptive Server supports only threaded kernel mode on the Windows

platform.
The values for kernel mode are character data, so you must use 0 as a
placeholder for the second sp_configure parameter, which must be numeric,
and specify threaded or process as the third parameter. See Unit specification
using sp_configure on page 64.
Sybase assumes you use threaded mode in your production server, and includes
process mode for backward compatibilty. Process mode may not support
features for Adaptive Server 15.7 and later.
See Chapter 5, Managing Multiprocessor Servers, in the System
Administration Guide: Volume 2.

kernel resource memory

Summary information

158

Default value

Solaris 6184

Range of values

All other platforms 4096

75 2147483647

Status
Display level

Dynamic
Basic

Required role
Configuration group

System administrator
Phsyical Memory

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

kernel resource memory determines the size, in 2K pages, of the kernel resource
memory pool from which all thread pools and other kernel resources are
allocated memory. The amount of memory available for kernel resource
memory depends on the value of max memory. If max memory has an
insufficient amount of memory to allocate to kernel resource memory, Adaptive
Server issues an error message that includes the value to which you must raise
max memory.

The amount of kernel resoure memory required is determined mostly by a

combination of max online engines and number of user connections. For
configurations of 8 engines or fewer, Sybase recommends that you add one
page of kernel resource memory for every two user connections above 100.
For configurations of 9 engines or more, add 1 page of kernel resource memory
for every user connection.
For example, a server with 6 engines and 200 user connections should add a
value of 100 to the default value. A server with 16 engines and 10000 user
connections would add a value of 10000 to the default value.

license information
Summary information
Default value

25

Valid values
Status

0231
Dynamic

Display level
Required role

Comprehensive
System administrator

Configuration group

SQL Server Administration

license information allows Sybase system administrators to monitor the number

of user licenses used in Adaptive Server. Enabling this parameter only
monitors the number of licenses issued; it does not enforce the license
agreement.

If license information is set to 0, Adaptive Server does not monitor license use.
If license information is set to a number greater than 0, the housekeeper chores
task monitors the number of licenses used during the idle cycles in Adaptive
Server. Set license information to the number of licenses specified in your
license agreement.
If the number of licenses used is greater than the number to which license
information is set, Adaptive Server writes this message to the error log:

System Administration Guide: Volume 1

159

Configuration parameters

WARNING: Exceeded configured number of user licenses

At the end of each 24-hour period, the maximum number of licenses used
during that time is added to the syblicenseslog table. The 24-hour period
restarts if Adaptive Server is restarted.
See Chapter 3, Managing Adaptive Server Logins and Database Users, in the
Security Administration Guide.

lock address spinlock ratio

Summary information
Default value
Range of values

100
12147483647

Status
Display level

Static
Comprehensive

Required role
Configuration group

System administrator
Lock Manager

For Adaptive Servers running with multiple engines, the address lock spinlock
ratio sets the number of rows in the internal address locks hash table that are
protected by one spinlock.
Adaptive Server manages the acquiring and releasing of address locks using an
internal hash table with 1031 rows (known as hash buckets). This table can use
one or more spinlocks to serialize access between processes running on
different engines.
The default value for address lock spinlock ratio defines 11 spinlocks for the
address locks hash table. The first 10 spinlocks protect 100 rows each, and the
eleventh spinlock protects the remaining 31 rows. If you specify a value of
1031 or greater for address lock spinlock ratio, Adaptive Server uses only 1
spinlock for the entire table.

lock hashtable size

Summary information

160

Default value
Range of values

2048
12147483647

Status
Display level

Static
Comprehensive

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

Summary information
Required role
Configuration groups

System administrator
Lock Manager, Memory Use

lock hashtable size specifies the number of hash buckets in the lock hash table.
This table manages all row, page, and table locks, and all lock requests. Each
time a task acquires a lock, the lock is assigned to a hash bucket, and each lock
request for that lock checks the same hash bucket. Setting this value too low
results in large numbers of locks in each hash bucket and slows the searches.
On Adaptive Servers with multiple engines, setting this value too low can also
lead to increased spinlock contention. Do not set the value to less than the
default value, 2048.
lock hashtable size must be a power of 2. If the value you specify is not a power
of 2, sp_configure rounds the value to the next highest power of 2 and prints an
informational message.

The optimal hash table size is a function of the number of distinct objects
(pages, tables, and rows) that can be locked concurrently. The optimal hash
table size is at least 20 percent of the number of distinct objects that need to be
locked concurrently. See the Performance and Tuning Series: Locking and
Concurrency Control.
However, if you have a large number of users and have had to increase the
number of locks parameter to avoid running out of locks, use sp_sysmon to
check the average hash chain length at peak periods. If the average length of
the hash chains exceeds 4 or 5, consider increasing the value of lock hashtable
size from its current setting to the next power of 2.
The hash chain length may be high during large insert batches, such as bulk
copy operations. This is expected behavior, and does not require you to reset
lock hashtable size.

lock scheme
Summary information
Default value
Range of values

allpages
allpages, datapages, datarows

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
Lock Manager

System Administration Guide: Volume 1

161

Configuration parameters

lock scheme sets the default locking scheme to be used by create table and
select into commands when a lock scheme is not specified in the command.

The values for lock scheme are character data, so you must use 0 as a
placeholder for the second parameter, which must be numeric, and specify
allpages, datapages, or datarows as the third parameter:
sp_configure "lock scheme", 0, datapages

lock shared memory

Summary information
Default value
Valid values

0 (off)
0 (off), 1 (on)

Status
Display level

Static
Comprehensive

Required role
Configuration group

System administrator
Physical Memory

lock shared memory disallows swapping of Adaptive Server pages to disk and

allows the operating system kernel to avoid the servers internal page locking
code. This can reduce disk reads, which are expensive.
Not all platforms support shared memory locking. Even if your platform does,
lock shared memory may fail due to incorrectly set permissions, insufficient

physical memory, or for other reasons. See operating system documentation for
your platform for information on shared memory locking.

lock spinlock ratio

Summary information

162

Default value
Range of values

85
12147483647

Status
Display level

Static
Comprehensive

Required role
Configuration groups

System administrator
Lock Manager, Memory Use

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

Adaptive Server manages the acquiring and releasing of locks using an internal
hash table with a configurable number of hash buckets. On SMP systems, this
hash table can use one or more spinlocks to serialize access between processes
running on different engines. To set the number of hash buckets, use lock
hashtable size.
For Adaptive Servers running with multiple engines, lock spinlock ratio sets a
ratio that determines the number of lock hash buckets that are protected by one
spinlock. If you increase lock hashtable size, the number of spinlocks
increases, so the number of hash buckets protected by one spinlock remains the
same.
The Adaptive Server default value for lock spinlock ratio is 85. With lock
hashtable size set to the default value of 2048, the default spinlock ratio defines
26 spinlocks for the lock hash table. See Chapter 5, Managing Mulitprocessor
Servers, in System Administration Guide: Volume 2.
sp_sysmon reports on the average length of the hash chains in the lock hash
table. See the Performance and Tuning Series: Monitoring Adaptive Server
with sp_sysmon.

lock table spinlock ratio

Summary information
Default value

20

Range of values
Status

12147483647
Static

Display level
Required role

Comprehensive
System administrator

Configuration group

Lock Manager

For Adaptive Servers running with multiple engines, table lock spinlock ratio
sets the number of rows in the internal table locks hash table that are protected
by one spinlock.
Adaptive Server manages the acquiring and releasing of table locks using an
internal hash table with 101 rows (known as hash buckets). This table can use
one or more spinlocks to serialize access between processes running on
different engines.

System Administration Guide: Volume 1

163

Configuration parameters

The Adaptive Server default value for table lock spinlock ratio is 20, which
defines 6 spinlocks for the table locks hash table. The first 5 spinlocks protect
20 rows each; the sixth spinlock protects the last row. If you specify a value of
101 or greater for table lock spinlock ratio, Adaptive Server uses only 1 spinlock
for the entire table.

lock timeout pipe active

Summary information
Default value
Range of values

1 (off)
0 (on), 1 (off)

Status

Dynamic

Display level
Required role

Comprehensive
System administrator

Configuration group

Monitoring

lock timeout pipe active controls whether Adaptive Server collects lock timeout
messages. If lock timeout pipe active and lock timeout pipe max messages are
enabled, Adaptive Server collects the data for each lock timeout that occurs.

Retrieve the lock timeout messages from the monLockTimeout monitor table.

lock timeout pipe max messages

Summary information
Default value

Range of values
Status

0 - 2147483648
Dynamic

Display level
Required role

Comprehensive
System administrator

Configuration group

Monitoring

lock timeout pipe max messages controls the maximum number of rows per
engine in the lock timeout pipe, which determines the maximum number of
rows that can be returned by the monLockTimeout monitoring table.

164

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

lock wait period

Summary information
Default value
Range of values

2147483647
0 2147483647

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
Lock Manager

lock wait period limits the number of seconds that tasks wait to acquire a lock
on a table, data page, or data row. If the task does not acquire the lock within
the specified time period, Adaptive Server returns error message 12205 to the
user and rolls back the transaction.

The lock wait option of the set command sets a session-level number of seconds
that a task waits for a lock. It overrides the server-level setting for the session.
lock wait period, used with the session-level setting set lock wait nnn, is
applicable only to user-defined tables. These settings have no influence on
system tables.

At the default value, all processes wait indefinitely for locks. To restore the
default value, reset the value to 2147483647 or enter:
sp_configure "lock wait period", 0, "default"

log audit logon failure

Summary information
Default value
Range of values

0 (off)
0 (off), 1 (on)

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
Error Log

log audit logon failure specifies whether to log unsuccessful Adaptive Server

logins to the Adaptive Server error log and, on Windows servers, to the
Windows Event Log, if event logging is enabled.

System Administration Guide: Volume 1

165

Configuration parameters

log audit logon success

Summary information
Default value
Range of values

0 (off)
0 (off), 1 (on)

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
Error Log

log audit logon success specifies whether to log successful Adaptive Server
logins to the Adaptive Server error log and, on Windows servers, to the
Windows Event Log, if event logging is enabled.

max async i/os per engine

Summary information
Default value
Range of values

Platform dependent
1 platform-dependent value

Status
Display level

Static
Comprehensive

Required role
Configuration group

System administrator
O/S Resources

max async i/os per engine specifies the maximum number of outstanding
asynchronous disk I/O requests for a single engine at one time.

On the Linux platform

On the Linux platform, max async i/os per engine controls the number of
asynchronous IOs reserved from the operating system.
In threaded mode, Adaptive Server reserves the value for max async i/os per
engine for the engine server. In process mode, Adaptive Server reserves the
value for max async i/os per engine for each engine. For example, if you set max
async i/os per engine to 4096 and there are 4 engines, in threaded mode
Adaptive Server reserves 4096 I/Os, but in process mode Adaptive Server
reserves 16384 I/Os.
Your system may benefit from using a number greater than the default value.

166

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

You can use sp_sysmon to help tune max async i/os per engine. sp_sysmons
disk i/o section contains information about the maximum number of
outstanding IOs for each engine during the sample period and the number of
I/Os that were delayed because of engine or operating system limits. Generally,
any I/Os delayed by engine limits indicate that you should increase the value
of max async i/os per engine.
Whether Adaptive Server can perform asynchronous IO on a device depends
on whether or not this device support kernel asynchronous I/O (KAIO). The
Linux kernel requires that you implement kernel asynchronous I/O support at
the file system level. Most major file systems provide support for kernel
asynchronous I/O, including ext3, xfs, jfs, and raw devices. The tmpfs file
system does not support kernel asynchronous I/O. If the device does not
support kernel asynchronous I/O, Adaptive Server cannot perform
asyncronous IO on that device, and instead reverts to standard synchronous IO
for all reads and writes to that device. Adaptive Server prints a message similar
to the following in the error log indicating that the device has switched to
synchronous IO:
00:00000:00001:2006/12/15 11:47:17.98 kernel Virtual device
'/dev/shm/tempdb.dat' does not support kernel asynchronous i/o. Synchronous i/o
will be used for this device.

max async i/os per server

Summary information
Default value

Platform dependent

Range of values
Status

1 platform dependent value

Static

Display level
Required role

Comprehensive
System administrator

Configuration group

O/S Resources

max async i/os per server specifies the maximum number of asynchronous disk
I/O requests that can be outstanding for Adaptive Server at one time. This limit
is not affected by the number of online engines per Adaptive Server. max async
i/os per engine limits the number of outstanding I/Os per engine.

System Administration Guide: Volume 1

167

Configuration parameters

Most operating systems limit the number of asynchronous disk I/Os that can be
processed at any one time; some operating systems limit the number per
operating system process, some limit the number per system, and some do
both. If an application exceeds these limits, the operating system returns an
error message. Because operating system calls are relatively expensive, it is
inefficient for Adaptive Server to attempt to perform asynchronous I/Os that
get rejected by the operating system.
To avoid this, Adaptive Server maintains a count of the outstanding
asynchronous I/Os per engine and per server; if an engine issues an
asynchronous I/O that would exceed either max async i/os per engine or max
async i/os per server, Adaptive Server delays the I/O until enough outstanding
I/Os have completed to fall below the exceeded limit.
For example, assume an operating system limit of 200 asynchronous I/Os per
system and 75 per process and an Adaptive Server with three online engines.
The engines currently have a total of 200 asynchronous I/Os pending,
distributed according to the following table:
Engine

Number of
I/Os pending

60

Engine 0 delays any further asynchronous I/Os until the total for the server is under the
operating system per-system limit and then continues issuing asynchronous I/Os.

75

65

Engine 1 delays any further asynchronous I/Os until the per-engine total is under the
operating system per-process limit and then continues issuing asynchronous I/Os.
Engine 2 delays any further asynchronous I/Os until the total for server is under the
operating system per-system limit and then continues issuing asynchronous I/Os.

Outcome

All I/Os (both asynchronous and synchronous) require a disk I/O structure, so
the total number of outstanding disk I/Os is limited by the value of disk i/o
structures. It is slightly more efficient for Adaptive Server to delay the I/O
because it cannot get a disk I/O structure than because the I/O request exceeds
max i/os per server. Set max async i/os per server equal to the value of disk i/o
structures. See disk i/o structures on page 116.
If the limits for asynchronous I/O can be tuned on your operating system, make
sure they are set high enough for Adaptive Server. There is no penalty for
setting them as high as needed.
Use sp_sysmon to see if the per server or per engine limits are delaying I/O on
your system. If sp_sysmon shows that Adaptive Server exceeded the limit for
outstanding requests per engine or per server, raise the value of the
corresponding parameter. See the Performance and Tuning Series: Monitoring
Adaptive Server with sp_sysmon.

168

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

max cis remote connections

Summary information
Default value
Range of values

0
02147483647

Status
Display level

Dynamic
Basic

Required role
Configuration group

System administrator
Component Integration Services

max cis remote connections specifies the maximum number of concurrent

Client-Library connections that can be made to remote servers by Component

Integration Services.
By default, Component Integration Services allows up to four connections per
user to be made simultaneously to remote servers. If you set the maximum
number of users to 25, as many as 100 simultaneous Client-Library
connections are allowed by Component Integration Services.
If this number does not meet the needs of your installation, you can override
the setting by specifying exactly how many outgoing Client-Library
connections you want the server to be able to make at one time.

max concurrently recovered db

Summary information
Default value
Valid values

0
1 number of engines at start-up minus 1

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
Backup/Recovery

max concurrently recovered db determines the degree of parallelism. The

minimum value is 1, which uses serial recovery, but you can also use the
default value of 0, to use a self-tuning approach. The maximum value is the
number of engines at start-up minus 1. max concurrently recovered db is also
limited by the value of the configuration parameter number of open databases.

System Administration Guide: Volume 1

169

Configuration parameters

max memory
Summary information
Default value
Range of values

Platform-dependent
Platform-dependent minimum 2147483647

Status
Display level

Dynamic
Basic

Required role
Configuration groups

System administrator
Memory Use, Physical Memory

max memory specifies the maximum amount of total physical memory that you
can configure Adaptive Server to allocate. max memory must be greater than

the total logical memory consumed by the current configuration of Adaptive

Server.
There is no performance penalty for configuring Adaptive Server to use the
maximum memory available to it on your computer. However, assess the other
memory needs on your system, or Adaptive Server may not be able to acquire
enough memory to start.
See see Chapter 3, Configuring Memory, in System Administration Guide:
Volume 2.
If Adaptive Server cannot start

When allocate max shared memory is set to 1, Adaptive Server must have the
amount of memory available that is specified by max memory. If the memory
is not available, Adaptive Server does not start. If this occurs, reduce the
memory requirements for Adaptive Server by manually changing the value of
max memory in the servers configuration file. You can also change the value
of allocate max shared memory to 0 so that not all memory required by max
memory is required at start-up.
You may also want to reduce the values for other configuration parameters that
require large amounts of memory. Then restart Adaptive Server to use the
memory specified by the new values. If Adaptive Server fails to start because
the total of other configuration parameter values is higher than the max memory
value, see Chapter 3, Configuring Memory, in System Administration
Guide: Volume 2 for information about configuration parameters that use
memory.

170

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

max native threads per engine

Summary information
Default value
Maximum values

50
50 1000

Status
Display level

Dynamic
Intermediate

Required role
Configuration group

System administrator
User Environment

Note max native threads per engine is ignored in threaded mode.

max native threads per engine defines the maximum number of native threads
the server spawns per engine. When the limit for the native threads is reached,
Adaptive Server sessions that require a native thread sleep until another session
releases a native thread.

max nesting level

In Adaptive Server 15.0.3 and later, the maximum nesting level has been
increased to 100, and the default value to 50.
Summary information
Default value
Range of values

50
16 100

Status
Display level

Static
Comprehensive

Required role
Configuration group

System administrator
User environment

maximum nesting level sets the maximum nesting level for stored procedures

and triggers. Each increased nesting level requires about 160 bytes of
additional memory. For example, if you increase the nesting level from 16 to
26, Adaptive Server requires an additional 1600 bytes of memory.

max network packet size

Summary information
Default value

System Administration Guide: Volume 1

512

171

Configuration parameters

Summary information
Range of values
Status

51265024
Static

Display level
Required role

Intermediate
System administrator

Configuration group

Network Communication

max network packet size specifies the maximum network packet size that can
be requested by clients communicating with Adaptive Server.

If some of your applications send or receive large amounts of data across the
network, these applications can achieve significant performance improvement
by using larger packet sizes. Two examples are large bulk copy operations and
applications that read or write large text, unitext, and image values.
Generally, you want:

default network packet size to be small for users who perform short queries,

and

max network packet size to be large enough to allow users who send or
receive large volumes of data to request larger packet sizes.

max network packet size must always be as large as, or larger than, the default
network packet size. Values that are not even multiples of 512 are rounded

down.
For client applications that explicitly request a larger network packet size to
receive, you must also configure additional network memory. See additional
network memory on page 82.
Open Client Server cannot accept a network packet size greater than 64K.
See bcp and isql in the Utility Guide for information on using larger packet sizes
from these programs. Open Client Client-Library documentation includes
information on using variable packet sizes.
Choosing packet sizes

For best performance, choose a server packet size that works efficiently with
the underlying packet size on your network. The goals are:

172

Reducing the number of server reads and writes to the network

Reducing unused space in network packets (increasing network

throughput)

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

For example, if your network packet size carries 1500 bytes of data, setting the
Adaptive Server packet size to 1024 (512*2) will probably achieve better
performance than setting it to 1536 (512*3). Figure 5-3 shows how four
different packet size configurations would perform in such a scenario.
Figure 5-3: Factors in determining packet size

Underlying network packets: 1500 bytes after overhead

Packet size 512
Used
1024 bytes
Unused
476 bytes
% Used:
68%
2 server reads
Packet size 1024
Used
1024 bytes
Unused
476 bytes
% Used:
68%
1 server read

Depending on amount of data, network packets may have

1 or 2 packets

Should yield improved performance over default of 512

Packet size 2560

Used
2560 bytes
Unused
440 bytes
% Used
85%
2 server reads
Possibly the best option of illustrated choices
Packet size 1536
Used
1536 bytes
Unused
1464 bytes
% Used
51%
2 server reads
Probably the worst option of illustrated choices
Key:
Overhead

Data

Unused

After you determine the available data space of the underlying packets on your
network, perform your own benchmark tests to determine the optimum size for
your configuration.

System Administration Guide: Volume 1

173

Configuration parameters

Use sp_sysmon to see how changing max network packet size affects network
I/O management and task switching. For example, try increasing max network
packet size and then checking sp_sysmon output to see how this affects bcp for
large batches. See the Performance and Tuning Series: Monitoring Adaptive
Server with sp_sysmon.

max number network listeners

Summary information
Default value
Range of values

5
02147483647

Status

Static

Display level
Required role

Comprehensive
System administrator

Configuration groups

Memory Use, Network Communication

max number network listeners specifies the maximum number of network

listeners allowed by Adaptive Server at one time.

Each master port has one network listener. Generally, there is no need to have
multiple master ports, unless your Adaptive Server must communicate over
more than one network type. Some platforms support both socket and TLI
(Transport Layer Interface) network interfaces. See the Configuration Guide
for your platform for information on supported network types.

max online engines

Summary information
Default value

Range of values

1 1024 (threaded mode)

1 128 (process mode)

174

Status
Display level

Static
Intermediate

Required role
Configuration groups

System administrator
Memory Use, Processors

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

max online engines places an upper limit of the number of engine threads that

can be brought online. It does not take into account the number of CPUs
available at start-up, and allows users to add CPUs at a later date. See Chapter
5, Managing Mulitprocessor Servers, in System Administration Guide:
Volume 2 for a detailed discussion of how to set this parameter for your SMP
environment.
At start-up, Adaptive Server starts with a single engine thread and completes
its initialization, including recovery of all databases. Its final task is to allocate
additional server engines. Each engine accesses common data structures in
shared memory.
When tuning the max engines online parameter:

Never have more engine threads online than there are CPUs.

max engines online must be large enough to allow Adaptive Server to bring
all the engine thread pools online simultaneously.

Depending on overall system load (including applications other than

Adaptive Server), you may achieve optimal throughput by leaving some
CPUs free to run non-Adaptive Server processes.

You can achieve better throughput by running fewer engine threads with
high CPU use, rather than by running more engine threads with low CPU
use.

Scalability is application-dependent. Conduct extensive benchmarks on

your application to determine the best configuration of online engine
threads.

In process mode, use sp_engine to take engine threads offline or to bring

them online. You can take all engine threads offline except engine zero.

See Chapter 3, Using Engines and CPUs in the Performance and Tuning
Series: Basics .

max online Q engines

Summary information
Default value

Range of values
Status

0 127
Static

Display level
Required role

Comprehensive
System administrator

System Administration Guide: Volume 1

175

Configuration parameters

Summary information
Configuration group

Processors

(Process mode only) max online Q engines specifies the maximum number of
Q engines you can have online, and is required for MQ (see the Active
Messaging documentation).
The maximum number of Q engines depends on the value of max online
engines:

max online Q engines cannot be greater than max online engines minus
number of engines at startup. That is, if the value for max online engines is
57, and the value for number of engines at startup is 28, the value for max
online Q engines cannot be greater than 29.

Setting max online Q engines reserves the high range of max online engines
for Q engines. Once you set max online Q engines, Adaptive Server
engines can no longer use the engines in the range that is reserved for Q
engines. For example, if you set max online engines to 10 and set max
online Q engines to 4:

Adaptive Server issues an error if you attempt to bind an engine group

to engines 6, 7, 8, or 9

You cannot change the value of number of engines at startup to 7, 8, or

9

You cannot add engines 6, 7, 8, or 9 to an engine group

Because setting max online Q engines may affect existing production

environments, Sybase recommends that you increase the value of max online
engines by the same amount that you increase max online Q engines. For
example, if you increase max online Q engines by 4, also increase max online
engines by 4.

max parallel degree

176

Summary information
Default value

Range of values
Status

1255
Dynamic

Display level
Required role

Basic
System administrator

Configuration group

Query Tuning

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

max parallel degree specifies the server-wide maximum number of worker

processes allowed per query. This is called the maximum degree of

parallelism.
If max parallel degree is too low, the performance gain for a given query may
not be as significant as possible; if max parallel degree is too high, the server
may compile plans that require more processes than are actually available at
execution time, or the system may become saturated, resulting in decreased
throughput. To enable parallel partition scans, set this parameter to be equal to
or greater than the number of partitions in the table you are querying.
The value of this parameter must be less than or equal to the current value of
number of worker processes.

If you set max parallel degree to 1:

Adaptive Server scans all tables or indexes serially.

Adaptive Server forces serial query execution and the optimizer may
select plans with a higher parallel degree than if it is disabled.

Changing max parallel degree causes all query plans in the procedure cache to
be invalidated, and new plans are compiled the next time you execute a stored
procedure or trigger.
See Chapter 9, Parallel Sorting in the Performance and Tuning Series: Query
Processing and Abstract Plans.

max pci slots

Summary information
Default value

Range of values
Status

0 30
Static

Display level
Required role

Comprehensive
System administrator

Configuration group

User Environment

Sets the maximum number of PCI slots Adaptive Server allows. The values
are:

System Administration Guide: Volume 1

177

Configuration parameters

0, 1 default bridge with one PCA.

Note JVM support requires a single slot. Do not increase the number of

slots.

2 30 allocated for future releases.

For more information about PCI slots, see Java in Adaptive Server Enterprise.

max query parallel degree

Summary information
Default value
Range of values

1
1 255

Status
Display level

Static
Comprehensive

Required role
Configuration group

System administrator
Query Tuning

Used when Adaptive Server is in compatibility mode. Defines the number of

worker processes to use for a given query. This parameter is relevant only if
you do not want to enable parallelism globally. The value for number of worker
process cannot be less than the value for max query parallel degree.
See Chapter 5, Parallel Query Processing, in the Performance and Tuning
Series: Query Processing and Abstract Plans.
For more information about compatibility mode, see the Migration Technology
Guide.

max repartition degree

Summary information

178

Default value
Range of values

1
1 value of max parallel degree

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
Query Tuning

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

max repartition degree configures the amount of dynamic repartitioning

Adaptive Server requires, which enables Adaptive Server to use horizontal
parallelism. However, if the number of partitions is too large, the system is
flooded with worker processes that compete for resources, which degrades
performance. The value for max repartition degree enforces the maximum
number of partitions created for these resources. If all of the tables and indexes
are unpartitioned, Adaptive Server uses the value for max repartition degree to
provide the number of partitions to create as a result of repartitioning the data.

max resource granularity

Summary information
Default value

10

Range of values
Status

1 100
Dynamic

Display level
Required role

Comprehensive
System administrator

Configuration group

Query Tuning

max resource granularity indicates the maximum percentage of the systems

resources a query can use. This parameter is not enforced at execution time, but
is only a guide for the query optimizer, and does not prevent the query
processor from running queries in parallel. The query engine can avoid some
memory-intensive strategies by using max resource granularity as a guide.

max scan parallel degree

Summary information
Default value

Range of values
Status

1255
Dynamic

Display level
Required role

Basic
System administrator

Configuration group

Query Tuning

max scan parallel degree specifies the server-wide maximum degree of

parallelism for hash-based scans which may be used for the following access
methods:

Parallel index scans for partitioned and nonpartitioned tables

System Administration Guide: Volume 1

179

Configuration parameters

Parallel table scans for nonpartitioned tables

max scan parallel degree applies per table or index; that is, if max scan parallel
degree is 3, and one table in a join query is scanned using a hash-based table

scan and the second can best be accessed by a hash-based index scan, the query
can use 9 worker processes (as long as max scan parallel degree is set to 9 or
higher.
The optimizer uses max scan parallel degree as a guideline when it selects the
number of processes to use for parallel, nonpartition-based scan operations. It
does not apply to parallel sort. Because there is no partitioning to spread the
data across devices, parallel processes can be accessing the same device during
the scan. This can cause additional disk contention and head movement, which
may degrade performance. To prevent multiple disk accesses from becoming a
problem, use max scan parallel degree to reduce the maximum number of
processes that can access the table in parallel.
If this number is too low, the performance gain for a given query is not as
significant as possible; if the number is too large, the server may compile plans
that use enough processes to make disk access less efficient. A general rule is
to set this parameter to no more than 2 or 3, because it takes only 2 to 3 worker
processes to fully utilize the I/O of a given physical device.
Set the value of max scan parallel degree to less than or equal to the current
value of max parallel degree. Adaptive Server returns an error if you specify a
number larger than the max parallel degree value.
If you set max scan parallel degree to 1, Adaptive Server does not perform hashbased scans.
Changing max scan parallel degree causes all query plans in the procedure
cache to be invalidated, and new plans are compiled the next time you execute
a stored procedure or trigger.

max SQL text monitored

180

Summary information
Default value

Range of values
Status

02147483647
Static

Display level
Required role

Comprehensive
System administrator

Configuration groups

Memory Use, Monitoring

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

max SQL text monitored specifies the amount of memory allocated per user
connection for saving SQL text to memory shared by Adaptive Server Monitor.

If you do not allocate enough memory for the batch statements, the text you
want to view may be truncated. Sybase recommends that you use an initial
value of 1024 bytes of memory per user connection.
The total memory allocated from shared memory for the SQL text is the
product of max SQL text monitored multiplied by the currently configured
number of user connections.

max transfer history

Summary information
Default value

10

Range of values
Status

1 255
Dynamic

Display level
Required role

Comprehensive
System administrator

Configuration group

Adaptive Server Administration

max transfer history controls how many transfer history entries Adaptive Server
retains in the spt_TableTransfer table in each database. For each table tracked,
spt_TableTransfer retains:

(N successful entries) + (N unsuccessful entries)

Where N is the value for max transfer history.
Lowering this parameter does not automatically remove any entries from
spt_TableTransfer. Entries are removed for a given transferred table the next

time you initiate a transfer for that table. The tables successful transfer entries
are cleared if the transfer succeeds. If the transfer is unsuccessful, its failed
transfer entries are cleared.
For example, if a table has 12 successful and 9 unsuccessful history entries in
spt_configure, and you change max transfer history to 5, the next successful
transfer of that table places 5 successful entries in spt_configure, but
spt_configure retains the previous 9 failed entries.

System Administration Guide: Volume 1

181

Configuration parameters

maximum dump conditions

Summary information
Default value
Range of values

10
10100

Status
Display level

Static
Intermediate

Required role
Configuration group

System administrator
Group Diagnostics

maximum dump conditions sets the maximum number of conditions you can

specify under which Adaptive Server generates a dump of data in shared

memory.
Note This parameter is included for use only by Sybase Technical Support. Do

not modify it unless you are instructed to do so by Sybase Technical Support.

max buffers per lava operator

Summary information
Default value
Range of values

2048
500 65535

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
Query Tuning

maximum buffers per lava opeator sets an upper limit for the number of buffers
used by Lava operators that perform sorting or hashing (which are expensive
in terms of processing). Lava operators use buffers from the sessions tempdb
data cache pool as a work area for processing rows.

Lava operators often recurse through their input streams. Sorting requires
subsequent merge passes until there are enough buffers available to merge all
of the remaining runs. Hashing requires subsequent passes to build hash tables
on any spilled sets until all of the remaining data can fit into an in-memory hash
table. Some queries require less I/O if you increase max buffers per lava
operator. This is particularly true for queries that use the HASH DISTINCT,
HASH VECTOR AGGREGATE, and HASH UNION operators.

182

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

Be careful when you increase the default value of maximum buffers per lava
operator for servers with many concurrent users: Adaptive Server may allocate
more buffers solely for expensive operators, reducing the number of buffers
available for caching users tables and other sessions worktables. Use
sp_sysmon to analyze tempdb's data caching effectiveness.
maximum buffers per lava operator works with max resource granularity to limit

the number of buffers used. The limit is set to the minimum of:

The value of maximum buffers per lava operator, or,

(max resource granularity) X (the number of data buffers in tempdbs

pagesize pool)

See number of sort buffers on page 216 for information about setting the
amount of memory allocated for sort buffers.

maximum failed logins

Summary information
Default value

Range of values
Status

-1 32767
Dynamic

Display level
Required role

10
System security officer

Configuration group

Security Related

maximum failed logins allows you to set the server-wide maximum number of
failed login attempts for logins and roles.

A value of -1 indicates that the failed login count in the syslogins column
logincount is updated whenever an authentication failure occurs, but that the
account is not locked. Compare with a 0 (zero) value, which avoids
incrementing the column for every failed authentication and avoids locking the
account due to authentication failures.
See the Reference Manual: Procedures for information about using
sp_modifylogin to change the maximum failed logins for a specific role. See the
Reference Manual: Commands for information about using alter role to change
the maxiumum failed logins.

System Administration Guide: Volume 1

183

Configuration parameters

maximum job output

Summary information
Default value
Range of values

32768
02147483647

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
SQL Server Administration

maximum job output sets limit, in bytes, on the maximum output a single job can
produce. If a job produces more output than specified in maximum job output,
all the data returned above the value you enter is discarded.

memory alignment boundary

Summary information
Default value
Range of values

Logical page size

2048a 16384
a. Minimum determined by servers logical page size

Status

Static

Display level
Required role

Comprehensive
System administrator

Configuration group

Cache Manager

memory alignment boundary determines the memory address boundary on

which data caches are aligned.

Some machines perform I/O more efficiently when structures are aligned on a
particular memory address boundary. To preserve this alignment, values for
memory alignment boundary should always be powers of two between the
logical page size and 2048K.
Note The memory alignment boundary parameter is included for support of

certain hardware platforms. Do not modify it unless you are instructed to do so

by Sybase Technical Support.

184

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

memory per worker process

Summary information
Default value
Range of values

1024
10242147483647

Status
Display level

Dynamic
Basic

Required role
Configuration group

System administrator
Memory Use

memory per worker process specifies the amount of memory, in bytes, used by
worker processes. Each worker process requires memory for messaging during
query processing. This memory is allocated from a shared memory pool; the
size of this pool is memory per worker process multiplied by number of worker
processes. For most query processing, the default size is more than adequate.
If you use dbcc checkstorage, and have set number of worker processes to 1, you
may need to increase memory per worker process to 1792 bytes.

See Chapter 3, Configuring Memory, in System Administration Guide:

Volume 2.

messaging memory
Summary information
Default value
Range of values

400
60 2147483647

Status
Display level

Dynamic
Intermediate

Required role
Configuration group

System administrator
Memory Use, Physical Memory

Configures the amount of memory available for Sybase messaging.

metrics elap max

Summary information
Default value

Range of values
Status

0 2147483647
Dynamic

System Administration Guide: Volume 1

185

Configuration parameters

Summary information
Display level
Required role

Comprehensive
System administrator

Configuration group

Query Tuning

metrics elap max configures maximum elapsed time and thresholds for QP

metrics

metrics exec max

Summary information
Default value

Range of values
Status

0 2147483647
dynamic

Display level
Required role

Comprehensive
System administrator

Configuration group

Query Tuning

metrics exec max configures maximum execution time and thresholds for QP

metrics.

metrics lio max

Summary information
Default value

Range of values
Status

0 2147483647
Dynamic

Display level
Required role

Comprehensive
System administrator

Configuration group

Query Tuning

metrics lio max configures maximum logical I/O and thresholds for QP metrics.

metrics pio max

186

Summary information
Default value

Range of values

0 2147483647

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

Summary information
Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator

metrics pio max configures maximum physical I/O and thresholds for QP

metrics.

min pages for parallel scan

Summary information
Default value
Range of values

200
20 - 2147483647

Status
Display level

dynamic
Comprehensive

Required role
Configuration group

System administrator
Query Tuning

min pages for parallel scan controls the number of tables and indexes that

Adaptive Server can access in parallel. If the number of pages in a table is

below the value you set, the table is accessed serially. min pages for parallel scan
does not consider page size. If Adaptive Server accesses the indexes and tables,
Adaptive Server attempts to repartition the data, if that is appropriate, and to
use parallelism above the scans, if that is appropriate.

minimum password length

Summary information
Default value
Range of values

6
0 30

Status
Display level

Dynamic
10

Required role
Configuration group

System security officer

Security Related

System Administration Guide: Volume 1

187

Configuration parameters

minimum password length allows you to customize the length of server-wide

password values or per-login or per-role password values. The per-login or perrole minimum password length value overrides the server-wide value. Setting
minimum password length affects only the passwords you create after you have
set the value; existing password lengths are not changed.
Use minimum password length to specify a server-wide value for minimum
password length for both logins and roles. For example, to set the minimum
password length for all logins and roles to 4 characters, enter:
sp_configure "minimum password length", 4

To set minimum password length for a specific login at creation, use

sp_addlogin. For example, to create the new login joe with the password
Djdiek3, and set minimum password length for joe to 4, enter:
sp_addlogin joe, "Djdiek3", minimum password length=4

To set minimum password length for a specific role at creation, use create role.
To create the new role intern_role with the password temp244 and set the
minimum password length for intern_role to 0, enter:
create role intern_role with passwd "temp244", minimum password length 0

The original password is seven characters, but the password can be changed to
one of any length because the minimum password length is set to 0.
Use sp_modifylogin to set or change minimum password length for an existing
login. sp_modifylogin only effects user roles, not system roles. For example, to
change minimum password length for the login joe to 8 characters, enter:
sp_modifylogin "joe", @option="minimum password length", @value="8"
Note The value parameter is a character datatype; therefore, quotes are

required for numeric values.

To change the value of the overrides for minimum password length for all logins
to 2 characters, enter:
sp_modifylogin "all overrides", "minimum password length", @value="2"

To remove the overrides for minimum password length for all logins, enter:
sp_modifylogin "all overrides", @option="minimum password length", @value="-1"

Use alter role to set or change the minimum password length for an existing role.
For example, to set the minimum password length for physician_role, an
existing role, to 5 characters, enter:

188

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

alter role physician_role set minimum password length 5

To override the minimum password length for all roles, enter:

alter role "all overrides" set minimum password length -1

mnc_full_index_filter
Summary information
Default value
Range of values

2
02
0 disable.
1 enable.

Status

2 set according to the optimization goal setting.

Dynamic

Display level
Required roles

Comprehensive
System administrator

Configuration group

Query Tuning

mnc_full_index_filter prevents Adaptive Server from considering noncovered

indexes that do not have a limiting search argument at the server level, if there
is:

A column in the index

A predicate that does not have a histogram

You can use mnc_full_index_filter on data-only-locked (DOL) tables in which

you have the intelligent index scan, even though the intelligent index scan
manufactures search arguments.
Changing the value of mnc_full_index_filter does not increase the amount of
memory Adaptive Server uses.
mnc_full_index_filter is not enabled for any specific optional goal; the only way
to obtain the behavior is to explicitly enable it.

msg confidentiality reqd

Summary information
Default value
Range of values

0 (off)
0 (off), 1 (on)

Status

Dynamic

System Administration Guide: Volume 1

189

Configuration parameters

Summary information
Display level
Required role

Intermediate
System security officer

Configuration group

Security Related

msg confidentiality reqd requires that all messages into and out of Adaptive
Server be encrypted. The use security services parameter must be 1 for
messages to be encrypted.

msg integrity reqd

Summary information
Default value

0 (off)

Range of values
Status

0 (off), 1 (on)
Dynamic

Display level
Required role

Intermediate
System security officer

Configuration group

Security Related

msg integrity reqd requires that all messages be checked for data integrity. use
security services must be 1 for message integrity checks to occur. If msg
integrity reqd is set to 1, Adaptive Server allows the client connection to

succeed unless the client is using one of the following security services:
message integrity, replay detection, origin checks, or out-of-seq checks.

net password encryption required

Summary information
Default value

Range of values
Status

02
Dynamic

Display level
Required role

Intermediate
System security officer

Configuration group

Security Related

net password encryption reqd restricts login authentication to use only RSA
encryption algorithm or the Sybase proprietary algorithm. Table 5-3 describes
valid values for net password encryption reqd.

190

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

Table 5-3: Values and descriptions for net password encryption reqd
Value
0
1

Description
Allows the client to choose the encryption algorithm used for login passwords on the network, including
no password encryption.
Restricts clients to use either RSA or Sybase proprietary encryption algorithms to encrypt login passwords
on the network. This provides an incrementally restrictive setting that allows clients who have previously
connect to reconnect with the Sybase proprietary algorithm and new clients to connect with the stronger
RSA algorithm. A client that attempts to connect without using password encryption fails.
Restricts clients to use only the RSA encryption algorithms to encrypt login passwords on the network.
This provides strong RSA encryption of passwords. Clients that attempt to connect without using the RSA
encryption fail.

When a connection is refused because network password encryption is

required, the client receives:
Msg 1640, Level 16, State 2:
Adaptive Server requires encryption of the login
password on the network.

number of alarms
Summary information
Default value
Range of values

40
40 2147483647

Status
Display level

Dynamic
Comprehensive

Required role
Configuration groups

System administrator
Memory Use, SQL Server Administration

number of alarms specifies the number of alarm structures allocated by

Adaptive Server.
The Transact-SQL command waitfor defines a specific time, time interval, or
event for the execution of a statement block, stored procedure, or transaction.
Adaptive Server uses alarms to correctly execute waitfor commands. Other
internal processes require alarms.
When Adaptive Server needs more alarms than are currently allocated, this
message is written to the error log:
uasetalarm: no more alarms available

System Administration Guide: Volume 1

191

Configuration parameters

The number of bytes of memory required for each alarm structure is small. If
you raise the number of alarms value significantly, adjust max memory
accordingly.

number of aux scan descriptors

Summary information
Default value
Range of values

200
02147483647

Status
Display level

Dynamic
Comprehensive

Required role

System administrator

Configuration groups

Memory Use, SQL Server Administration

number of aux scan descriptors sets the number of auxiliary scan descriptors
available in a pool shared by all users on a server.

Each user connection and each worker process has 48 scan descriptors
exclusively allocated to it. Of these, 16 are reserved for user tables, 12 are
reserved for worktables, and 20 are reserved for system tables (with 4 of these
set aside for rollback conditions). A descriptor is needed for each table
referenced, directly or indirectly, by a query. For user tables, a table reference
includes:

All tables referenced in the from clause of the query

All tables referenced in a view named in the query (the view itself is not
counted)

All tables referenced in a subquery

All tables that need to be checked for referential integrity (these are used
only for inserts, updates, and deletes)

A table created with select...into

All worktables created for the query

If a table is referenced more than once (for example, in a self-join, in more than
one view, or in more than one subquery) the table is counted each time. If the
query includes a union, each select statement in the union query is a separate
scan. If a query runs in parallel, the coordinating process and each worker
process needs a scan descriptor for each table reference.

192

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

When the number of user tables referenced by a query scan exceeds 16, or the
number of worktables exceeds 12, scan descriptors from the shared pool are
allocated. Data-only-locked tables also require a system table descriptor for
each data-only-locked table accessed with a table scan (but not those accessed
with an index scan). If more than 16 data-only-locked tables are scanned using
table scans in a query, auxiliary scan descriptors are allocated for them.
If a scan needs auxiliary scan descriptors after it has used its allotted number,
and there are no descriptors available in the shared pool, Adaptive Server
displays an error message and rolls back the user transaction.
If none of your queries need additional scan descriptors, you may still want to
leave number of aux scan descriptors set to the default value in case your system
requirements grow. Set it to 0 only if you are sure that users on your system will
never run queries on more than 16 tables and that your tables will always have
few or no referential integrity constraints. See Monitoring scan descriptor
usage on page 193.
If your queries need more scan descriptors, use one of these methods to remedy
the problem:

Rewrite the query, or break it into steps using temporary tables. For dataonly-locked tables, consider adding indexes if there are many table scans.

Redesign the tables schema so that it uses fewer scan descriptors, if it uses
a large number of referential integrity constraints. You can find how many
scan descriptors a query would use by enabling set showplan, noexec on
before running the query.

Increase the number of aux scan descriptors setting.

The following sections describe how to use sp_monitorconfig to monitor the

current and high-water-mark usage to avoid running out of descriptors, and
how to estimate the number of scan descriptors you need.
Monitoring scan descriptor usage
sp_monitorconfig reports the number of unused (free) scan descriptors, the
number of auxiliary scan descriptors currently being used, the percentage that
is active, and the maximum number of scan descriptors used since the server
was last started. Run it periodically, at peak periods, to monitor scan descriptor
use.

This example shows scan descriptor use with 500 descriptors configured:
sp_monitorconfig "aux scan descriptors"
Usage information at date and time: Apr 22 2002

System Administration Guide: Volume 1

2:49PM.

193

Configuration parameters

Name
Num_free
Reuse_cnt
Instance_Name
------------------------------------------ -------------------number of aux
260
0
NULL

Num_active

Pct_act

Max_Used

----------

-------

-----------

240

48.00

427

Only 240 auxiliary scan descriptors are being used, leaving 260 free. However,
the maximum number of scan descriptors used at any one time since the last
time Adaptive Server was started is 427, leaving about 20 percent for growth
in use and exceptionally heavy use periods. Re-used does not apply to scan
descriptors.
Estimating and configuring auxiliary scan descriptors

To get an estimate of scan descriptor use:

1

Determine the number of table references for any query that references
more than 16 user tables, or for those that have a large number of
referential constraints, by running the query with set showplan and set
noexec enabled. If auxiliary scan descriptors are required, showplan
reports the number needed:
Auxiliary scan descriptors required: 17

The reported number includes all auxiliary scan descriptors that are
required for the query, including those for all worker processes. If your
queries involve only referential constraints, you can also use
sp_helpconstraint, which displays a count of the number of referential
constraints per table.
2

For each query that uses auxiliary scan descriptors, estimate the number of
users who would run the query simultaneously and multiply. If 10 users are
expected to run a query that requires 8 auxiliary descriptors, a total of 80
will be needed at any one time.

Add the per-query results to calculate the number of needed auxiliary scan
descriptors.

number of backup connections

194

Summary information
Default value

Range of values

1 32768

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

Summary information
Status
Display level

Dynamic
Basic

Required role
Configuration group

System administrator
User Environment

number of backup connections sets the maximum number of user connections

Backup Server establishes to dump or load in-memory databases. The value of
number of backup connections restricts the maximum number of stripes for an
archived database because Backup Server requires one user connection per
stripe when you run dump or load database, and requires an extra connection
to run the dump database command.
number of backup connections is a limit, and does not consume any resources.
Setting number of backup connections to 0 means that Backup Server can use
the maximum number of user connections.

number of ccbs
Summary information
Default value
Range of values

0
0 100

Status
Display level

Static

Required role
Configuration group

Diagnostics

Reserved for future use.

number of checkpoint tasks

Summary information
Default value
Valid values

1
1 8

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
Backup/Recovery

System Administration Guide: Volume 1

195

Configuration parameters

number of checkpoint tasks configures parallel checkpoints. The value of

number of checkpoint tasks must be less than or equal to the value of number
of engines at start-up. The maximum value is limited by the value of the
configuration parameters number of engines online at startup and number of
open databases, with a maximum of 8.

The default value sets serial checkpoints as the default behavior.

number of devices
Summary information
Default value
Range of values

10
12,147,483,647

Status

Dynamic

Display level
Required role

Basic
System administrator

Configuration groups

Disk I/O, Memory Use

number of devices controls the number of database devices Adaptive Server can
use. It does not include devices used for database or transaction log dumps.

When you execute disk init, you can also assign the virtual device number (the
vdevno), although this value is optional. If you do not assign the vdevno,
Adaptive Server assigns the next available virtual device number.
If you do assign the virtual device number, each device number must be unique
among the device numbers used by Adaptive Server. The number 0 is reserved
for the master device. You can enter any unused device number that falls in the
valid range of values.
To determine which numbers are currently in use, enter:
select vdevno from master..sysdevices
where status & 2 = 2

Here, status 2 specifies physical disk.

Setting the number of devices on UNIX

If you are using a large number of devices on UNIX platforms, Sybase

recommends that you set the appropriate number of devices and user
connections in the configuration file. Attempting to configure a large number
of devices dynamically using sp_configure may fail.

196

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

The number of network connections and number of devices configuration

parameters use operating system file descriptors. You must increase the
number of operating system file descriptors to use a large value for either
configuration parameter.
What constitutes a large number of devices is platform-specific. These are the
upper limits for the number of file descriptors for Adaptive Server:

HPIA 64, HPPA 64, AIX 64, Windows Set by Adaptive Server.

IBM RISC 64-bit 2000

IBM PLinux 2048

AMD64 Linux 10000

IA32-bit Linux 10000

IA 64-bit Linux 1024

Note Changing the number of file descriptors requires you restart Adaptive

Server for the change to take affect.

If there are insufficient number of file descriptors, Adaptive Server issues an
error message similar to:
Msg 5893, Level 16, State 1:Procedure 'sp_configure', Line 1234:
The sum, ( ((number of user connections) + (number of remote sites) + (max
cisremote connections) + (number of java sockets)) / (max online engines)) +
(number of devices ) + (max number network listeners),
must be no greater than '958'.
Msg 5849, Level 16, State 1:Procedure 'sp_configure', Line 1234:
Verification failed for parameter 'number of devices'.
(return status = 1)

If you then set the number of file descriptors to 2048 with the command
ulimit -n 2048 and restart the server, Adaptive Server reports this message
in the error log when you increase the value for the number of devices:
kernel Using 2048 file descriptors.

number of disk tasks

Summary information
Default value

System Administration Guide: Volume 1

197

Configuration parameters

Summary information
Range of values

UNIX: 1 64
Windows: 1

Status
Display level

Dynamic
Basic

Required role
Configuration groups

System administrator
Disk I/O, Processors

number of disk tasks controls the number of tasks dedicated to polling and
completing disk I/Os.

number of dtx participants

Summary information
Default value
Valid values

500
100 2147483647

Status
Display level

Dynamic
10

Required role
Configuration groups

System administrator
DTM Administration, Memory Use

number of dtx participants sets the total number of remote transactions that the

Adaptive Server transaction coordination service can propagate and coordinate

simultaneously. A DTX participant is an internal memory structure that the
coordination service uses to manage a remote transaction branch. As
transactions are propagated to remote servers, the coordination service must
obtain new DTX participants to manage those branches.
Setting number of dtx participants to a number smaller than the default reduces
the number of remote transactions that the server can manage. If no DTX
participants are available, new distributed transactions cannot start. In-progress
distributed transactions may abort if no DTX participants are available to
propagate a new remote transaction.
Setting number of dtx participants to a number larger than the default increases
the number of remote transaction branches that Adaptive Server can handle,
but also consumes more memory.

198

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

Optimizing the number of DTX participants for your system

During a peak period, use sp_monitorconfig to examine the use of DTX

participants:
sp_monitorconfig "number of dtx participants"
Usage information at date and time: Apr 22 2002 2:49PM.
Name
Num_free
Num_active Pct_act
Reuse_cnt
Instance_Name
--------------------------------- ---------- ---------------- -------------------number of dtx
80
20
4.00
0
NULL

Max_Used
----------210

If the num_free value is zero or very low, new distributed transactions may be
unable to start due to a lack of DTX participants. Consider increasing the
number of dtx participants value.
A low Max_used value may indicate that unused DTX participants are
consuming memory that could be used by other server functions. Consider
reducing the value of number of dtx participants.

number of dump threads

Summary information
Default value

Disabled

Range of values
Status

1 (disabled, no parallelism) 8 (fully parallel)

Dynamic

Display level
Required role

Intermediate
System administrator

Configuration group

Group Diagnostics

number of dump threads controls the number of threads that Adaptive Server
spawns to perform a memory dump. Using the appropriate value for number of
dump threads can reduce the amount of time the engines are halted during the
memory dump.

When you are determining the number of threads for memory:

Use a value of 8 if the machine has enough free memory for the file system
cache to hold the entire memory dump.

System Administration Guide: Volume 1

199

Configuration parameters

If you do not know whether the machine has enough free memory, the
value for number of dump threads depends on many factors, including the
speed of the I/O system, the speed of the disks, the controllers cache,
whether the dump file lives in a logical volume manager created on several
disks, and so on.

Disable parallel processing (by assigning a value of 1) if you do not halt

the engines when performing memory dumps, described below.
When Adaptive Server performs a memory dump, the number of files it
creates is the sum of the number of memory segments that it has allocated
multiplied by the number of threads configured. Adaptive Server uses
separate threads to write on separate files. When this job completes, the
engines are restarted, and the files are merged into the target dump file.
Because of this, the time to dump the shared memory in parallel is greater
than doing it serially.

If you halt the engines during the memory dump, using a value other than
1 may reduce the amount of time the engines spend stopped while
dumping the memory.

number of engines at startup

Summary information
Default value

Range of values
Status

1 number of CPUs on machine

Static

Display level
Required role

Basic
System administrator

Configuration groups

Java Services, Memory Use, Processors

Note When configured for threaded mode, Adaptive Server ignores the
number of engines at startup configuration parameter. In threaded mode,

Adaptive Server uses the size of the defined thread pools to determine the
number of online engines at startup. If the configuration file contains no thread
pool configuration information (for example, while you are upgrading from an
earlier version), Adaptive Server uses the existing value for number of engines
at startup to determine the size the default engine pool.

200

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

number of engines at startup is used exclusively during start-up to set the

number of engines brought online. It allows great flexibility, subject to the

restriction that you cannot set the value of number of engines at startup to a
value greater than the number of CPUs on your machine, or to a value greater
than the configuration of max online engines. Users who do not intend to bring
engines online after start-up should set max online engines and number of
engines at startup to the same value. A difference between number of engines at
startup and max online engines wastes approximately 1.8 MB of memory per
engine.
Adaptive Server allows users to take all engines offline, except engine zero.

number of histogram steps

Summary information
Default value

20

Range of values
Status

3 2147483647
Dynamic

Display level
Required role

Comprehensive
System administrator

Configuration group

Query Tuning

number of histogram steps specifies the number of steps in a histogram.

number of index trips

Summary information
Default value

Range of values
Status

065535
Dynamic

Display level
Required role

Comprehensive
System administrator

Configuration group

Cache Manager

number of index trips specifies the number of times an aged index page traverses
the most recently used/least recently used (MRU/LRU) chain before it is
considered for swapping out. As you increase the value of number of index trips,
index pages stay in cache for longer periods of time.

System Administration Guide: Volume 1

201

Configuration parameters

A data cache is implemented as an MRU/LRU chain. As the user threads

access data and index pages, these pages are placed on the MRU end of the
caches MRU/LRU chain. In some high transaction environments (and in some
benchmarks), you may want to keep index pages in cache, since they will
probably be needed again soon. Setting number of index trips higher keeps index
pages in cache longer; setting it lower allows index pages to be swapped out of
cache sooner.
You need not set the number of index trips for relaxed LRU pages. See Chapter
4, Configuring Data Caches, in the System Administration Guide: Volume 2.
Note If the cache used by an index is relatively small (especially if it shares
space with other objects) and you have a high transaction volume, do not set
number of index trips too high. The cache can flood with pages that do not age
out, and this may lead to the timing out of processes that are waiting for cache
space.

Before changing the value of number of index trips to a number other than 0,
make sure the application has sufficient cache to store all index, OAM, and
data pages. Consult Sybase Technical Support before changing the value of
number of index trips.

number of java sockets

Summary information
Default value
Valid values

0
0 32767

Status
Display level

Dynamic
Comprehensive

Required role
Configuration groups

System administrator
Java Services, Memory Use

number of java sockets enables the Java VM and the java.net classes Sybase

supports.

number of large i/o buffers

Summary information
Default value

202

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

Summary information
Valid values
Status

1256
Dynamic

Display level
Required role

Comprehensive
System administrator

Configuration groups

Disk I/O, Memory Use, SQL Server Administration

number of large i/o buffers sets the number of allocation unit-sized buffers

reserved for performing large I/O for certain Adaptive Server utilities. These
large I/O buffers are used primarily by the load database command, which uses
one buffer to load the database, regardless of the number of stripes it specifies.
load database then uses as many as 32 buffers to clear the pages for the
database it is loading. These buffers are not used by load transaction. To
perform more than six load database commands concurrently, configure one
large I/O buffer for each load database command.
create database and alter database use these buffers for large I/O while clearing
database pages. Each instance of create database or load database can use as

many as 32 large I/O buffers.

These buffers are also used by disk mirroring and by some dbcc commands.
Note In Adaptive Server version 12.5.0.3 and later, the size of the large I/O

buffers is one allocation (256 pages), not one extent (8 pages). The server thus
requires more memory allocation for large buffers. For example, a disk buffer
that required memory for 8 pages in earlier versions now requires memory for
256 pages.

number of locks
Summary information
Default value
Range of values

5000
10002147483647

Status
Display level

Dynamic
Basic

Required role
Configuration groups

System administrator
Lock Manager, Memory Use

number of locks sets the total number of available locks for all users on
Adaptive Server.

System Administration Guide: Volume 1

203

Configuration parameters

The total number of locks needed by Adaptive Server depends on the number
of concurrent and parallel processes, and the types of actions performed by the
transactions. To see how many locks are in use at a particular time, use sp_lock.
For serial operation, Sybase suggests that you start by assigning 20 locks for
each active, concurrent connection.
Parallel execution requires more locks than serial execution. For example, if
you find that queries use an average of five worker processes, try increasing by
one-third the number of locks configured for serial operation.
If the system runs out of locks, Adaptive Server displays a server-level error
message. If users report lock errors, you may need to increase number of locks;
but remember that locks use memory. See Chapter 3, Configuring Memory,
in the System Administration Guide Volume 2.
Note Datarows locking may require that you change the value for number of
locks. See the Performance and Tuning Series: Locking and Concurrency

Control.

number of mailboxes
Summary information
Default value
Range of values

30
302147483647

Status
Display level

Dynamic
Comprehensive

Required role
Configuration groups

System administrator
Memory Use, SQL Server Administration

number of mailboxes specifies the number of mailbox structures allocated by

Adaptive Server. Mailboxes, which are used with messages, are used internally
by Adaptive Server for communication and synchronization between kernel
service processes. Mailboxes are not used by user processes. Do not modify
this parameter unless instructed to do so by Sybase Technical Support.

number of messages
Summary information
Default value

204

64

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

Summary information
Range of values
Status

02147483647
Dynamic

Display level
Required role

Comprehensive
System administrator

Configuration groups

Memory Use, SQL Server Administration

number of messages specifies the number of message structures allocated by

Adaptive Server. Messages, which are used with mailboxes, are used internally
by Adaptive Server for communication and synchronization between kernel
service processes. Messages are also used to coordinate between a family of
processes in parallel processing. Do not modify this parameter unless
instructed to do so by Sybase Technical Support.

number of network tasks

Summary information
Default value

Range of values

UNIX: 1 64

Status
Display level

Dynamic
Basic

Required role
Configuration groups

System administrator
Disk I/O, Processors

Windows: 1

number of network tasks controls the number of tasks dedicated to polling and

completing network I/Os.

number of oam trips

Summary information
Default value
Range of values

0
065535

Status
Display level

Dynamic
Comprehensive

Required role

System administrator

System Administration Guide: Volume 1

205

Configuration parameters

number of oam trips specifies the number of times an object allocation map
(OAM) page traverses the MRU/LRU chain before it is considered for
swapping out. The higher the value of number of oam trips, the longer aged
OAM pages stay in cache.

Each table, and each index on a table, has an OAM page, which holds
information on pages allocated to the table or index and is checked when a new
page is needed for the index or table. (See page utilization percent on page
227. ) A single OAM page can hold allocation mapping for between 2,000 and
63,750 data or index pages.
The OAM pages point to the allocation page for each allocation unit where the
object uses space. The allocation pages, in turn, track the information about
extent and page usage within the allocation unit.
In some environments and benchmarks that involve significant allocations of
space (that is, massive bulk copy operations), keeping OAM pages in cache
longer improves performance. Setting number of oam trips to a higher value
keeps OAM pages in cache.
Note If the cache is relatively small and used by a large number of objects, do
not set number of oam trips too high. This may result in the cache being flooded
with OAM pages that do not age out, and user threads may begin to time out.

Before changing the value of number of oam trips to a number other than 0,
make sure the application has sufficient cache to store all index, OAM, and
data pages. Consult Sybase Technical Support before changing the value of
number of oam trips.

number of open databases

Summary information
Default value
Range of values

12
6 2147483647

Status
Display level

Dynamic
Basic

Required role
Configuration groups

System administrator
Memory Use, Meta-Data Caches, SQL Server
Administration

number of open databases sets the maximum number of databases that can be
open simultaneously on Adaptive Server.

206

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

When you calculate a value, include the system databases master, model,
sybsystemprocs, and tempdb. If you have installed auditing, include the
sybsecurity database. Also, count the sample databases pubs2 and pubs3, the
syntax database sybsyntax, and the dbcc database dbccdb if they are installed.
If you are planning to make a substantial change, such as loading a large
database from another server, use sp_helpconfig to calculate an estimated
metadata cache size by using sp_helpconfig. sp_helpconfig displays the amount
of memory required for a given number of metadata descriptors, as well as the
number of descriptors that can be accommodated by a given amount of
memory. A database metadata descriptor represents the state of the database
while it is in use or cached between uses.

Optimizing the number of open databases

If Adaptive Server displays a message saying that you have exceeded the
allowable number of open databases, adjust the value.
1

Use sp_countmetadata to find the total number of database metadata

descriptors:
sp_countmetadata "open databases"

The best time to run sp_countmetadata is when there is little activity on the
server. Running sp_countmetadata during a peak time can cause
contention with other processes.
Suppose Adaptive Server reports the following information:
There are 50 databases, requiring 1719 Kbytes of
memory. The 'open databases' configuration parameter
is currently set to 500.

Configure number of open databases with the value of 50:

sp_configure "number of open databases", 50

This new configuration number is only a starting point; base the ideal size
on the number of active metadata database cache descriptors, not the total
number of databases.
3

During a peak period, find the number of active metadata descriptors:

sp_monitorconfig "open databases"

Usage information at date and time: Apr 22 2002 2:49PM.
Name
Num_free
Num_active Pct_act
Reuse_cnt
Instance_Name
--------------------------------- ---------- ---------------- --------------------

System Administration Guide: Volume 1

Max_Used
-----------

207

Configuration parameters

number of open
0

50

20

.00

26

NULL

In this example, 20 metadata database descriptors are active; the

maximum number of descriptors that have been active since the server was
last started is 26.
See sp_monitorconfig in the Reference Manual: Procedures for more
information.
4

Configure number of open databases to 26, plus additional space for 10

percent more (about 3), for a total of 29:
sp_configure "number of open databases", 29

If there is a lot of activity on the server, for example, if databases are being
added or dropped, periodically run sp_monitorconfig. Reset the cache size
as the number of active descriptors changes.

number of open indexes

Summary information
Default value
Range of values

500
1002147483647

Status
Display level

Dynamic
Basic

Required role
Configuration groups

System administrator
Memory Use, Meta-Data Caches

number of open indexes sets the maximum number of indexes that can be used

simultaneously on Adaptive Server.

If you are planning to make a substantial change, such as loading databases
with a large number of indexes from another server, use sp_helpconfig to
calculate an estimated metadata cache size. sp_helpconfig displays the amount
of memory required for a given number of metadata descriptors, as well as the
number of descriptors that can be accommodated by a given amount of
memory. An index metadata descriptor represents the state of an index while it
is in use or cached between uses.

208

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

Optimizing number of open indexes

If the default value of number of open indexes is insufficient, Adaptive Server

displays a message after trying to reuse active index descriptors, and you must
adjust this value.
1

Use sp_countmetadata to find the total number of index metadata

descriptors:
sp_countmetadata "open indexes"

The best time to run sp_countmetadata is when there is little activity in the
server. Running sp_countmetadata during a peak time can cause
contention with other processes.
Suppose Adaptive Server reports the following information:
There are 698 user indexes in all database(s),
requiring 286.289 Kbytes of memory. The 'open
indexes' configuration parameter is currently set to
500.

Configure the number of open indexes parameter to 698:

sp_configure "number of open indexes", 698

This new configuration is only a starting point; base the ideal size on the
number of active index metadata cache descriptors, not the total number
of indexes.
3

During a peak period, find the number of active index metadata

descriptors:

sp_monitorconfig "open indexes"

Usage information at date and time: Apr 22 2002 2:49PM.
Name
Num_free
Num_active Pct_act
Reuse_cnt
Instance_Name
--------------------------------- ---------- ---------------- -------------------number of open
182
516
73.92
0
NULL

Max_Used
----------590

In this example, 590 is the maximum number of index descriptors that

have been used since the server was last started.
See sp_monitorconfig in the Reference Manual: Procedures.
4

Configure the number of open indexes configuration parameter to 590, plus

additional space for 10 percent more (59), for a total of 649:

System Administration Guide: Volume 1

209

Configuration parameters

sp_configure "number of open indexes", 649

If there is a lot of activity on the server, for example, if tables are being
added or dropped, periodically run sp_monitorconfig. Reset the cache size
as the number of active descriptors changes.

number of open objects

Summary information
Default value
Range of values

500
1002147483647

Status
Display level

Dynamic
Basic

Required role
Configuration groups

System administrator
Memory Use, Meta-Data Caches, SQL Server
Administration

number of open objects sets the maximum number of objects that can be open

simultaneously on Adaptive Server.

If you are planning to make a substantial change, such as loading databases
with a large number of objects from another server, use sp_helpconfig to
recalculate an estimated metadata cache size. sp_helpconfig displays the
amount of memory required for a given number of metadata descriptors, as
well as the number of descriptors that can be accommodated by a given amount
of memory. An object metadata descriptor represents the state of an object
while it is in use, or cached between uses.

Optimizing number of open objects

If the default number of open objects is insufficient, Adaptive Server displays a

message after trying to reuse active object descriptors.
1

Use sp_countmetadata to find the total number of object metadata cache

descriptors:
sp_countmetadata "open objects"

The best time to run sp_countmetadata is when there is little activity in the
server. Running sp_countmetadata during a peak time can cause
contention with other processes.
Suppose Adaptive Server reports this information:
There are 696 user objects in all database(s),
requiring 1166 Kbytes of memory. The 'open objects'

210

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

configuration parameter is currently set to 500.

Configure number of open objects to account for the number of open

objects:
sp_configure "number of open objects", 1407

1407 covers the 1340 user objects, plus 5 percent to accommodate

temporary tables.
This new configuration is only a starting point; base the ideal number on
the active object metadata cache descriptors, not the total number of
objects.
3

During a peak period, find the number of active metadata cache

descriptors:

sp_monitorconfig "open objects"

Usage information at date and time: Aug 20 2007 1:32PM..
Name
Num_free Num_active Pct_act Max_Used Reuse_cnt
Instance_Name
--------------------- -------- ---------- ------- -------- ---------------------------number of open objects
560
847
71.40
1497
0
NULL

In this example, 1497 is the maximum number of object descriptors that

have been used since the server was last started.
4

Configure the number of open objects to 1397, plus 10 percent (140), for a
total of 1537:
sp_configure "number of open objects", 1537

If there is a lot of activity on the server, for example, if tables are being added
or dropped, periodically run sp_monitorconfig. Reset the cache size as the
number of active descriptors changes. See sp_monitorconfig in the Reference
Manual: Procedures.

number of open partitions

Summary information
Default value

500

Range of values
Status

100 2147483647
Dynamic

Display level
Required role

Basic
System administrator

System Administration Guide: Volume 1

211

Configuration parameters

Summary information
Configuration groups

Memory Use, Meta-Data Caches

Specifies the number of partitions that Adaptive Server can access at one time.
Optimizing the
number of open
partitions parameter
for your system

If the default value of number of open partitions is insufficient, Adaptive Server

displays a message after trying to reuse active partition descriptors. You must
adjust this value.
1

Use sp_countmetadata to find the total number of open partitions. For

example:
sp_countmetadata "open partitions"

The best time to run sp_countmetadata is when there is little activity in the
server. Running sp_countmetadata during a peak time can cause
contention with other processes.
Suppose Adaptive Server reports the following information:
There are 42 user partitions in all database(s),
requiring 109 Kbytes of memory. The 'open
partitions' configuration parameter is currently set
to 110.

Configure number of open partitions to 110, as reported by

sp_countmetadata:
sp_configure "number of open partitions", 110

During a peak period, find the number of active metadata cache

descriptors, for example:

sp_monitorconfig "open partitions"

Usage information at date and time: Jun 30 2008
Name
Num_free
Reuse_cnt
Instance_Name
------------------------------------------ -------------------number of open partitions
27
0
NULL

3:15PM.

Num_active

Pct_act

Max_Used

----------

-------

-----------

57

51.8

83

In this example, 83 is the maximum number of partition descriptors that

have been used since the server was last started.
4

Configure the number of open partitions to 83, plus 10 percent (8), for a
total of 91:
sp_configure "number of open partitions", 91

212

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

If there is a lot of activity on the server, for example, if tables are being added
or dropped, periodically run sp_monitorconfig. Reset the cache size as the
number of active descriptors changes. See sp_monitorconfig in the Reference
Manual: Procedures.

number of pre-allocated extents

Summary information
Default value
Range of values

2
132

Status
Display level

Dynamic
Comprehensive

Required role

System administrator

Configuration group

SQL Server Administration

number of pre-allocated extents specifies the number of extents (eight pages)

allocated in a single trip to the page manager. Currently, this parameter is used
only by bcp to improve performance when copying in large amounts of data.
By default, bcp allocates two extents at a time and writes an allocation record
to the log each time.

Setting number of pre-allocated extents means that bcp allocates the specified
number of extents each time it requires more space, and writes a single log
record for the event.
An object may be allocated more pages than actually needed, so the value of
number of pre-allocated extents should be low if you are using bcp for small
batches. If you are using bcp for large batches, increase the value of number of
pre-allocated extents to reduce the amount of overhead required to allocate
pages and to reduce the number of log records.

System Administration Guide: Volume 1

213

Configuration parameters

Using a value of 32 for the number of pre-allocated extents

Using a value of 32 for number of pre-allocated extents has a special

significance for configuration and impacts the space allocations Adaptive
Server performs internally. If you set number of pre-allocated extents to 32,
Adaptive Server reserves an entire allocation unit worth of extents for utility
operations like bcp-in and select into, both of which use the large-scale
allocation scheme of space reservation. This greatly improves the performance
of these utilities, particularly when you run them concurrently on multiple
nodes. Consequently, using a value of 32 guarantees that each node of a cluster
is able to work independently on its own allocation unit without interference
from the other nodes.
In earlier versions of Adaptive Server, the number of pre-allocated extents
parameter specified the number of extents reserved in a single allocation call
for tables of all sizes.
With this version of Adaptive Server, the value of number of pre-allocated
extents is ignored for large tables with 240 or more pages for these commands
only:

alter table table_name add column_name . . .

alter table table_name modify column_name . . .

alter table table_name drop column_name . . .

alter table lock . . .

reorg rebuild

When you run these command on tables larger than 240 pages, Adaptive Server
reserves an entire allocation unit (32 extents), which greatly improves
performance, particularly when you run them concurrently on multiple nodes.
The value of number of pre-allocated extents continues to be observed for the
above commands for tables with fewer than 240 pages, and for all commands
(such as select into, bcp, alter table partition) for tables of all sizes.

number of Q engines at startup

214

Summary information
Default value

Range of values
Status

0 127
Static

Display level

Comprehensive

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

Summary information
Required role
Configuration group

System administrator
Processors

number of Q engines at startup, which specifies the number of Q engines that

are online when the server starts, is required for MQ. You may need to increase
max online engines to accommodate the number of max online Q engines.

number of remote connections

Summary information
Default value

20

Range of values

532767

Status
Display level

Static
Intermediate

Required role
Configuration groups

System administrator
Memory Use, Network Communication

number of remote connections specifies the number of logical connections that

can simultaneously be open to and from an Adaptive Server. Each
simultaneous connection to XP Server for ESP execution uses up to one remote
connection each. See Chapter 7, Managing Remote Servers.

number of remote logins

Summary information
Default value
Range of values

20
032767

Status
Display level

Static
Intermediate

Required role
Configuration groups

System administrator
Memory Use, Network Communication

number of remote logins controls the number of active user connections from
Adaptive Server to remote servers. Each simultaneous connection to XP Server
for ESP execution uses up to one remote login each. Set this parameter to the
same (or a lower) value as number of remote connections. See Chapter 7,
Managing Remote Servers.

System Administration Guide: Volume 1

215

Configuration parameters

number of remote sites

Summary information
Default value
Range of values

10
032767

Status
Display level

Static
Intermediate

Required role
Configuration groups

System administrator
Memory Use, Network Communication

number of remote sites determines the maximum number of remote sites that

can simultaneously access Adaptive Server. Each Adaptive Server-to-XP

Server connection uses one remote site connection.
Internally, number of remote sites determines the number of site handlers that
can be active at any one time; all server accesses from a single site are managed
with a single site handler. For example, if you set number of remote sites to 5,
and each site initiates three remote procedure calls, sp_who shows 5 site
handler processes for the 15 processes. See Chapter 7, Managing Remote
Servers.

number of sort buffers

Summary information
Default value
Range of values

500
032767

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
SQL Server Administration

number of sort buffers specifies the amount of memory allocated for buffers
used to hold pages read from input tables and perform index merges during
sorts. number of sort buffers is used only for parallel sorting.

Adaptive Server allocates sort buffers from the cache to which the table is
bound. If the cache is not bound to a table, Adaptive Server allocates the sort
buffers from the cache to which the database is bound. If the cache is not bound
to a table or a database, Adaptive Server allocates sort buffers from the default
data cache.
Parallel sorts are used when you:

216

Adaptive Server Enterprise

CHAPTER 5

Run updates statistics

Create indexes

Setting Configuration Parameters

See Chapter 10, Using Statistics to Improve Performance, in the

Performance and Tuning Series: Query Processing and Abstract Plans.
The value you use for number of sort buffers depends on the page size of the
server.
Sybase recommends that you leave this parameter set to the default except
when you are creating indexes in parallel.
Setting the value too high can rob nonsorting processes of access to the buffer
pool in caches being used to perform sorts.
If you configure a high number of sort buffers, a sort on a large table may
require more procedure cache. The effect is more pronounced with tables that
have smaller row sizes, because the number of rows per page is higher.
This equation estimates the amount of procedure cache required (in bytes):
(Number of sort buffers) X (rows per page) X 100
If you do not configure enough procedure cache for the number of sort buffers,
the sort may fail with error message 701. If this occurs, reconfigure Adaptive
Server with a lower number of sort buffers and retry the sort.
See max buffers per lava operator on page 182 for information about setting
an upper limit for the number of buffers used by an operator.

number of user connections

Summary information
Default value

25

Range of values
Status

52147483647
Dynamic

Display level
Required role

Basic
System administrator

Configuration groups

Memory Use, User Environment

number of user connections sets the maximum number of user connections that
can simultaneously be connected to Adaptive Server. It does not refer to the
maximum number of processes; that number depends not only on the value of
this parameter but also on other system activity.

System Administration Guide: Volume 1

217

Configuration parameters

Upper limit to the

maximum number of
user connections

The maximum allowable number of file descriptors per process is operatingsystem-dependent; see the configuration documentation for your platform.
The number of file descriptors available for Adaptive Server connections is
stored in the global variable @@max_connections. You can report the
maximum number of file descriptors your system can use with:
select @@max_connections

The return value represents the maximum number of file descriptors allowed
by the system for your processes, minus overhead. Overhead increases with the
number of engines. For more information on how multiprocessing affects the
number file descriptors available for Adaptive Server connections, see Chapter
5, Managing Mulitprocessor Servers, in System Administration Guide:
Volume 2.
In addition, you must reserve a number of connections for the following items,
which you also set with configuration parameters:

The database devices, including mirror devices

Site handlers

Network listeners

The number of user connections + (number of devices * max online engines * 2)

+ number of remote sites + max number network listeners cannot be greater than
the value of @@max_connections.
Reserved connections

One connection from the configured number of connections is reserved for

temporary administrative tasks to make sure that database administrators can
connect to Adaptive Server. A reserved connection has a total login time of 15
minutes, and can be is allocated only to a user who has the sa_role. Adaptive
Server terminates the connection after 15 minutes to ensure the availability of
the reserved connection at an installation with multiple database
administrators.
Adaptive Server also automatically uses this reserved connection when a client
uses the last resource for connecting to Adaptive Server.
If Adaptive Server is using a reserved connection, the following informational
message appears when the user logs in to Adaptive Server:

There are not enough user connections available; you are being connected
using a temporary administrative connection which will time out after '15'
minutes. Increase the value of th 'number of user connections' parameter

Adaptive Server also prints a message similar to the following to the error log
when the final connection to Adaptive Server terminates due to a timeout:

218

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

00:00000:00008:2003/03/14 11:25:31.36 server Process '16' has been

terminated as it exceeded the maximum login time allowed for such processes.
This process used a connection reserved for system administrators and has a
maximum login period of '15' minutes
Optimizing max
number of user
connections

There is no formula to determine how many connections to allow for each user.
You must estimate this number, based on the system and user requirements.
You must also take into account that on a system with many users, connections
needed only occasionally or transiently can generally be shared among users.
The following processes require user connections:

One connection is needed for each user running isql.

Application developers use one connection for each editing session.

The number of connections required by users running an application

depends on how the application has been programmed. Users executing
Open Client programs need one connection for each open DB-Library
dbprocess or Client-Library cs_connection.
Note Sybase suggests that you estimate the maximum number of
connections used by Adaptive Server and update number of user
connections as you add physical devices or users to the system.
Periodically use sp_who to determine the number of active user

connections on your Adaptive Server.

Certain other configuration parameters, including stack size and default network
packet size, affect the amount of memory for each user connection.
User connections for
shared memoryEJB
Server

Adaptive Server uses the value of number of user connections to establish the
number of shared-memory connections for EJB Server. Thus, if number of user
connections is 30, Adaptive Server establishes 10 shared-memory connections
for EJB Server. Shared-memory connections are not a subset of user
connections, and are not subtracted from the number of user connections.
To increase the number of user connections for shared memory, you must:
1

Increase number of user connections to a number one-third of which is the

number of desired shared-memory connections.

Restart Adaptive Server.

Although number of user connections is a dynamic configuration parameter,

you must restart the server to change the number of user connections for shared
memory. See the EJB Server Users Guide.

System Administration Guide: Volume 1

219

Configuration parameters

With Adaptive Server version 12.5.3 ESD #2, no sockets are automatically
reserved for EJB. However, you can enable trace flag 1642 to revert to the
functionality of earlier version, reserving one-third of the sockets for EJB.
Enable traceflag 1642 to set up the EJB server. For this version of Adaptive
Server, you can ignore this message, "hbc_ninit: No sockets
available for HBC", in the error log if the EJB server is not configured.
In Adaptive Server version 12.5.3 and later, if the EJB server is enabled and
HBC sockets are not available, "hbc_ninit: No sockets available
for HBC" is reported. If traceflag 1642 is not enabled, set the flag, and restart
Adaptive Server. If the EJB server is not enabled, then no message is reported
and Adaptive Server automatically disables the sockets reserved for EJB
server.

number of worker processes

Summary information
Default value

Range of values
Status

02147483647
Dynamic

Display level
Required role

Basic
System administrator

Configuration groups

Memory Use, Query Tuning

number of worker processes specifies the maximum number of worker

processes that Adaptive Server can use at any one time for all simultaneously
running parallel queries.
Adaptive Server issues a warning message at start-up if there is insufficient
memory to create the specified number of worker processes. memory per
worker process controls the memory allocated to each worker process.
If you have not configured number of worker processes for a sufficient number
of threads from the worker thread pool, Adaptive Server adjusts query plans at
runtime to use fewer worker threads. If Adaptive Server cannot adjust the
queries at runtime, the queries recompile serially. However, alter table and
execute immediate commands are aborted if they do not have sufficient worker
threads.

220

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

o/s file descriptors

Summary information
Default value
Range of values

0
Site-specific

Status
Display level

Read-only
Comprehensive

Required role
Configuration group

System administrator
O/S Resources

o/s file descriptors indicates the maximum per-process number of file

descriptors configured for your operating system. This parameter is read-only

and cannot be configured through Adaptive Server.
Many operating systems allow you to configure the number of file descriptors
available per process. See your operating system documentation.
The number of file descriptors available for Adaptive Server connections,
which is less than the value of o/s file descriptors, is stored in the variable
@@max_connections. See Upper limit to the maximum number of user
connections on page 218.

object lockwait timing

Summary information
Default value
Range of values

0 (off)
0 (off), 1 (on)

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
Monitoring

object lockwait timing controls whether Adaptive Server collects timing

statistics for requests of locks on objects.

open index hash spinlock ratio

Summary information
Default value

100

Range of values

12147483647

System Administration Guide: Volume 1

221

Configuration parameters

Summary information
Status
Display level

Dynamic
Basic

Required role
Configuration groups

System administrator
Memory Use, Meta-Data Cache

open index hash spinlock ratio sets the number of index metadata descriptor
hash tables that are protected by one spinlock. This parameter is used only in
multiprocessing systems only.

All the index descriptors belonging to a table are accessible through a hash
table. When you run a query on the table, Adaptive Server uses hash tables to
look up the necessary index information in its sysindexes rows. A hash table is
an internal mechanism used by Adaptive Server to retrieve information
quickly.
Usually, you do not need to change this parameter. In rare instances, however,
you may need to reset it if Adaptive Server demonstrates contention from hash
spinlocks. See the Performance and Tuning Series: Monitoring Adaptive
Server with sp_sysmon.
For more information about configuring spinlock ratios, see Chapter 5,
Managing Multiprocessor Servers, in the System Administration Guide:
Volume 2.

open index spinlock ratio

Summary information
Default value
Range of values

100
1214748364

Status
Display level

Dynamic
Comprehensive

Required role
Configuration groups

System administrator
Memory Use, Meta-Data Cache

open index spinlock ratio specifies the number of index metadata descriptors

that are protected by one spinlock.

Adaptive Server uses a spinlock to protect an index descriptor, since more than
one process can access the contents of the index descriptor. open index spinlock
ratio is used only in multiprocessing systems.

222

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

The value specified for this parameter defines the ratio of index descriptors per
spinlock.
If one spinlock is shared by too many index descriptors, it can cause spinlock
contention. Use sp_sysmon to get a report on spinlock contention. See the
Performance and Tuning Series: Monitoring Adaptive Server with sp_sysmon.
If sp_sysmon output indicates an index descriptor spinlock contention of more
than 3 percent, try decreasing the value of open index spinlock ratio.
See Chapter 5, Managing Mulitprocessor Servers, in System Administration
Guide: Volume 2.

open object spinlock ratio

Summary information
Default value

100

Range of values
Status

12147483647
Dynamic

Display level
Required role

Comprehensive
System administrator

Configuration group

Meta-Data Cache

open object spinlock ratio specifies the number of object descriptors that are

protected by one spinlock. Adaptive Server uses a spinlock to protect an object

descriptor, since more than one process can access the contents of the object
descriptor. open object spinlock ratio is used only in multiprocessing systems..
The default value for this parameter is 100; 1 spinlock for each 100 object
descriptors configured for your server. If your server is configured with only
one engine, Adaptive Server sets only 1 object descriptor spinlock, regardless
of the number of object descriptors.
If one spinlock is shared by too many object descriptors, it causes spinlock
contention. Use sp_sysmon to get a report on spinlock contention. See the
Performance and Tuning Series: Monitoring Adaptive Server with sp_sysmon.
If sp_sysmon output indicates an object descriptor spinlock contention of more
than 3 percent, try decreasing the value of the open object spinlock ratio
parameter.
See Chapter 5, Managing Mulitprocessor Servers, in System Administration
Guide: Volume 2.

System Administration Guide: Volume 1

223

Configuration parameters

optimization goal
Summary information
Default value
Range of values

allrows_mix

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
Query Tuning

allrows_oltp, allrows_dss

Optimization goals are a convenient way to match user query demands with the
best optimization techniques, ensuring optimal use of the optimizers time and
resources. Adaptive Server allows users to configure for two optimization
goals, which you can specify at three tiers: server level, session level, and
query level.
The server-level optimization goal is overridden at the session level, which is
overridden at the query level.
These optimization goals allow you to choose an optimization strategy that best
fits your query environment:

allrows_oltp the most useful goal for purely OLTP queries.

allrows_dss the most useful goal for operational DSS queries of

medium-to-high complexity.

optimization timeout limit

Summary information
Default value
Range of values

10
0 1000

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
Query Tuning

optimization timeout limit specifies the amount of time, as a fraction of the

estimated execution time of the query, that Adaptive Server can spend
optimizing a query.
A value of 0 indicates there is no optimization timeout.

224

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

optimizer level
Summary information
Default value
Range of values

ase_default

ase_current Enables all optimizer changes

through the current release
ase_default Disables all optimizer changes since
version 1503 ESD #1
ase1503esd2 Enables all optimizer changes
through version 15.0.3 ESD #2

Status

ase1503esd3 Enables all optimizer changes

through version 15.0.3 ESD #3
Dynamic

Display level
Required role

System administrator

Configuration group

Query Tuning

optimizer level determines the level of optimization the query processor uses.
See Chapter 7, Controlling Optimization, in the Performance and Tuning
Series: Query Processing and Abstract Plans for information about
optimization levels.

The values for optimizer level are character data, so you must use 0 as a
placeholder for the second parameter, which must be numeric.

page lock promotion HWM

Summary information
Default value
Range of values

200
22147483647

Status
Display level

Dynamic
Intermediate

Required role
Configuration groups

System administrator
Lock Manager, SQL Server Administration

page lock promotion HWM (high-water mark), with page lock promotion LWM
(low-water mark) and page lock promotion PCT (percentage), specifies the

number of page locks permitted during a single scan session of a page-locked

table or index before Adaptive Server attempts to escalate from page locks to
a table lock.

System Administration Guide: Volume 1

225

Configuration parameters

When the number of page locks acquired during a scan session exceeds page
lock promotion HWM, Adaptive Server attempts to acquire a table lock. page
lock promotion HWM value cannot be higher than number of locks.
For more detailed information on scan sessions and setting up page lock
promotion limits, see Chapter 2, Locking Configuration and Tuning, in the
Performance and Tuning Series: Locking and Concurrency Control.
The default value for page lock promotion HWM is appropriate for most
applications. To avoid table locking, you may want to increase the value. For
example, if you know that there are regular updates to 500 pages of an allpageslocked or datapages-locked table containing thousands of pages, increase
concurrency for the tables by setting page lock promotion HWM to 500.
You can also configure lock promotion of page-locked tables and views at the
object level. See sp_setrowlockpromote in the Reference Manual: Procedures.
Use sp_sysmon to see how changing page lock promotion HWM affects the
number of lock promotions. sp_sysmon reports the ratio of exclusive page to
exclusive table lock promotions and the ratio of shared page to shared table
lock promotions. See the Performance and Tuning Series: Monitoring
Adaptive Server with sp_sysmon.

page lock promotion LWM

Summary information
Default value

200

Range of values
Status

2value of page lock promotion HWM

Dynamic

Display level
Required role

Intermediate
System administrator

Configuration groups

Lock Manager, SQL Server Administration

page lock promotion LWM (low-water mark), with page lock promotion HWM
(high-water mark) and the page lock promotion PCT, specify the number of page

locks permitted during a single scan session of a page locked table or an index
before Adaptive Server attempts to promote from page locks to a table lock.
The page lock promotion LWM sets the number of page locks below which
Adaptive Server does not attempt to issue a table lock on an object. page lock
promotion LWM must be less than or equal to page lock promotion HWM.

226

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

The default value for page lock promotion LWM is sufficient for most
applications. If Adaptive Server runs out of locks (except for an isolated
incident), increase number of locks.
See the Performance and Tuning Series: Locking and Concurrency Control.
You can also configure page lock promotion at the object level. See
sp_setpglockpromote in the Reference Manual: Procedures.

page lock promotion PCT

Summary information
Default value
Range of values

100
1100

Status

Dynamic

Display level
Required role

Intermediate
System administrator

Configuration groups

Lock Manager, SQL Server Administration

If the number of locks held on an object is between page lock promotion LWM
(low-water mark) and page lock promotion HWM (high-water mark). page lock
promotion PCT sets the percentage of page locks (based on the table size) above
which Adaptive Server attempts to acquire a table lock.
See Chapter 2, Locking Configuration and Tuning, in the Performance and
Tuning Series: Locking and Concurrency Control.
The default value for page lock promotion PCT is appropriate for most
applications.
You can also configure lock promotion at the object level for page locked
objects. See sp_setpglockpromote in the Reference Manual: Procedures.

page utilization percent

Summary information
Default value

95

Range of values
Status

1100
Dynamic

Display level
Required role

Comprehensive
System administrator

Configuration group

Disk I/O

System Administration Guide: Volume 1

227

Configuration parameters

page utilization percent is used during page allocations to control whether

Adaptive Server scans a tables object allocation map (OAM) to find unused
pages or simply allocates a new extent to the table. See number of oam trips
on page 205 for more information on the OAM. The page utilization percent
parameter is a performance optimization for servers with very large tables; it
reduces the time needed to add new space.

If you set page utilization percent to 100, Adaptive Server scans through all
OAM pages to find unused pages allocated to the object before allocating a
new extent. When this parameter is set lower than 100, Adaptive Server
compares the page utilization percent setting to the ratio of used and unused
pages allocated to the table, as follows:
100 * used pages/(used pages + unused pages)

If page utilization percent is lower than the ratio, Adaptive Server allocates a
new extent instead of searching for the unused pages.
For example, when inserting data into a 10GB table that has 120 OAM pages
and only 1 unused data page:

A page utilization percent of 100 tells Adaptive Server to scan through all
120 OAM pages to locate an unused data page.

A page utilization percent of 95 allows Adaptive Server to allocate a new

extent to the object, because 95 is lower than the ratio of used pages to
used and unused pages.

A low page utilization percent value results in more unused pages. A high page
utilization percent value slows page allocations in very large tables, as Adaptive
Server performs an OAM scan to locate each unused page before allocating a
new extent. This increases logical and physical I/O.
If page allocations (especially in the case of large inserts) seem to be slow,
lower the value of page utilization percent, but reset it after inserting the data. A
lower setting affects all tables on the server and results in unused pages in all
tables.
Fast bulk copy ignores the page utilization percent setting and always allocates
new extents until there are no more extents available in the database.

partition groups
Summary information
Default value
Range of values

228

1024
12147483647

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

Summary information
Status
Display level

Dynamic
Comprehensive

Required role
Configuration groups

System administrator
Memory Use, Meta-Data Cache

partition groups specifies the maximum number of partition groups that can be
allocated by Adaptive Server. Partition groups are internal structures used by
Adaptive Server to control access to individual partitions of a table. Partition
groups are used while upgrading or during a load database upgrade to
unpartition Adaptive Server 12.5.x and earlier partitions.

The default value allows a maximum 1024 open partition groups and a
maximum of 2147483647 open partitions. The actual number of partitions may
be slightly less, due to the grouping of partitions.

partition spinlock ratio

Summary information
Default value
Range of values

10
12147483647

Status
Display level

Dynamic
Comprehensive

Required role
Configuration groups

System administrator
Memory Use, Meta-Data Cache

For Adaptive Servers running with multiple engines, partition spinlock ratio sets
the number of rows in the partition descriptors that are protected by one
spinlock.
Adaptive Server manages access to table partitions using partition descriptors.
Each partition descriptor stores information about a partition (for example, the
last page of the partition) that processes must use when accessing that partition.
Configure partition descriptors using number of open partitions.
The default value of partition spinlock ratio sets 1 spinlock for every 10 partition
caches. Decreasing the value of partition spinlock ratio may have little impact on
the performance of Adaptive Server. The default setting is correct for most
servers.
See Chapter 5, Managing Mulitprocessor Servers, in System Administration
Guide: Volume 2.

System Administration Guide: Volume 1

229

Configuration parameters

pci memory size

Summary information
Default value
Valid values

64MB
0 2147483647

Status
Display level

Dynamic
Intermediate

Required role
Configuration group

System Administrator
User Environment

pci memory size sets the size of the pluggable component interface (PCI)

memory pool. All pluggable component adapter (PCA) and JVM plug-ins
running under the PCI Bridge share a single dedicated PCI memory pool. If
you set pci memory size to less than the default, Adaptive Server uses the
default size.
This memory pool is fully dedicated to the PCI bridge and any running
pluggable component. Like all other memory pools, Adaptive Server controls
this memory pool. However, unlike other memory pools, the PCI memory pool
is allocated when you initialize the PCI Bridge and does not grow after that
time.

per object statistics active

Summary information
Default value
Range of values

0 (off)
0 (off), 1 (on)

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
Monitoring

per object statistic active controls whether Adaptive Server collects statistics for

each object.

percent database for history

230

Summary information
Default value

20

Valid values

0 100

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

Summary information
Status
Display level

Dynamic
Intermediate

Required role
Configuration group

System administrator
SQL Server Administration

percent database for history specifies the percentage of the total space available
in sybmgmtdb that is reserved for the js_history table. Increase percent database
for history if there are more jobs running, or to store historical records about
executed jobs for future queries.

percent database for output

Summary information
Default value
Valid values

30
0 100

Status
Display level

Dynamic
Intermediate

Required role
Configuration group

System administrator
SQL Server Administration

percent database for output specifies the percentage of the total space available
in sybmgmtdb that is reserved for job output. Increase the default value if there
are more jobs running or jobs that produce lot of output that must be stored for
querying.

percent history free

Summary information
Default value
Valid values

30
0 100

Status
Display level

Dynamic
Intermediate

Required role
Configuration group

System administrator
SQL Server Administration

System Administration Guide: Volume 1

231

Configuration parameters

percent history free specifies the percentage of reserved space in sybmgmtdb to

be kept free For example, if you use the default value, Adaptive Server starts
purging the oldest history records to make room for new records when 70
percent of sybmgmtdb is filled.

percent output free

Summary information
Default value
Valid values

50
0 100

Status
Display level

Dynamic
Intermediate

Required role

System administrator

Configuration group

SQL Server Administration

Specifies the percentage of reserved space kept free in sybmgmtdb that is

reserved for Job Scheduler output. For example, if you use the default value,
Adaptive Server starts purging the oldest history records to make room for new
records when 50 percent of sybmgmtdb is filled.

performance monitoring option

Summary information
Default value

0 (off)

Range of values
Status

0 (off), 1 (on)
Dynamic

Display level
Required role

Comprehensive
System administrator

Configuration group

Monitoring

performance monitoring option enables the license for the BMC DBXray
graphical performance monitoring and diagnostic tool .

permission cache entries

232

Summary information
Default value

64

Range of values

12147483647

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

Summary information
Status
Display level

Dynamic
Comprehensive

Required role
Configuration groups

System administrator
Memory Use, User Environment

permission cache entries determines the number of cache protectors per task,
increasing the amount of memory for each user connection and worker process.

Information about user permissions is held in the permission cache. When

Adaptive Server checks permissions, it looks first in the permission cache; if it
does not find what it needs, it looks in the sysprotects table. This process is
significantly faster if Adaptive Server finds the information it needs in the
permission cache and does not have to read sysprotects.
However, Adaptive Server looks in the permission cache only when it is
checking user permissions, not when permissions are being granted or revoked.
When a permission is granted or revoked, the entire permission cache is
flushed. This is because existing permissions have timestamps that become
outdated when new permissions are granted or revoked.
If users on your Adaptive Server frequently perform operations that require
their permissions to be checked, you may see a small performance gain by
increasing the value of permission cache entries. This effect is not likely to be
significant enough to warrant extensive tuning.
If users on your Adaptive Server frequently grant or revoke permissions, avoid
setting permission cache entries to a large value. The space used for the
permission cache would be wasted, since the cache is flushed with each grant
and revoke command.

plan text pipe active

Summary information
Default value
Range of values

0
01

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
Monitoring

System Administration Guide: Volume 1

233

Configuration parameters

plan text pipe active determines whether Adaptive Server collects query plan
text. If both plan text pipe active and plan text pipe max messages are enabled,

Adaptive Server collects the plan text for each query. You can use
monSysPlanText to retrieve the query plan text for all user tasks.

plan text pipe max messages

Summary information
Default value
Range of values

0
02147483647

Status
Display level

Dynamic
Comprehensive

Required role

System administrator

Configuration groups

Memory Use, Monitoring

plan text pipe max messages determines the number of query plan text

messages Adaptive Server stores per engine. The total number of messages in
the monSQLText table is the value of sql text pipe max messages multiplied by
the number of engines running.

print deadlock information

Summary information
Default value

0 (off)

Valid values
Status

0 (off), 1 (on), 2 (on, print summary)

Dynamic

Display level
Required role

Intermediate
System administrator

Configuration groups

Lock Manager, SQL Server Administration

print deadlock information prints deadlock information to the error log.

If you are experiencing recurring deadlocks, setting print deadlock information

to 1 provides you with detailed information in the error log that can be useful
in tracing the cause of the deadlocks. However, setting print deadlock
information to 1 can degrade Adaptive Server performance. For this reason, set
print deadlock information on only when you are trying to determine the cause
of deadlocks.

234

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

Use sp_sysmon output to determine whether deadlocks are occurring in your

application. If they are, set print deadlock information to 1 to learn more about
why they are occurring. See the Performance and Tuning Series: Monitoring
Adaptive Server with sp_sysmon.
A value of 2 allows you to print a summary of deadlock information to the error
log (as opposed to the detailed information a value of 1 provides). For example:
Deadlock Id 34: Process (Familyid 0, Spid 70) was waiting for a 'exclusive page'
lock on page 10858346 of the 'equineline_job' table in database 18 but process
(Familyid 0, Spid 88) already held a 'exclusive page' lock on it.
Deadlock Id 34: Process (Familyid 0, Spid 88) was waiting for a 'exclusive page'
lock on page 11540986 of the 'equineline_job' table in database 18 but process
(Familyid 0, Spid 70) already held a 'update page' lock on it.

print recovery information

Summary information
Default value

0 (off)

Valid values
Status

0 (off), 1 (on)
Dynamic

Display level
Required role

Intermediate
System administrator

Configuration group

Backup/Recovery

print recovery information determines what information Adaptive Server

displays on the console during recovery. (Recovery is performed on each

database at Adaptive Server start-up and when a database dump is loaded.) The
default value means that Adaptive Server displays only the database name and
a message saying that recovery is in progress. A value of 1 indicates that
Adaptive Server displays information about each individual transaction
processed during recovery, including whether it was aborted or committed.

procedure cache size

Summary information
Default value

7000

Range of values
Status

7000 2147483647
Dynamic

Display level
Required role

Basic
System administrator

System Administration Guide: Volume 1

235

Configuration parameters

Summary information
Configuration groups

Memory Use, SQL Server Administration

Specifies the size of the procedure cache, in 2K pages. Adaptive Server uses
the procedure cache while running stored procedures. If the server finds a copy
of a procedure already in the cache, it does not need to read it from the disk.
Adaptive Server also uses space in the procedure cache to compile queries
while creating stored procedures.
Since the optimum value for procedure cache size differs from application to
application, resetting it may improve Adaptive Server performance. For
example, if you run many different procedures or ad hoc queries, your
application uses the procedure cache more heavily, so you may want to
increase this value.
Warning! If procedure cache size is too small, Adaptive Server performance

degrades.
If you are upgrading, procedure cache size is set to the size of the original
procedure cache at the time of upgrade.

procedure deferred compilation

Summary information
Default value

1(enabled)

Range of values
Status

01
dynamic

Display level
Required role

System administrator

Configuration groups

Query tuning

When this parameter is enabled, compiling of statements that reference local

variables or temporary tables inside a stored procedure is postponed to
execution time, so that the optimization of those statements can use runtime
values, instead of estimations or magic numbers.

236

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

process wait events

Summary information
Default value
Range of values

0 (off)
0 (off), 1 (on)

Status
Display level

Dynamic
Comprehensive

Required role
Configuration groups

System administrator
Memory Use, Monitoring

process wait events controls whether Adaptive Server collect statistics for each
wait event for every task. You can get wait information for a specific task using
monProcessWaits.

See Chapter 17, Using Stored Procedures, in the Transact-SQL Users Guide.

prod-consumer overlap factor

Summary information
Default value
Range of values

20

Status
Display level

dynamic
Comprehensive

Required role
Configuration group

System administrator

prod-consumer overlap factor affects optimization. Adaptive Server changes the

group by algorithm, and you cannot use set statistics I/O with parallel plans.

quorum heartbeat interval

Summary information
Default value

Valid values
Status

1 60
Dynamic

Display level
Required role

Comprehensive
System administrator

Configuration group

Shared Disk Cluster

System Administration Guide: Volume 1

237

Configuration parameters

quorum heartbeat interval specifies the number of seconds between quorum

heartbeats. Setting quorum heartbeat interval to a lower number increases the

heartbeat overhead but speeds the detection of a lost disk link, resulting in a
quicker termination of an instance for which you have set I/O fencing or that
has lost its SAN link. Setting quorum heartbeat interval to a high number
reduces heartbeat overhead, but delays the detection of a lost disk link.

quorum heartbeat retries

Summary information
Default value
Valid values

2
0 32767

Status

Dynamic

Display level
Required role

Comprehensive
System administrator

Configuration group

Shared Disk Cluster

quoted identifier enhancements

Summary information
Default value
Valid values

0 (off)
0 (off), 1(on)

Status
Display level

Static
Comprehensive

Required role
Configuration group

System administrator
Application Functionality

quoted identifier enhancement enables Adaptive Server to use quoted identifiers

for:

Tables

Views

Column names

Index names

System procedure parameters

See the Reference Manual: Blocks.

238

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

read committed with lock

Summary information
Default value
Valid values

0 (off)
0 (off), 1(on)

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
Lock Manager

read committed with lock determines whether an Adaptive Server using

transaction isolation level 1 (read committed) holds shared locks on rows or

pages of data-only-locked tables during select queries. For cursors, read
committed with lock applies only to read-only cursors declared.
For transaction isolation level 1, select queries on allpages-locked tables
continue to hold locks on the page at the current position. Any updatable cursor
on a data-only-locked table also holds locks on the current page or row. See the
Performance and Tuning Series: Basics.

recovery interval in minutes

Summary information
Default value
Range of values

5
132767

Status
Display level

Dynamic
Basic

Required role
Configuration group

System administrator
Backup/Recovery

recovery interval in minutes sets the maximum number of minutes per database

that Adaptive Server uses to complete its recovery procedures in case of a

system failure. The recovery procedure rolls transactions backward or forward,
starting from the transaction that the checkpoint process indicates as the oldest
active transaction. The recovery process has more or less work to do,
depending on the value of recovery interval in minutes.

System Administration Guide: Volume 1

239

Configuration parameters

Adaptive Server estimates that 6000 rows in the transaction log require 1
minute of recovery time. However, different types of log records can take more
or less time to recover. If you set recovery interval in minutes to 3, the checkpoint
process writes changed pages to disk only when syslogs contains more than
18,000 rows since the last checkpoint.
Note The recovery interval has no effect on long-running, minimally logged
transactions (such as create index) that are active when Adaptive Server fails.

It may take as much time to reverse these transactions as it took to run them.
To avoid lengthy delays, dump each database after index maintenance
operations.
Adaptive Server uses the recovery interval in minutes setting and the amount of
activity on each database to decide when to checkpoint each database. When
Adaptive Server checkpoints a database, it writes all dirty pages (data pages
in cache that have been modified) to disk. This may create a brief period of high
I/O, called a checkpoint spike. The checkpoint also performs other
maintenance tasks, including truncating the transaction log for each database
for which the truncate log on chkpt option has been set. About once per minute,
the sleeping checkpoint process wakes up, checks the truncate log on chkpt
setting, and checks the recovery interval to determine if a checkpoint is needed.
Figure 5-4 shows the logic used by Adaptive Server during this process.

240

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

Figure 5-4: The checkpoint process

Sleep

Checkpoint
process
wakes up
and polls
each database

Is truncate log
on chkpt set?

Yes

Checkpoint
performed

No
No

Time to
checkpoint?

Logs
truncated

Yes
Checkpoint
performed
You may want to change the recovery interval if your application and its use
change. For example, you may want to shorten the recovery interval when
there is an increase in update activity on Adaptive Server. Shortening the
recovery interval causes more frequent checkpoints, with smaller, more
frequent checkpoint spikes, and slows the system slightly. However, setting the
recovery interval too high may cause the recovery time to be unacceptably
long. You can reduce the spikes caused by checkpointing by reconfiguring the
housekeeper freewrite percent parameter. See housekeeper free write percent
on page 149. For more information on the performance implications of
recovery interval in minutes, see Chapter 5, Memory Use and Performance, in
the Performance and Tuning Series: Basics.
Use sp_sysmon to determine how a particular recovery interval affects the
system. See the Performance and Tuning Series: Monitoring Adaptive Server
with sp_sysmon.

System Administration Guide: Volume 1

241

Configuration parameters

remote server pre-read packets

Summary information
Default value
Range of values

3
3255

Status
Display level

Static
Intermediate

Required role
Configuration groups

System administrator
Memory Use, Network Communication

remote server pre-read packets determines the number of packets that are preread by a site handler during connections with remote servers.

To reduce the required number of connections, communication between two

servers is managed through a single site handler. The site handler can pre-read
and keep track of data packets for each user process before the receiving
process is ready to accept them.
The default value for remote server pre-read packets is appropriate for most
servers. Increasing the value uses more memory; decreasing the value can slow
network traffic between servers. See Chapter 7, Managing Remote Servers.

restricted decrypt permission

Summary information
Default value
Range of values

0
0 (off), 1 (on)

Status
Display level

Dynamic
Basic

Required role
Configuration group

System security officer

Security Related

restricted decrypt permission enables or disables restricted decrypt permission

in all databases. You must have the sso_role to set this parameter.

When restricted decrypt permission is set to 0 (off), decrypt permission on

encrypted columns acts the same as in versions earlier than 15.0.2:

242

The table owner or the SSO explicitly grants decrypt permission.

However, with grant option on decrypt permission is supported.

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

Decrypt permission is granted implicitly to table owners and the SSO, as

well as to any user through a chain of ownership. For example, if user Fred
owns the proc1 stored procedure, which selects data from the encrypted
column fred.table1.col1, and if Fred grants exec permission on proc1 to
Harry, then Harry has implicit decrypt permission on fred.table1.col1

Decrypt permission is not needed for alter table decrypt. because the table
owner has implicit decrypt permission on encrypted columns.

When restricted decrypt permission is set to 1 (on):

Decrypt permission is granted implicitly only to the SSO.

The SSO can grant decrypt permission using the with grant option
parameter. This allows the SSO to decide who can grant decrypt
permission in the system. For example, if the SSO wants user1 to be able
to grant decrypt permission on user3.user3_tab, the SSO issues:
grant decrypt on user3.user3_tab to user1
with grant option

If you use a system encryption password, Sybase recommends that, to

protect data privacy, you do not grant decrypt permission to the DBO to.
Access to keys through user passwords prevents the DBO and other parties
from accessing the data unless they have a keys password; however, you
may find it convenient for the DBO to decide which users should see the
decrypted data. If you are not protecting keys and data with user-specified
passwords, the SSO should retain the sole responsibility to grant decrypt
permission.

Table ownership does not give a user implicit decrypt permission. That is,
if you create a table with encrypted columns, you do not have decrypt
permission on them unless it is explicitly granted to you.

No user is implicitly granted decrypt permission through an ownership

chain. For example, if Fred owns the proc1 stored procedure, which selects
data from the encrypted column fred.table1.col1, and if Fred grants exec
permission on proc1 to Harry, then Harry must also have explicit decrypt
permission on fred.table1.col1 to see the data.

Aliased users assume the permissions of the user to whom they are aliased.
Similarly, a user with sa_role, who is implicitly aliased to the DBO in any
database, inherits any decrypt permissions that have been explicitly
granted to the DBO.

Decrypt permission is required for alter table decrypt statement because the
table owner does not have implicit decrypt permission on the table.

System Administration Guide: Volume 1

243

Configuration parameters

If you change restricted decrypt permission from 0 to 1, currently executing

statements that use implicit decrypt permission finish; however any subsequent
statements that use implicit decrypt permission fail with this error until the
SSO grants the user decrypt permission on the necessary columns:
Msg 10330 "DECRYPT permission denied on object object_name, database
database_name, owner owner_name."

If you change restricted decrypt permission from 1 to 0, the rows that reflect
explicit grants remain in the sysprotects system table. However, these rows
have no effect on implicitly granted decrypt permissions because Adaptive
Server does not check sysprotects to make sure decrypt permission can be
implicitly granted. sp_helprotect displays misleading information for only
those users who were granted or revoked explicit decrypt permission before
you reconfigure the system, and who now have implicit decrypt permission.
Sybase recommends that, to keep the system consistent, you revoke any
explicit decrypt permissions granted to users before you switch between
enabling or disabling restricted decrypt permission to keep the system
consistent.
See the Encrypted Columns Users Guide for more information about decrypt
permissions.

row lock promotion HWM

Summary information
Default value

200

Range of values
Status

22147483647
Dynamic

Display level
Required role

Intermediate
System administrator

Configuration groups

Lock Manager, SQL Server Administration

row lock promotion HWM (high-water mark), with row lock promotion LWM (lowwater mark) and row lock promotion PCT specifies the maximum number of row
locks permitted during a single scan session of a table or an index before
Adaptive Server attempts to escalate from row locks to a table lock.

When the number of locks acquired during a scan session exceeds row lock
promotion HWM, Adaptive Server attempts to acquire a table lock. The lock
promotion HWM value cannot be higher than the number of locks value.

244

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

See Chapter 2, Locking Configuration and Tuning, in Performance and

Tuning Series: Locking and Concurrency Control.
The default value for row lock promotion HWM is appropriate for most
applications. To avoid table locking, you may want to increase the value of row
lock promotion HWM.. For example, if you know that there are regular updates
to 500 rows on a table that has thousands of rows, you can increase
concurrency for the tables by setting row lock promotion HWM to around 500.
You can also configure row lock promotion at the object level. See
sp_setpglockpromote in the Reference Manual: Procedures.

row lock promotion LWM

Summary information
Default value

200

Range of values
Status

2value of row lock promotion HWM

Dynamic

Display level
Required role

Intermediate
System administrator

Configuration groups

Lock Manager, SQL Server Administration

row lock promotion LWM (low-water mark), with the row lock promotion HWM
(high-water mark) and row lock promotion PCT specifies the number of row

locks permitted during a single scan session of a table or an index before

Adaptive Server attempts to promote from row locks to a table lock.
row lock promotion LWM sets the number of locks below which Adaptive Server
does not attempt to acquire a table lock on the object. The row lock promotion
LWM must be less than or equal to row lock promotion HWM.

The default value for row lock promotion LWM is sufficient for most
applications. If Adaptive Server runs out of locks (except for an isolated
incident), increase number of locks.
See the Performance and Tuning Series: Locking and Concurrency Control.
You can also configure lock promotion at the object level. See
sp_setpglockpromote in the Reference Manual: Procedures.

System Administration Guide: Volume 1

245

Configuration parameters

row lock promotion PCT

Summary information
Default value
Range of values

100
1100

Status
Display level

Dynamic
Intermediate

Required role
Configuration groups

System administrator
Lock Manager, SQL Server Administration

If the number of locks held on an object is between row lock promotion LWM
(low-water mark) and row lock promotion HWM (high-water mark), row lock
promotion PCT sets the percentage of row locks (based on the number of rows
in the table) above which Adaptive Server attempts to acquire a table lock.
The default value for row lock promotion PCT is appropriate for most
applications.
For more information on setting up lock promotion limits, see Chapter 2,
Locking Configuration and Tuning, in Performance and Tuning Series:
Locking and Concurrency Control.
You can also configure row lock promotion at the per-object level. See
sp_sterowlockpromote in the Reference Manual: Procedures.

rtm thread idle wait period

Summary information
Default value
Range of values

600
600 4026531839

Status
Display level

Dynamic
Intermediate

Required role
Configuration groups

System administrator
SQL Server Administration

rtm thread idle wait period defines the time, in seconds, a native thread used by

Adaptive Server waits when it has no work to do. When the time set for a native
thread is reached, the thread automatically fades out.

246

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

runnable process search count

Summary information
Default value
Range of values

2000 (default value of 3 for the Cluster Edition)

0 2147483647

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
SQL Server Administration

Note runnable process search count functions only when you configure
Adaptive Server for process kernel mode; it is nonfunctional for threaded
kernel mode. Use alter thread pool pool_name with idle timeout = n instead.
runnable process search count specifies the number of times an engine loops
while looking for a runnable task before relinquishing the CPU to the operating
system.

Adaptive Server engines check the run queue for runnable tasks whenever a
task completes or exceeds its allotted time on the engine. At times, there are no
tasks in the run queues. An engine can either relinquish the CPU to the
operating system or continue to check for a task to run. Setting runnable
process search count higher causes the engine to loop more times, thus holding
the CPU for a longer time. Setting the runnable process search count lower
causes the engine to release the CPU sooner.
If your machine is a uniprocessor that depends on helper threads to perform
I/O, you may see some performance benefit from setting runnable process
search to perform network I/O, disk I/O, or other operating system tasks. If a
client, such as a bulk-copy operation, is running on the same machine as a
single CPU server that uses helper threads, you may need to allow both the
server and the client access to the CPU.
Note If you are having performance problems, try setting runnable process
search count to 3.

For Adaptive Servers running on uniprocessor machines that do not use helper
threads, and for multiprocessor machines, the default value should provide
good performance.

System Administration Guide: Volume 1

247

Configuration parameters

With a runnable process search count value of 3, the Cluster Edition can better
share the system CPU with other processes running on the same machine.
However, if you set runnable process search count to 3 and Adaptive Server is
running as a standalone process, users may experience delays in server
response times. In this case, reset runnable process search count to 2000.
Use sp_sysmon to determine how the runnable process search count parameter
affects the Adaptive Server use of CPU cycles, engine yields to the operating
system, and blocking network checks. See the Performance and Tuning Series:
Monitoring Adaptive Server with sp_sysmon.
runnable process search count versus idle timeout
runnable process search count and alter thread pool ...idle timeout both indicate
how Adaptive Server looks for work:

runnable process search count specifies the number of loops Adaptive

Server spends looking for work, and is a server-wide parameter.

alter thread pool ...idle timeout specifies the period of time Adaptive Server
spends looking for work, and is tuned according to individual thread pools.
idle timeout is more consistent across processors with varying speeds.

sampling percent
Summary information
Default value

Range of values
Status

0 100 percent
Dynamic

Display level
Required role

Comprehensive
System or database administrator

Configuration group

Query Tuning

sampling percent is the numeric value of the sampling percentage, such as 5 for

5%, 10 for 10%, and so on.

To reduce I/O contention and resources, run update statistics using a sampling
method, which can reduce the I/O and time when your maintenance window is
small and the data set is large. If you are updating a large data set or table that
is in constant use, being truncated and repopulated, you may want to perform
a statistical sampling to reduce the time and the size of the I/O.
Use caution with sampling, since the results are not fully accurate. Balance
changes to histogram values against the savings in I/O.

248

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

Although a sampling of the data set may not be completely accurate, usually
the histograms and density values are reasonable within an acceptable range.
When you are deciding whether or not to use sampling, consider the size of the
data set, the time constraints you are working with, and if the histogram
produced is as accurate as needed.
The percentage to use when sampling depends on your needs. Test various
percentages until you receive a result that reflects the most accurate
information on a particular data set.
Statistics are stored in the system tables systabstats and sysstatistics.

secure default login

Summary information
Default value

Range of values

0 (followed by another parameter naming the default

login)

Status
Display level

Dynamic
Intermediate

Required role
Configuration group

System security officer

Security Related

secure default login specifies a default login for all users who are
preauthenticated but who do not have a login in master..syslogins.

Establish the secure default login with:

sp_configure "secure default login", 0, default_login_name

where:

secure default login is the name of the parameter.

0 is a required parameter because the second parameter of sp_configure

must be a numeric value.

default_login_name is the name of the default login for a user who is

unknown to Adaptive Server, but who has already been authenticated by a
security mechanism. The login name must be a valid login in
master..syslogins.

System Administration Guide: Volume 1

249

Configuration parameters

select for update

Summary information
Default value
Valid values

0 (off)
0 (off), 1(on)

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
Application Functionality

select for update enables Adaptive Server to exclusively lock rows for

subsequent updates within the same transaction, and for updatable cursors,
preventing other concurrent tasks from updating these rows and from blocking
the subsequent update
See Queries: Selecting Data from a Table in the Transact-SQL Users Guide.

select on syscomments.text
Summary information
Default value
Range of values

0 (off)
0 (off), 1 (on)

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System security officer

Security Related

select on syscomments.text enables protection of the text of database objects

through restriction of the select permission on the text column of the
syscomments table. The default value sets select permission to public. Set the
value to 0 to restrict select permission to the object owner and the system

administrator.

send doneinproc tokens

250

Summary information
Default value

1 (on)

Range of values
Status

0 (off), 1 (on)
Dynamic

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

Summary information
Display level
Required role

Comprehensive
System administrator

Configuration group

Network Communication

send doneinproc tokens enables or disables Adaptive Server for sending

doneinproc packets (TDS messages that are sent after each select statement
performed in a stored procedure). send doneinproc tokens replaces dbcc tune
doneinproc and trace flag 292. Currently running queries immediately take

note of any change in the option.

Setting send doneinproc tokens to 1 is safe in most cases. However some stored
procedures are executed using asynchronous commands from CT-Lib, and
using a value of 0 may cause state-machine errors in some CT-Lib applications.

session migration timeout

Summary information
Default value

600

Valid values
Status

0 32767
Dynamic

Display level
Required role

Comprehensive
System administrator

Configuration group

Shared Disk Cluster

session migration timeout specifies the amount of time available for a client to

complete a migration by connecting to the target instance. If the client does not
migrate to the target instance in the time alloted, Adaptive Server fails the
connection.

session tempdb log cache size

Summary information
Default value

The logical page size, in bytes

Range of values
Status

The logical page size up to 2147483647

Static

Display level
Required role

Comprehensive
System administrator

Configuration group

User Environment

System Administration Guide: Volume 1

251

Configuration parameters

session tempdb log cache size specifies the size for each session tempdb log
cache. Adaptive Server uses the tempdb log cache to buffer the user transaction
log records for the session involving tempdb objects. The size for session
tempdb log cache is determined by the Adaptive Server page size. Configuring
a reasonable size for session tempdb log cache size can help prevent context
switches due to a session having to flush the tempdb user log cache.

Adaptive Server includes two log caches for each user connection: the session
tempdb log cache, and the user log cache, which is determined by the user log
cache size parameter.

shared memory starting address

Summary information
Default value

Range of values
Status

Platform-specific
Static

Display level
Required role

Comprehensive
System administrator

Configuration group

Physical Memory

shared memory starting address determines the virtual address where Adaptive

Server starts its shared memory region.

It is unlikely that you will ever have to reconfigure shared memory starting
address. Do so only after consulting with Sybase Technical Support.
number of worker processes, max parallel degree, and max scan parallel degree
control parallel query processing at the server level. Using the parallel_degree,
process_limit_action, and scan_parallel_degree options to the set command can
limit parallel optimization at the session level, and using the parallel keyword
of the select command can limit parallel optimization of specific queries.

size of auto identity column

252

Summary information
Default value

10

Range of values
Status

138
Dynamic

Display level
Required role

Intermediate
System administrator

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

Summary information
Configuration group

SQL Server Administration

size of auto identity column sets the precision of IDENTITY columns that are
automatically created with the sp_dboption auto identity and unique
auto_identity index options.

The maximum value that can be inserted into an IDENTITY column is

10precision -1. After an IDENTITY column reaches its maximum value, all
further insert statements return an error that aborts the current transaction.
If you reach the maximum value of an IDENTITY column, you can increase it
with a modify operation in the alter table command. See the Transact-SQL
Users Guide for examples.
You can also use the create table command to create a table that is identical to
the old one, but with a larger precision for the IDENTITY column. After you
have created the new table, use the insert command or bcp to copy data from
the old table to the new one.

size of global fixed heap

Summary information
Default values

150 pages (32-bit version)

Minimum values

300 pages (64-bit version)

10 pages (32-bit version)

Status

20 pages (64-bit version)

Dynamic

Display level
Required role

Comprehensive
System administrator

Configuration groups

Java Services, Memory Use

size of global fixed heap specifies the memory space for internal data structures

and other needs.

If you change size of the global fixed heap, change the total logical memory by
the same amount.

System Administration Guide: Volume 1

253

Configuration parameters

size of process object heap

Summary information
Default values

1500 pages (32-bit version)

Minimum values

45 pages (32-bit version)

3000 pages (64-bit version)

90 pages (64-bit version)
Status
Display level

Dynamic
Basic

Required role
Configuration groups

System administrator
Java Services, Memory Use

size of process object fixed heap specifies the total memory space for all

processes using the Java VM.

If you change size of process object fixed heap, change the total logical memory
by the same amount.

size of shared class heap

Summary information
Default values

1536 pages (32-bit version)

Minimum values

3072 pages (64-bit version)

650 pages (32-bit version)
1300 pages (64-bit version)

Status
Display level

Dynamic
Basic

Required role
Configuration groups

System administrator
Java Services, Memory Use

size of shared class heap specifies the shared memory space for all Java classes
called into the Java VM. Adaptive Server maintains the shared class heap
server-wide for both user-defined and system-provided Java classes.

If you change the size of shared class heap, change the total logical memory by
the same amount.

254

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

size of unilib cache

Summary information
Default value
Range of values

0
02147483647

Status
Display level

Dynamic
Comprehensive

Required role
Configuration groups

System administrator
Memory Use, Unicode

size of unilib cache specifies the memory used in bytes rounded up to the nearest

1K in addition to the minimum overhead size, which provides enough memory

to load a single copy of the largest Unilib conversion table plus the largest
Unilib sort table. Asian clients may need to increase size of unilib cache by an
extra 100K for every additional character set they want to support via Unicodebased conversion.

sproc optimize timeout limit

Summary information
Default value
Range of values

40
0 4000

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
Query Tuning

sproc optimize timeout limit specifies the amount of time Adaptive Server can

spend optimizing a stored procedure as a fraction of the estimated execution

time.

SQL batch capture

Summary information
Default value

0 (off)

Range of values
Status

0 (off), 1(on)
Dynamic

Display level
Required role

Comprehensive
System administrator

System Administration Guide: Volume 1

255

Configuration parameters

Summary information
Configuration group

Monitoring

SQL batch capture controls whether Adaptive Server collects SQL text. If both
SQL batch capture and max SQL text monitored are enabled, Adaptive Server

collects the SQL text for each batch for each user task.

SQL Perfmon Integration (Windows only)

Summary information
Default value
Valid values

1 (on)
0 (off), 1 (on)

Status

Static

Display level
Required role

Intermediate
System administrator

Configuration group

SQL Server Administration

SQL Perfmon Integration enables and disables the ability to monitor Adaptive

Server statistics from the Windows Performance Monitor.

Adaptive Server must be registered as a Windows Service to support monitor
integration. This occurs automatically when you start Adaptive Server using
the Services Manager in the Sybase for the Windows program group, and is the
default configuration when you use the Sybase installer or the syconfig utility
to create an Adaptive Server.
The 15.7 version of sybperf exposes a set of Adaptive Server counters that are
more useful for monitoring Adaptive Server performance.

sql server clock tick length

Summary information
Default value

256

Platform-specific

Range of values

Platform-specific minimum1000000, in multiples of

default value

Status
Display level

Static
Comprehensive

Required role
Configuration group

System administrator
SQL Server Administration

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

sql server clock tick length specifies the duration of the servers clock tick, in
microseconds. Both the default value and the minimum value are platformspecific. Adaptive Server rounds values up to an even multiple of n, where n is
the platform-specific clock-tick default value. Use sp_helpconfig or
sp_configure to find the current values for sql server clock tick length.

In mixed-use applications with some CPU-bound tasks, decrease the value of

sql server clock tick length to:

Help I/O-bound tasks a value of 20,000 is reasonable for this. Shortening

the clock-tick length means that CPU-bound tasks exceed the allotted time
on the engine more frequently per unit of time, which allows other tasks
greater access to the CPU

Marginally increase response times Adaptive Server runs its service

tasks once per clock tick. Decreasing the clock-tick length means that the
service tasks are run more frequently per unit of time

Increasing sql server clock tick length favors CPU-bound tasks, because they
execute longer between context switches. The maximum value of 1,000,000
may be appropriate for primarily CPU-bound applications. However, any I/Obound tasks may suffer as a result. You can mitigate this somewhat by tuning
cpu grace time (see cpu grace time on page 104) and time slice (see time
slice on page 271).
Note Changing the value of sql server clock tick length can have serious effects
on Adaptive Server performance. Consult with Sybase Technical Support
before resetting this value.

sql text pipe active

Summary information
Default value
Range of values

0
01

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
Monitoring

System Administration Guide: Volume 1

257

Configuration parameters

sql text pipe active controls whether Adaptive Server collects SQL text. If this
option is enabled and sql text pipe max messages is set, Adaptive Server
collects the SQL text for each query. Use monSysSQLText to retrieve the SQL

text for all user tasks.

sql text pipe max messages

Summary information
Default value
Range of values

0
02147483647

Status
Display level

Dynamic
Comprehensive

Required role

System administrator

Configuration groups

Memory Use, Monitoring

sql text pipe max messages specifies the number of SQL text messages
Adaptive Server stores per engine. The total number of messages in the
monSQLText table is the value of sql text pipe max messages multiplied by the
number of engines running.

stack guard size

Summary information
Default value

4096

Range of values
Status

02147483647
Static

Display level
Required role

Comprehensive
System administrator

Configuration groups

Memory Use, User Environment

stack guard size sets the size, in bytes, of the stack guard area, which is an

overflow stack of configurable size at the end of each stack. Adaptive Server
allocates one stack for each user connection and worker process when it starts.
These stacks are located contiguously in the same area of memory, with a guard
area at the end of each stack. At the end of each stack guard area is a
guardword, which is a 4-byte structure with a known pattern. Figure 5-5
illustrates how a process can corrupt a stack guardword.

258

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

Figure 5-5: Process about to corrupt stack guardword

Stack 2
guard area

Stack guardwords
Stack 2

Stack 1
guard area

Process overflowing stack

area, about to corrupt
stack guardword

Stack 1

Process begins
executing
Adaptive Server periodically checks to see whether the stack pointer for a user
connection has entered the stack guard area associated with that user
connections stack. If it has, Adaptive Server aborts the transaction, returns
control to the application that generated the transaction, and generates error
3626:
The transaction was aborted because it used too much
stack space. Either use sp_configure to increase the
stack size, or break the query into smaller pieces.
spid: %d, suid: %d, hostname: %.*s, application name:
%.*s

System Administration Guide: Volume 1

259

Configuration parameters

Adaptive Server also periodically checks the guardword pattern to see if it has
changed, thus indicating that a process has overflowed the stack boundary.
When this occurs, Adaptive Server prints these messages to the error log and
shuts down:
kernel: *** Stack overflow detected: limit: 0x%lx sp: 0x%lx
kernel: *** Stack Guardword corrupted
kernel: *** Stack corrupted, server aborting

In the first message, limit is the address of the end of the stack guard area,
and sp is the current value of the stack pointer.
In addition, Adaptive Server periodically checks the stack pointer to see
whether it is completely outside both the stack and the stack guard area for the
pointers process. If it is, Adaptive Server shuts down, even if the guardword
is not corrupted. When this happens, Adaptive Server prints the following
messages to the error log:
kernel: *** Stack overflow detected: limit: 0x%lx sp: 0x%lx
kernel: *** Stack corrupted, server aborting

The default value for stack guard size is appropriate for most applications.
However, if you experience server shutdown from either stack guardword
corruption or stack overflow, increase stack guard size by a 2K increment. Each
configured user connection and worker process has a stack guard area; thus,
when you increase stack guard size, you use up that amount of memory,
multiplied by the number of user connections and worker processes you have
configured.
Rather than increasing stack guard size to avoid stack overflow problems,
consider increasing stack size (see stack size on page 260). The stack guard
area is intended as an overflow area, not as an extension to the regular stack.
Adaptive Server allocates stack space for each task by adding the values of the
stack size and stack guard size parameters. stack guard size must be configured
in multiples of 2K. If the value you specify is not a multiple of 2K, sp_configure

verification routines round the value up to the next highest multiple.

stack size
Summary information
Default value
Range of values

260

Platform-specific
Platform-specific minimum2147483647

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

Summary information
Status
Display level

Static
Basic

Required role
Configuration group

System administrator
User Environment

stack size specifies the size, in bytes, of the execution stacks used by each user
process on Adaptive Server. To find the stack size values for your platform, use
sp_helpconfig or sp_configure. stack size must be configured in multiples of 2K.
If the value you specify is not a multiple of 2K, sp_configure verification
routines round the value up to the next highest multiple.

An execution stack is an area of Adaptive Server memory where user processes

keep track of their process context and store local data.
Certain queries can contribute to the probability of a stack overflow. Examples
include queries with extremely long where clauses, long select lists, deeply
nested stored procedures, and multiple selects and updates that holdlock. When
a stack overflow occurs, Adaptive Server prints an error message and rolls back
the transaction. See stack guard size on page 258, and see the
Troubleshooting and Error Messages Guide for more information on specific
error messages.
The two options for remedying stack overflows are to break the large queries
into smaller queries and to increase stack size. Changing stack size affects the
amount of memory required for each configured user connection and worker
process. See total logical memory on page 272.
If you have queries that exceed the size of the execution stack, you may want
to rewrite them as a series of smaller queries, especially if there are only a small
number of such queries, or if you run them infrequently.
There is no way to determine how much stack space a query requires without
actually running the query. Stack space for each user connection and worker
process is preallocated at start-up.
Therefore, determining the appropriate value for stack size is an empirical
process. Test your largest and most complex queries using the default value for
stack size. If they run without generating error messages, the default is
probably sufficient. If they generate error messages, begin by increasing stack
size by a small amount (2K). Re-run your queries and see if the amount you
have added is sufficient. If it is not, continue to increase stack size until queries
run without generating error messages.

System Administration Guide: Volume 1

261

Configuration parameters

If you are using CIS, or if Java is enabled in the database and you want to use
methods that call JDBC, Sybase recommends that you increase the default by
50 percent. If you are not using JDBC or CIS, the standard default value is
usually sufficient.

start mail session (Windows only)

Summary information
Default value
Valid values

0 (off)
0 (off), 1 (on)

Status
Display level

Dynamic
Comprehensive

Required role

System administrator

Configuration group

Extended Stored Procedure

start mail session enables and disables the automatic initiation of an Adaptive

Server mail session when you start Adaptive Server.

A value of 1 configures Adaptive Server to start a mail session the next time
Adaptive Server is started. A value of 0 configures Adaptive Server not to start
a mail session at the next restart.
If start mail session is 0, you can start an Adaptive Server mail session
explicitly, using the xp_startmail system ESP.
Before setting start mail session, you must prepare your Windows system by
creating a mailbox and mail profile for Adaptive Server. Then, create an
Adaptive Server account for Sybmail. See the Configuration Guide for
Windows.

start xp server during boot

Summary information
Default value

0 (off)

Range of values
Status

0 (off), 1 (on)
Static

Display level
Required role
Configuration group

262

Extended Stored Procedures

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

start xp server during boot determines whether XP Server starts when Adaptive

Server starts.
When set to 1, XP Server starts when Adaptive Server starts. If you set start xp
server during boot to 0, XP Server does not start until you run xp_cmdshell.

startup delay
Summary information
Default value
Range of values

0 (off)
0 (off), 1 (on)

Status
Display level

Dynamic
Comprehensive

Required role

System administrator

Configuration group

Query Tuning

startup delay controls when RepAgent is started during the server start. By
default, RepAgent starts at the same time as Adaptive Server. Adaptive Server
writes a message to the error log stating the wait time.

statement cache size

Summary information
Default value

Valid values
Status

Size of cache in 2K pages

Dynamic

Display level
Required role

Comprehensive
System administrator

Configuration groups

Memory Use, SQL Server Administration

statement cache size increases the server allocation of procedure cache

memory and limits the amount of memory from the procedure cache pool used
for cached statements.

Note If you enable the statement cache, you must configure set chained on/off

in its own batch.

Because cached statements are transformed into lightweight stored procedures,
statement caching requires additional open object descriptors.

System Administration Guide: Volume 1

263

Configuration parameters

statement pipe active

Summary information
Default value
Range of values

0 (off)
0 (off), 1 (on)

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
Monitoring

statement pipe active controls whether Adaptive Server collects statement-level

statistics. If both statement pipe active and statement pipe max messages are
enabled, Adaptive Server collects the statement statistics for each query. Use
monSysStatement to retrieve the statistics for all executed statements.

statement pipe max messages

Summary information
Default value
Range of values

0
02147483647

Status
Display level

Dynamic
Comprehensive

Required role
Configuration groups

System administrator
Memory Use, Monitoring

statement pipe max messages determines the number of statement statistics

messages Adaptive Server stores per engine. The total number of messages in
the monSQLText table is the value of sql text pipe max messages multiplied by
the number of engines running.

statement statistics active

264

Summary information
Default value

0 (off)

Range of values
Status

0 (off), 1 (on)
Dynamic

Display level
Required role

Comprehensive
System administrator

Configuration group

Monitoring

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

statement statistic active controls whether Adaptive Server collects monitoring

table statement-level statistics. Use monProcessStatement to get statement
statistics for a specific task.

streamlined dynamic SQL

Summary information
Default value
Valid values

0 (off)
0 (off), 1(on)

Status
Display level

Dynamic
Comprehensive

Required role

System administrator

Configuration group

Application Functionality

quoted identifier enhancement enables the statement cache to store dynamic

SQL statements
See Memory Use and Performance in the Performance and Tuning Series:
Basics.

strict dtm enforcement

Summary information
Default value

0 (off)

Valid values
Status

0 (off), 1(on)
Static

Display level
Required role

10
System administrator

Configuration group

DTM Administration

strict dtm enforcement determines whether or not Adaptive Server transaction

coordination services strictly enforce the ACID properties of distributed
transactions.

In environments where Adaptive Server should propagate and coordinate

transactions only to other Adaptive Servers that support transaction
coordination, set strict dtm enforcement on. If a transaction attempts to update
data in a server that does not support transaction coordination services,
Adaptive Server aborts the transaction.

System Administration Guide: Volume 1

265

Configuration parameters

In heterogeneous environments, you may want to make use of servers that do

not support transaction coordination. This includes earlier versions of Adaptive
Server and non-Sybase database stores configured using CIS. Under these
circumstances, set strict dtm enforcement off to allow Adaptive Server to
propagate transactions to legacy Adaptive Servers and other data stores. This
does not, however, ensure that the remote work of these servers is rolled back
or committed with the original transaction.

suppress js max task message

Summary information
Default value
Valid values

0 (off)
0 (off), 1(on)

Status

Dynamic

Display level
Required role

10
System administrator

Configuration group

Application Functionality

strict dtm enforcement prevents Adaptive Server from printing the Job
Scheduler js maxtask error messages to the errorlog.

suspend audit when device full

Summary information
Default value

0 (off)

Range of values
Status

0 (off), 1 (on)
Dynamic

Display level
Required role

Intermediate
System security officer

Configuration group

Security Related

suspend audit when device full determines what Adaptive Server does when an
audit device becomes completely full.

Note If you have two or more audit tables, each on a separate device other than

the master device, and you have a threshold procedure for each audit table
segment, the audit devices should never become full. Only if a threshold
procedure is not functioning properly does the full condition occur.

266

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

Choose one of these values:

0 truncates the next audit table and starts using it as the current audit
table when the current audit table becomes full. If you set suspend audit
when device full to 0, you ensure that the audit process is never suspended.
However, you incur the risk that older audit records are lost if they have
not been archived.

1 suspends the audit process and all user processes that cause an
auditable event. To resume normal operation, the system security officer
must log in and set up an empty table as the current audit table. During this
period, the system security officer is exempt from normal auditing. If the
system security officers actions would generate audit records under
normal operation, Adaptive Server sends an error message and
information about the event to the error log.

syb_sendmsg port number

Summary information
Default value

Valid values
Status

0, or 102465535, or system limit

Dynamic

Display level
Required role

Comprehensive
System administrator

Configuration group

Network Communication

syb_sendmsg port number specifies the port number that Adaptive Server uses
to send messages to a UDP (User Datagram Protocol) port with sp_sendmsg or
syb_sendmsg.

If more than one engine is configured, a port is used for each engine, numbered
consecutively from the port number specified. If the port number is set to the
default value, 0 Adaptive Server assigns port numbers.
Note Sending messages to UDP ports is not supported on Windows.

A system security officer must set the allow sendmsg configuration parameter
to 1 to enable sending messages to UDP ports. To enable UDP messaging, a
system administrator must set allow sendmsg to 1. See allow sendmsg on
page 87. For more information on UDP messaging, see sp_sendmsg in the
Reference Manual: Procedures.

System Administration Guide: Volume 1

267

Configuration parameters

sysstatistics flush interval

Summary information
Default value
Valid values

0
0 32767

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
SQL Server Administration

sysstatiscts flush interval determines the length of the interval, in minutes,

between flushes of sysstatistics.

Adaptive Server dynamically maintains the statistics for the number of rows
and columns modified in a table as part of any DML statement and flushes
them according to the value of sysstatistics flush interval.
Adaptive Server uses these statistics for query optimization since they are more
accurate. The datachange function determines the amount of data that is
changed at the table, column, or partition level since the last update statistics,
and initiates updating statistics on the object.
The in-memory statistics are always flushed to disk during a polite shutdown
of the server. You can configure sysstatistics flush interval to flush these inmemory statistics to disk by the house keeper task at regular intervals. Set
sysstatistics flush interval to 0 to disable this housekeeper task.

systemwide password expiration

Summary information
Default value
Range of values

0
032767

Status
Display level

Dynamic
Intermediate

Required role
Configuration group

System security officer

Security Related

systemwide password expiration sets the number of days that passwords remain
in effect after they are changed. If systemwide password expiration is set to 0,
passwords do not expire.

268

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

The password expires when the number of specified days passes. For example,
if you create a new login on August 1, 2007 at 10:30 a.m., with a password
expiration interval of 30 days, the password expires on August 31, 2007 at
10:30 a.m.
An accounts password is considered expired if an interval greater than
number_of_days has passed since the last time the password for that account
was changed.
When the number of days remaining before expiration is less than 25 percent
of the value of systemwide password expiration or 7 days, whichever is greater,
each time the user logs in, a message displays, giving the number of days
remaining before expiration. Users can change their passwords anytime before
expiration.
When an accounts password has expired, the user can still log in to Adaptive
Server but cannot execute any commands until he or she has used sp_password
to change his or her password. If the system security officer changes the users
password while the account is in sp_password-only mode, the account returns
to normal after the new password is assigned.
This restriction applies only to login sessions established after the password
has expired. Users who are logged in when their passwords expire are not
affected until the next time they log in.

tape retention in days

Summary information
Default value

Range of values
Status

0365
Dynamic

Display level
Required role

Intermediate
System administrator

Configuration group

Backup/Recovery

tape retention in days specifies the number of days you intend to retain each

tape after it has been used for either a database or a transaction log dump. This
parameter can keep you from accidentally overwriting a dump tape.
For example, if you have set tape retention in days to 7 days, and you attempt
to use the tape before 7 days have elapsed since the last time you dumped to
that tape, Backup Server issues a warning message.

System Administration Guide: Volume 1

269

Configuration parameters

You can override the warning using the with init option when executing the
dump command. Doing this causes the tape to be overwritten and all data on
the tape to be lost.
Both the dump database and dump transaction commands provide a retaindays
option, which overrides the tape retention in days value for a particular dump.
See Chapter 12, Backing Upa and Restoring User Databases, in System
Administration Guide: Volume 2.

tcp no delay
Summary information
Default value

1 (on)

Valid values
Status

0 (off), 1 (on)
Static

Display level
Required role

Comprehensive
System administrator

Configuration groups

Network Communication, O/S Resources

tcp no delay controls TCP (Transmission Control Protocol) packet batching.

The default value means that TCP packets are not batched.
TCP normally batches small logical packets into single, larger, physical
packets, and fills physical network frames with as much data as possible, which
improves network throughput in terminal emulation environments where users
mostly send keystrokes across the network.
However, applications that use small TDS (Tabular Data Stream) packets may
benefit from disabling TCP packet batching.
Note Disabling TCP packet batching means that packets are sent, regardless

of size; this increases the volume of network traffic.

text prefetch size

270

Summary information
Default value

16

Valid values
Status

065535
Dynamic

Display level

Comprehensive

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

Summary information
Required role
Configuration Group

System administrator
Network Communications

text prefetch size limits the number of pages of text, unitext, and image data that

can be prefetched into an existing buffer pool. Adaptive Server prefetches only
text, unitext, and image data that was created with Adaptive Server 12.x or was
upgraded using dbcc rebuild_text.

time slice
Summary information
Default value
Range of values

100
501000

Status
Display level

Dynamic
Comprehensive

Required role
Configuration group

System administrator
SQL Server Administration

time slice sets the number of milliseconds that the Adaptive Server scheduler
allows a task to run. If time slice is set too low, Adaptive Server may spend too
much time switching between tasks, which increases response time. If it is set
too high, CPU-intensive tasks may monopolize engines, which also increases
response time. The default value allows each task to run for 1/10 of a second
before relinquishing the CPU to another task.

See Chapter 3, Using Engines and CPUs in the Performance and Tuning
Series: Basics.
Use sp_sysmon to determine how time slice affects voluntary yields by
Adaptive Server engines. See the Performance and Tuning Series: Monitoring
Adaptive Server with sp_sysmon.

total data cache size

Summary information
Default value
Range of values

0
0 2147483647

Status
Display level

Calculated
Basic

System Administration Guide: Volume 1

271

Configuration parameters

Summary information
Required role
Configuration groups

System administrator
Cache Manager, Memory Use

total data cache size reports the amount of memory, in kilobytes, that is

currently available for data, index, and log pages. This parameter is a
calculated value that is not directly user-configurable.
The amount of memory available for the data cache can be affected by a
number of factors, including:

The amount of physical memory available on your machine

The values to which the following parameters are set:

total logical memory

number of user connections

total procedure cache percent

number of open databases

number of open objects

number of open indexes

number of devices

A number of other parameters also affect the amount of available memory, but
to a lesser extent.
For information on how Adaptive Server allocates memory and for information
on data caches, see Configuration parameters on page 80.

total logical memory

Summary information

272

Default value
Range of values

N/A
N/A

Status
Display level

Read-only
Intermediate

Required role
Configuration groups

System administrator
Memory Use, Physical Memory

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

total logical memory displays the total logical memory for the current

configuration of Adaptive Server. The total logical memory is the amount of

memory that the Adaptive Server current configuration uses. total logical
memory displays the memory that is required to be available, but which may or
may not be in use at any given moment. For information about the amount of
memory in use at a given moment, see total physical memory. You cannot use
total logical memory to set any of the memory configuration parameters.

total physical memory

Summary information
Default value
Range of values

N/A
N/A

Status

Read-only

Display level
Required role

Intermediate
System administrator

Configuration group

Memory Use

total physical memory is a read-only configuration parameter that displays the

total physical memory for the current configuration of Adaptive Server. The
total physical memory is the amount of memory that Adaptive Server is using
at a given moment in time. Configure Adaptive Server so that the value for max
memory is larger than the value for total logical memory, and the value for total
logical memory is larger than the value for total physical memory.

transfer utility memory size

Summary information
Default value

4096

Range of values
Status

0 2147483647
Dynamic

Display level
Required role

Intermediate
System administrator

Configuration group

SQL Server Administration

System Administration Guide: Volume 1

273

Configuration parameters

Adaptive Server maintains a memory pool for the transfer table command and
for tables marked for incremental transfer. This pool provides memory for
maintaining state information about current and past transfers, and for memory
used to write to and read from transfer files. transfer utility memory size
determines the size of this memory pool.
The units for this pool are in memory pages, which are blocks of 2048 bytes.
The default size is large enough to accommodate over 100 tables marked for
incremental transfer, all transferring simultaneously.
If your installation does not include tables marked for incremental transfer, and
does not use the transfer table command, you may set the size of this memory
pool to zero to reclaim this memory.

txn to pss ratio

Summary information
Default value

16

Valid values
Status

1 2147483647
Static

Display level
Required role

1
System administrator

Configuration groups

DTM Administration, Memory Use

Adaptive Server manages transactions as configurable server resources. Each

time a new transaction begins, Adaptive Server must obtain a free transaction
descriptor from a global pool that is created when the server is started.
Transaction descriptors are internal memory structures that Adaptive Server
uses to represent active transactions.
Adaptive Server requires one free transaction descriptor for:

274

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

The outer block of each server transaction. The outer block of a transaction
may be created explicitly when a client executes a new begin transaction
command. Adaptive Server may also implicitly create an outer transaction
block when clients use Transact-SQL to modify data without using begin
transaction to define the transaction.
Note Subsequent, nested transaction blocks, created with additional begin
transaction commands, do not require additional transaction descriptors.

Each database accessed in a multidatabase transaction. Adaptive Server

must obtain a new transaction descriptor each time a transaction uses or
modifies data in a new database.

txn to pss ratio determines the total number of transaction descriptors available

to the server. At start-up, this ratio is multiplied by the number of PSS

structures to create the transaction descriptor pool:
# of transaction descriptors = PSS structures * txn to pss ratio

The default value, 16, ensures compatibility with versions of Adaptive Server
earlier than 12.x which also allocated 16 transaction descriptors for each user
connection. In version 12.x and later, the number of simultaneous transactions
is limited only by the number of transaction descriptors available in the server.
Note You can have as many databases in a user transaction as there are in your
Adaptive Server installation. For example, if your Adaptive Server has 25
databases, you can include 25 databases in your user transactions.

Optimizing the txn to pss ratio for your system

During a peak period, use sp_monitorconfig to examine the use of transaction

descriptors:
sp_monitorconfig "txn to pss ratio"
Usage information at date and time: Apr 22 2002 2:49PM.
Name
Num_free
Num_active Pct_act
Reuse_cnt
Instance_Name
--------------------------------- ---------- ---------------- -------------------txn to pss ratio
784
80
10.20
0
NULL

System Administration Guide: Volume 1

Max_Used
----------523

275

Configuration parameters

If the num_used value is zero or very low, transactions may be delayed as

Adaptive Server waits for transaction descriptors to become free in the server.
In this case, consider increasing the value of txn to pss ratio.
If the Max_Used value is too low, unused transaction descriptors may be
consuming memory that can be used by other server functions. Consider
reducing the value of txn to pss ratio.

unified login required

Summary information
Default value
Range of values

0 (off)
0 (off), 1 (on)

Status

Dynamic

Display level
Required role

Intermediate
System security officer

Configuration group

Security Related

unified login required requires that all users who log in to Adaptive Server be
authenticated by a security mechanism. The use security services parameter
must be 1 to use the unified login security service.

upgrade version
Summary information
Default value

1100

Range of values
Status

02147483647
Dynamic

Display level
Required role

Comprehensive
System administrator

Configuration group

SQL Server Administration

upgrade version reports the version of the upgrade utility that upgraded your

master device. The upgrade utility checks and modifies this parameter during
an upgrade.
Warning! Do not reset upgrade version. Doing so may cause serious problems
with Adaptive Server.

276

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

You can determine whether an upgrade has been done on your master device
by using upgrade version without specifying a value:
sp_configure "upgrade version"

use security services

Summary information
Default value
Range of values

0 (off)
0 (off), 1 (on)

Status
Display level

Static
Intermediate

Required role
Configuration group

System security officer

Security Related

use security services specifies that Adaptive

Server uses network-based

security services. If the parameter is set to 0, none of the network-based
security services can be used.

user log cache size

Summary information
Default value
Range of values

Logical page size

2048a 2147483647
a. Minimum determined by servers logical page size

Status

Static

Display level
Required role

Intermediate
System administrator

Configuration groups

Memory Use, User Environment

user log cache size specifies the size, in bytes, for each users log cache. Its size

is determined by the servers logical page size. There is one user log cache for
each configured user connection and worker process. Adaptive Server uses
these caches to buffer the user transaction log records, which reduces the
contention at the end of the transaction log.

System Administration Guide: Volume 1

277

Configuration parameters

When a user log cache becomes full or another event occurs (such as when the
transaction completes), Adaptive Server flushes all log records from the user
log cache to the database transaction log. By first consolidating the log records
in each users log cache, rather than immediately adding each record to the
databases transaction log, Adaptive Server reduces contention of processes
writing to the log, especially for SMP systems that are configured with more
than one engine.
Note For transactions using a database with mixed data and log segments, the
user log cache is flushed to the transaction log after each log record. No
buffering takes place. If your databases do not have dedicated log segments, do
not increase the user log cache size.

Do not configure user log cache size to be larger than the maximum amount of
log information written by an applications transaction. Since Adaptive Server
flushes the user log cache when the transaction completes, any additional
memory allocated to the user log cache is wasted. If no transaction in your
server generates more than 4000 bytes of transaction log records, set user log
cache size no higher than that value. For example:
sp_configure "user log cache size", 4000

Setting user log cache size too high wastes memory. Setting it too low can
cause the user log cache to fill up and flush more than once per transaction,
increasing the contention for the transaction log. If the volume of transactions
is low, the amount of contention for the transaction log may not be significant.
Use sp_sysmon to understand how this parameter affects cache behavior. See
the Performance and Tuning Series: Monitoring Adaptive Server with
sp_sysmon.

user log cache spinlock ratio

Summary information

278

Default value
Range of values

20
12147483647

Status
Display level

Dynamic
Intermediate

Required role
Configuration groups

System administrator
Memory Use, User Environment

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

For Adaptive Servers running with multiple engines, user log cache spinlock
ratio specifies the ratio of user log caches per user log cache spinlock. There is
one user log cache for each configured user connection.
The default specifies 1 spinlock for each 20 user connections configured for
your server.
Use sp_sysmon to understand how this parameter affects cache behavior. See
the Performance and Tuning Series: Monitoring Adaptive Server with
sp_sysmon.

wait event timing

Summary information
Default value

Range of values
Status

01
Dynamic

Display level
Required role

Comprehensive
System administrator

Configuration groups

Memory Use, Monitoring

wait event timing controls whether Adaptive Server collects statistics for

individual wait events. A task may have to wait for a variety of reasons (for
example, waiting for a buffer read to complete). The monSysWaits table
contains the statistics for each wait event. The monWaitEventInfo table contains
a complete list of wait events.

workload manager cache size

Summary information
Default value

80

Valid values
Status

80 2147483647
Dynamic

Display level
Required role

Comprehensive
System administrator

Configuration group

Shared Disk Cluster

workload manager cache size specifies the maximum amount of memory, in 2K

pages, that the workload manager can use. See Chapter 6, Managing the
Workload, in the Cluster Users Guide.

System Administration Guide: Volume 1

279

Configuration parameters

xact coordination interval

Summary information
Default value
Valid values

60 (seconds)
1 2147483647 (seconds)

Status
Display level

Dynamic
10

Required role
Configuration group

System administrator
DTM Administration

xact coordination interval defines the length of time between attempts to resolve
transaction branches have been propagated to remote servers.

The coordinating Adaptive Server makes regular attempts to resolve the work
of remote servers participating in a distributed transaction. The coordinating
server contacts each remote server participating in the distributed transaction
in a serial manner, as shown in Figure 5-6. The coordination service may be
unable to resolve a transaction branch for a variety of reasons. For example, if
the remote server is not reachable due to network problems, the coordinating
server reattempts the connection after the time specified by xact coordination
level.
Figure 5-6: Resolving remote transaction branches

Adaptive Server
Coordination Service
OK
Adaptive Server 2

OK
Adaptive Server 3

Network problem
Adaptive Server 4

With the default value of xact coordination interval, 60, Adaptive Server
attempts to resolve remote transactions once every minute. Decreasing the
value may speed the completion of distributed transactions, but only if the
transactions are themselves resolved in less than a minute. Under normal
circumstances, there is no performance penalty to decreasing the value of xact
coordination interval.

280

Adaptive Server Enterprise

CHAPTER 5

Setting Configuration Parameters

Setting xact coordination interval to a higher number can slow the completion of
distributed transactions, and cause transaction branches to hold resources
longer than they normally would. Under normal circumstances, do not increase
the value of xact coordination interval beyond its default.

xp_cmdshell context
Summary information
Default value
Valid values

1
0, 1, 2

Status
Display level

Dynamic
Comprehensive

Required role

System administrator

Configuration group

Extended Stored Procedure

xp_cmdshell context sets the security context for the operating system
command to be executed using the xp_cmdshell system ESP. The values for the
context determines under which account the command runs:

0 command runs under XP Servers account.

1 command runs under users account.

2 command runs under XP Servers account only if the user has

administrator privileges.

Setting xp_cmdshell context to 1 restricts the xp_cmdshell security context to

users who have accounts at the operating system level. Its behavior is platformspecific. If xp_cmdshell context is set to 1, to use an xp_cmdshell ESP, an
operating system user account must exist for the Adaptive Server user name.
For example, an Adaptive Server user named sa cannot use xp_cmdshell
unless he or she has an operating-system-level user account named sa.
Starting XP Server as root automatically sets xp_cmdshell to 1 so secure access
is automatically enabled.
On Windows, when xp_cmdshell context is set to 1, xp_cmdshell succeeds only
if the user name of the user logging in to Adaptive Server is a valid Windows
user name with Windows system administration privileges on the system on
which Adaptive Server is running.

System Administration Guide: Volume 1

281

Configuration parameters

On other platforms, when xp_cmdshell context is set to 1, xp_cmdshell succeeds

only if Adaptive Server was started by a user with superuser privileges at the
operating system level. When Adaptive Server gets a request to execute
xp_cmdshell, it checks the uid of the user name of the ESP requestor and runs
the operating system command with the permissions of that uid.
If xp_cmdshell context is 0, the permissions of the operating system account
under which Adaptive Server is running are the permissions used to execute an
operating system command from xp_cmdshell. This allows users to execute
operating commands that they would not ordinarily be able to execute under
the security context of their own operating system accounts.

282

Adaptive Server Enterprise

CH A PTE R

Overview of Disk Resource

Issues

Topic
Device allocation and object placement
Commands for managing disk resources

Page
283
284

Considerations in storage management decisions

Status and defaults at installation time

286
287

System tables that manage storage

288

Many Adaptive Server defaults are set to reasonable values for aspects of
storage management, such as database, table, and index location, and how
much space is allocated for each one. Responsibility for storage allocation
and management is often centralized, and usually, the system
administrator has ultimate control over the allocation of disk resources to
Adaptive Server and the physical placement of databases, tables, and
indexes on those resources.

Device allocation and object placement

When configuring a new system, the system administrator must consider
several issues that have a direct impact on the number and size of disk
resources required. These device allocation issues refer to commands and
procedures that add disk resources to Adaptive Server.
Table 6-1: Device allocation topics
Task
Initialize and allocate a default pool of database devices

Chapter
Chapter 8, Initializing Database Devices

Mirror database devices for recovery

Chapter 2, Mirroring Database Devices, in

System Administration Guide: Volume 2

System Administration Guide: Volume 1

283

Commands for managing disk resources

After the initial disk resources have been allocated to Adaptive Server, the
system administrator, database owner, and object owners should consider how
to place databases and database objects on specific database devices. These
object placement issues determine where database objects reside on your
system and whether or not the objects share devices. Object placement tasks
are discussed throughout this manual, including the chapters shown in Table 62.
Table 6-2: Object placement topics
Task
Place databases on specific database devices
Place tables and indexes on specific database
devices

Chapter

Chapter 6, Creating and Managing User Databases, in

System Administration Guide: Volume 2
Chapter 6, Creating and Managing User Databases, in
System Administration Guide: Volume 2

Do not consider allocating devices separately from object placement. For

example, if you decide that a particular table must reside on a dedicated pair of
devices, first allocate those devices to Adaptive Server. The remaining sections
in this chapter provide an overview that spans both device allocation and object
placement issues, providing pointers to chapters where appropriate.

Commands for managing disk resources

Table 6-3 lists the major commands a system administrator uses to allocate disk
resources to Adaptive Server, and provides references to the chapters that
discuss those commands.
Table 6-3: Commands for allocating disk resources
Command
disk init
name = "dev_name"
physname = "phys_name"...

sp_deviceattr logicalname,
optname, optvalue
sp_diskdefault "dev_name"...
disk resize
name = device_name,
size = additional_space

284

Task

See

Makes a physical device available to a particular

Adaptive Server. Assigns a database device
name (dev_name) that is used to identify the
device in other Adaptive Server commands.
Changes the dsync setting of an existing
database device file.
Adds dev_name to the general pool of default
database space.
Dynamically increases the size of database
devices.

Chapter 8, Initializing
Database Devices

Chapter 8, Initializing
Database Devices
Chapter 8, Initializing
Database Devices
Chapter 8, Initializing
Database Devices

Adaptive Server Enterprise

CHAPTER 6

Command
disk mirror
name = "dev_name"
mirror = "phys_name"...

Overview of Disk Resource Issues

Task

See

Mirrors a database device on a specific physical

device.

Chapter 2, Mirroring
Database Devices, in
System
Administration
Guide: Volume 2

Table 6-4 lists the commands used in object placement. For information about
how object placement affects performance, see Chapter 1, Controlling
Physical Data Placement, in the Performance and Tuning Series: Physical
Database Tuning.
Table 6-4: Commands for placing objects on disk resources
Command
create database...on dev_name

or
alter database...on dev_name

create database...

or
alter database...

sp_addsegment seg_name,
dbname, devname

and

Task

See

Makes database devices available to a particular

Adaptive Server database. The log on clause to
create database places the databases logs on a
particular database device.

Chapter 6,
Creating and
Managing User
Databases, in
System
Administration
Guide: Volume 2
Chapter 6,
Creating and
Managing User
Databases, in
System
Administration
Guide: Volume 2

When used without the on dev_name clause,

these commands allocate space on the default
database devices.

Creates a segmenta named collection of

spacefrom the devices available to a
particular database.

Chapter 8, Creating
and Using Segments

Creates database objects, placing them on a

specific segment of the databases assigned disk
space.

Chapter 8, Creating
and Using Segments

When used without on seg_name, tables and

indexes occupy the general pool of space
allocated to the database (the default devices).

Chapter 8, Creating
and Using Segments

sp_extendsegment seg_name,
dbname, devname
create table...on seg_name

or
create index...on seg_name

create table...

or
create index...

System Administration Guide: Volume 1

in System
Administration
Guide: Volume 2

in System
Administration
Guide: Volume 2

in System
Administration
Guide: Volume 2

285

Considerations in storage management decisions

Considerations in storage management decisions

The system administrator must make many decisions regarding the physical
allocation of space to Adaptive Server databases. The major considerations in
these choices are:

Recovery disk mirroring and maintaining logs on a separate physical

device provide two mechanisms for full recovery in the event of physical
disk failures.

Performance for tables or databases where speed of disk reads and writes
is crucial, properly placing database objects on physical devices yields
performance improvements. Disk mirroring slows the speed of disk
writes.

Recovery
Recovery is the key motivation for using several disk devices. You can mirror
database devices to achieve nonstop recovery. You can also ensure full
recovery by storing a databases log on a separate physical device.

Keeping logs on a separate device

Unless a database device is mirrored, full recovery requires that a databases
transaction log be stored on a different device from the actual data (including
indexes) of a database. In the event of a hard disk failure, you can create an upto-date database by loading a dump of the database and then applying the log
records that were safely stored on another device. See Chapter 6, Creating and
Managing User Databases, in System Administration Guide: Volume 2 for
information about the log on clause of create database.

Mirroring
Nonstop recovery in the event of a hard disk failure is guaranteed by mirroring
all Adaptive Server devices to a separate physical disk. See Chapter 2,
Mirroring Database Devices, in System Administration Guide: Volume 2.

286

Adaptive Server Enterprise

CHAPTER 6

Overview of Disk Resource Issues

Performance
You can improve system performance by placing logs and database objects on
separate devices:

Placing a table on one hard disk and nonclustered indexes on another

ensures that physical reads and writes are faster, since the work is split
between two disk drives.

Splitting large tables across two disks can improve performance,

particularly for multiuser applications.

When log and data share devices, user log cache buffering of transaction
log records is disabled.

Partitioning provides multiple insertion points for a heap table, adds a

degree of parallelism to systems configured to perform parallel query
processing, and makes it possible to distribute a tables I/O across multiple
database devices.

See Chapter 1, Controlling Physical Data Placement, in the Performance and

Tuning Series: Physical Database Tuning for a detailed discussion of how
object placement affects performance.

Status and defaults at installation time

The installation program and scripts initialize the master device and set up the
master, model, sybsystemprocs, sybsecurity, and temporary databases for you.
When you install Adaptive Server, the system databases, system-defined
segments, and database devices are organized as follows:

The master, model, and tempdb databases are installed on the master
device.

The sybsystemprocs database is installed on a device that you specified.

Three segments are created in each database: system, default, and

logsegment.

System Administration Guide: Volume 1

287

System tables that manage storage

The master device is the default storage device for all user-created
databases.
Note After initializing new devices for default storage, remove the master
device from the default storage area with sp_diskdefault. Do not store user
databases and objects on the master device. See Designating default

devices on page 316.

If you install the audit database, sybsecurity, it is located on its own device.

System tables that manage storage

Two system tables in the master database, sysusages and sysdevices, and three
more in each user database (syssegments, sysindexes, and syspartitions) track
the placement of databases, tables (including the transaction log table, syslogs),
and indexes. The relationship between the tables is illustrated in Figure 6-1.

288

Adaptive Server Enterprise

CHAPTER 6

Overview of Disk Resource Issues

Figure 6-1: System tables that manage storage

vstart between low, high

SYSDEVICES
Transactions

SYSUSAGES
Transactions
One row
for each
fragment

Master
database

segmap

User
database

segmap

SYSSEGMENTS
Transactions

1
N
One row for
each data or
index
partition

One row for

each device

1
segment

One row
for each
segment

SYSPARTITIONS
Transactions

SYSINDEXES
Student Info

segment

One row for

each table,
index or
table with
text

The sysdevices table

The sysdevices table in the master database contains one row for each
database device and may contain a row for each dump device (tape, disk, or
operating system file) available to Adaptive Server.

System Administration Guide: Volume 1

289

System tables that manage storage

The disk init command adds entries for database devices to master..sysdevices.
Dump devices, added using sp_addumpdevice, are discussed in Chapter 11,
Developing a Backup and Recovery Plan, in the System Administration
Guide: Volume 2
sysdevices stores two names for each device:

A logical name or device name, used in all subsequent storagemanagement commands, is stored in the name column of sysdevices. This
is usually a user-friendly name, perhaps indicating the planned use for the
device, for example, logdev or userdbdev.

The physical name is the actual operating system name of the device. Use
this name only in the disk init command; after that, all Adaptive Server data
storage commands use the logical name.

Place a database or transaction log on one or more devices by specifying the

logical name of the device in the create database or alter database statement.
The log on clause to create database places a databases transaction log on a
separate device to ensure full recoverability. The log device must also have an
entry in sysdevices before you can use log on.
A database can reside on one or more devices, and a device can store one or
more databases. See Chapter 6, Creating and Managing User Databases, in
System Administration Guide: Volume 2 for information about creating
databases on specific database devices.

The sysusages table

The sysusages table in the master database keeps track of the space you assign
to all Adaptive Server databases.
create database and alter database allocate new space to the database by adding
a row to sysusages for each database device or device fragment. When you
allocate only a portion of the space on a device with create or alter database,
that portion is called a fragment.
sp_addsegment, sp_dropsegment, and sp_extendsegment change the segmap
column in sysusages for the device that is mapped or unmapped to a segment.
See Chapter 8, Creating and Using Segments in System Administration Guide:

Volume 2.

290

Adaptive Server Enterprise

CHAPTER 6

Overview of Disk Resource Issues

The syssegments table

The syssegments table, one in each database, lists the segments in a database.
A segment is a collection of the database devices and fragments available to a
particular database. Tables and indexes can be assigned to a particular
segmentand therefore to a particular physical deviceor can span a set of
physical devices.
create database makes default entries in syssegments. sp_addsegment and
sp_dropsegment to add and remove entries from syssegments.

The sysindexes table

The sysindexes table lists each table and index and the segment where each
table, clustered index, nonclustered index, and chain of text pages is stored. It
also lists other information such as the max_rows_per_page setting for the table
or index.
The create table, create index, and alter table commands create new rows in
sysindexes. Partitioning a table changes the function of sysindexes entries for
the table.

The syspartitions table

The syspartitions table lists each table and index partition and the segment
where the partition is stored. syspartitions maintains key storage management
information such as the first page of a data or index page chain, the last page
of a heap, the root page of an index partition, and so on.
Use create table, create index and alter table to create new rows in syspartitions.

System Administration Guide: Volume 1

291

System tables that manage storage

292

Adaptive Server Enterprise

CH A PTE R

Managing Remote Servers

This chapter discusses the steps the system administrator and system
security officer of each Adaptive Server must execute to enable remote
procedure calls (RPCs).
Topic
Overview

Page
293

Managing remote servers

Adding remote logins

294
300

Password checking for remote users

Getting information about remote logins

304
305

Configuration parameters for remote logins

305

Overview
Users on a local Adaptive Server can execute stored procedures on a
remote Adaptive Server. Executing an RPC sends the results of the remote
process to the calling process, which usually appears on the users screen.
To enable RPCs, the system administrator and system security officer of
each Adaptive Server must execute the following steps:

On the local server:

System security officer use sp_addserver to list the local server

and remote server in the system table master..sysservers.

List the remote server in the interfaces file or directory service

for the local server.

Restart the local server so the global variable @@servername is

set to the name of the local server. If this variable is not set
properly, users cannot execute RPCs from the local server on any
remote server.

On the remote server:

System Administration Guide: Volume 1

293

Managing remote servers

System security officer use sp_addserver to list the server

originating the RPC in the system table master..sysservers.

To allow the user who is originating the remote procedure access to

the server, a system security officer uses create login, and a system
administrator uses sp_addremotelogin.

Add the remote login name as a user of the appropriate database and
grant that login permission to execute the procedure. (If execute
permission is granted to public, the user does not need to be granted
specific permission.)

Figure 7-1 shows how to set up servers for remote access.

Figure 7-1: Setting up servers to allow remote procedure calls

The user joe on ROSE needs to access stored procedures on ZINNIA

ROSE
ZINNIA

sp_addserver ROSE, local

sp_addserver ZINNIA
interfaces files must have an entry
for ZINNIA

sp_addserver ROSE
create login joe
sp_addremotelogin ROSE, joe
sp_adduser joe (in the appropriate database)
grant execute on procedure_name to joe

For operating-system-specific information about handling remote servers, see

the installation documentation for your platform.

Managing remote servers

Table 7-1 lists the tasks related to managing remote servers, and the system
procedures you use to perform the tasks.

294

Adaptive Server Enterprise

CHAPTER 7

Managing Remote Servers

Table 7-1: Tasks related to managing remote servers

To
Add a remote server

Use

Manage remote server names

Change server connection options

sp_addserver

Display information about servers

Drop a server

sp_helpserver

sp_addserver
sp_serveroption
sp_dropserver

See
Adding a remote server on page 295
Managing remote server names on page 296
Setting server connection options on page 297
Getting information about servers on page 299
Dropping remote servers on page 299

Adding a remote server

A system security officer uses sp_addserver to add entries to the sysservers
table. On the server originating the call, you must add one entry for the local
server, and one for each remote server that your server will call.
When you create entries for a remote server, you can either:

Refer to them by the name listed in the interfaces file, or

Provide a local name for the remote server. For example, if the name in the
interfaces file is MAIN_PRODUCTION, you may want to call it simply
main.

The syntax is:

sp_addserver lname [{, local | null}
[, pname]]

where:

lname provides the local call name for the remote server. If this name
is not the same as the remote servers name in the interfaces file, provide
that name as the third parameter, pname.
The remote server must be listed in the interfaces file on the local machine.
If it is not listed, copy the interfaces file entry from the remote server and
append it to your existing interfaces file. Keep the same port numbers.

System Administration Guide: Volume 1

295

Managing remote servers

local identifies the server being added as a local server. The local value is
used only after starting up, or after a restart, to identify the local server
name so that it can appear in messages printed out by Adaptive Server. null
specifies that this server is a remote server.

Note For users to successfully run RPCs from the local server, add the
local server using the local option, and restart it. The restarting is required
to set the global variable @@servername.

pname is the remote server listed in the interfaces file for the server
named lname. This optional argument permits you to establish local
aliases for any other Adaptive Server, Open Server, or Backup Server that
you may need to communicate with. If you do not specify pname, to lname
is the default.

Examples of adding remote servers

This example creates an entry for the local server named DOCS:
sp_addserver DOCS, local

This example creates an entry for a remote server named GATEWAY:

sp_addserver GATEWAY

To run a remote procedure such as sp_who on the GATEWAY server, execute

either:
GATEWAY.sybsytemprocs.dbo.sp_who

or:
GATEWAY...sp_who

This example gives a remote server called MAIN_PRODUCTION the local

alias main:
sp_addserver main, null, MAIN_PRODUCTION

The user can then enter:

main...sp_who

Managing remote server names

The master.dbo.sysservers table has two server name columns:

296

Adaptive Server Enterprise

CHAPTER 7

Managing Remote Servers

srvname is the unique server name that users must supply when executing
remote procedure calls.

srvnetname is the servers network name, which must match the name in
the interfaces file.

To add or drop servers from your network, use sp_addserver to update the
servers network name in srvnetname.
For example, to remove the server MAIN from the network, and move your
remote applications to a server named TEMP, use the following statement to
change the network name, while keeping the local alias:
sp_addserver MAIN, null, TEMP
sp_addserver displays a message telling you that it is changing the network
name of an existing server entry.

Setting server connection options

sp_serveroption sets the server options timeouts, net password encryption, and
the security mechanism, which affect connections with remote servers.

The options you specify for sp_serveroption do not affect communication

between Adaptive Server and Backup Server.
The following sections describe timeouts, net password encryption.

Using the timeouts option

A system administrator can use the timeouts option to disable and enable the
normal timeout code used by the local server.
By default, timeouts is set to true, and the site handler process that manages
remote logins times out if there has been no remote user activity for one
minute. By setting timeouts to false on both of the servers involved in remote
procedure calls, the automatic timeout is disabled. to change timeouts to false
use:
sp_serveroption GATEWAY, "timeouts", false

System Administration Guide: Volume 1

297

Managing remote servers

After you set timeouts to false on both servers, when a user executes an RPC in
either direction, the site handler on each machine runs until one of the servers
is shut down. When the server is brought up again, the option remains false, and
the site handler is reestablished the next time a user executes an RPC. If users
execute RPCs frequently, it is probably efficient in terms of system resources
to set this option to false, since there is some system overhead involved in
setting up the physical connection.

Using the net password encryption option

A system security officer can use net password encryption to specify whether
connections with a remote server are to be initiated with a client-side password
encryption handshake or with the usual unencrypted password handshake
sequence. The default is false.
If net password encryption is set to true:
1

The initial login packet is sent without passwords.

The client indicates to the remote server that encryption is desired.

The remote server returns an encryption key, which the client uses to
encrypt its plain text passwords.

The client then encrypts its own passwords, and the remote server uses the
key to authenticate them when they arrive.

This example sets net password encription to true:

sp_serveroption GATEWAY, "net password encryption",
true

Using the security mechanism option

The security mechanism option enables Kerberos authentication for
connections to the remote server.
Note As of Adaptive Server release 15.7, Kerberos is the security mechanism
supported by Adaptive Server. The rpc security model A and rpc security model
B are no longer required for enabling security mechanism for remote
mechanisms and sp_helpserver will not display the rpc security model. For
backward compatibility, databases that use DCE mechanism upgraded to this
release will continue to work as earlier.

298

Adaptive Server Enterprise

CHAPTER 7

Managing Remote Servers

Getting information about servers

sp_helpserver reports on servers. Without an argument, sp_helpserver provides
information about all the servers listed in sysservers. When you include a
server name, sp_helpserver provides information about that server only:

sp_helpserver [server]
sp_helpserver checks for both srvname and srvnetname in the
master.sysremotelogins table.

For operating-system-specific information about setting up remote servers, see

the installation documentation for your platform.

Dropping remote servers

A system security officer can use sp_dropserver to drop servers from
sysservers:
sp_dropserver server [, droplogins]

where:

server is the name of the server you want to drop.

droplogins allows you to drop a remote server and all of that servers
remote login information. If you do not use droplogins, you cannot drop a

server that has remote logins associated with it.

The following statement drops the GATEWAY server and all of the remote
logins associated with it:
sp_dropserver GATEWAY, droplogins

You do not have to use droplogins to drop the local server; that entry does not
have remote login information associated with it.

System Administration Guide: Volume 1

299

Adding remote logins

Adding remote logins

The system security officer and system administrator of any Adaptive Server
share control over which remote users can access the server, and what identity
the remote users assume. The system administrator uses sp_addremotelogin to
add remote logins and sp_dropremotelogin to drop remote logins. The system
security officer uses sp_remoteoption to control whether password checking is
required.

Mapping users server IDs

Logins from a remote server can be mapped to a local server in three ways:

A particular remote login can be mapped to a particular local login name.

For example, user joe on the remote server might be mapped to
joesmith.

All logins from one remote server can be mapped to one local name. For
example, all users sending remote procedure calls from the MAIN server
might be mapped to remusers.

All logins from one remote server can use their remote names.

The first option can be combined with the other two options, and its specific
mapping takes precedence over the other two more general mappings. The
second and third options are mutually exclusive; you can use either of them,
but not both.
Changing the
mapping option

Use sp_dropremotelogin to remove the old mapping.

Use sp_addremotelogin to add remote logins:
sp_addremotelogin remoteserver [, loginame
[, remotename]]

If the local names are not listed in master..syslogins, use create login to add
them as Adaptive Server logins before you add the remote logins.
Only a system administrator can execute sp_addremotelogin. See the Reference
Manual: Procedures.

300

Adaptive Server Enterprise

CHAPTER 7

Managing Remote Servers

Mapping remote logins to particular local names

The following example maps the login pogo from a remote system to the
local login bob. The user logs in to the remote system as pogo. When
pogo executes remote procedure calls from GATEWAY, the local system
maps the remote login name to bob.
create logn bob with password itsA8secret
sp_addremotelogin GATEWAY, bob, pogo

Mapping all remote logins to one local name

The following example creates an entry that maps all remote login names to the
local name albert. All names are mapped to albert, except those with
specific mappings, as described in the previous section. For example, if you
mapped pogo to bob, and then the rest of the logins to albert, pogo still
maps to bob.
create logn albert with password itsA8secret
sp_addremotelogin GATEWAY, albert

If you use sp_addremotelogin to map all users from a remote server to the same
local name, use sp_remoteoption to specify the trusted option for those users.
For example, if all users from server GATEWAY that are mapped to albert
are to be trusted, specify:
sp_remoteoption GATEWAY, albert, NULL, trusted, true

If you do not specify logins as trusted, they cannot execute RPCs on the local
server unless they specify passwords for the local server when they log in to
the remote server. Users can run ct_remote_pwd to specify a password for
server-to-server connections when they use Open Client Client-Library. isql
and bcp do not permit users to specify a password for RPC connections. See
Password checking for remote users on page 304 for more information about
sp_remoteoption.
Warning! Do not map more than one remote login to a single local login, as it
reduces individual accountability on the server. Audited actions can be traced
only to the local server login, not to the individual logins on the remote server.

System Administration Guide: Volume 1

301

Adding remote logins

If you are using

network-based
security

If users are logged in to the remote server using unified login, the logins must
be designated as trusted on the local server, or they must specify passwords for
the server when they log in to the remote server.
Warning! Using the trusted mode of sp_remoteoption reduces the security of
your server, as passwords from such trusted users are not verified.

Keeping remote login names for local servers

To enable remote users to keep their remote login names while using a local
server:
1

Use create login to create a login for each login from the remote server.

Use sp_addremotelogin for the server to create an entry in

master..sysremotelogins with a null value for the remote login name and a
value of -1 for the suid. For example:
sp_addremotelogin GATEWAY

Example of remote user login mapping

This statement displays the local and remote server information recorded in
master..sysservers:
select
srvid
----0
1
2
3
4

srvid, srvname from sysservers

srvname
---------SALES
CORPORATE
MARKETING
PUBLICATIONS
ENGINEERING

The SALES server is local. The other servers are remote.

This statement displays information about the remote servers and users stored
in master..sysremotelogins:
select remoteserverid, remoteusername, suid
from sysremotelogins
remoteserverid
remoteusername
suid
--------------------------------

302

Adaptive Server Enterprise

CHAPTER 7

1
1
1
3
4

joe
nancy
NULL
NULL
NULL

Managing Remote Servers

1
2
3
4
-1

By matching the value of remoteserverid in this result and the value of srvid in
the previous result, you can find the name of the server for which the
remoteusername is valid. For example, in the first result, srvid 1 indicates the
CORPORATE server; in the second result, remoteserverid 1 indicates that same
server. Therefore, the remote user login names joe and nancy are valid on
the CORPORATE server.
The following statement shows the entries in master..syslogins:
select suid, name from syslogins
suid
name
----------------1
sa
2
vp
3
admin
4
writer

The results of all three queries together show:

The remote user name joe (suid 1) on the remote CORPORATE server
(srvid and remoteserverid 1) is mapped to the sa login (suid 1).

The remote user name nancy (suid 2) on the remote CORPORATE

server (srvid and remoteserverid 1) is mapped to the vp login (suid 2).

The other logins from the CORPORATE server (remoteusername

NULL) are mapped to the admin login (suid 3).

All logins from the PUBLICATIONS server (srvid and remoteserverid 3)

are mapped to the writer login (suid 4).

All logins from the ENGINEERING server (srvid and remoteserverid 4)

are looked up in master..syslogins by their remote user names (suid -1).

There is no remoteserverid entry for the MARKETING server in

sysremotelogins. Therefore, users who log in to the MARKETING server
cannot run remote procedure calls from that server.

System Administration Guide: Volume 1

303

Password checking for remote users

The remote user mapping procedures and the ability to set permissions for
individual stored procedures give you control over which remote users can
access local procedures. For example, you can allow the vp login from the
CORPORATE server to execute certain local procedures and all other logins
from CORPORATE to execute the procedures for which the admin login has
permission.
Note Typically, the passwords for users on the remote server must match

passwords on the local server.

Password checking for remote users

A system security officer can use sp_remoteoption to determine whether
passwords are checked when remote users log in to the local server. By default,
passwords are verified (this is the untrusted mode). In trusted mode, the local
server accepts remote logins from other servers and front-end applications
without user-access verification for the particular login.
When sp_remoteoption is used with arguments, it changes the mode for the
named user:
sp_remoteoption [remoteserver, loginame, remotename,
optname, {true | false}]

For example, to set trusted mode for the user bob, enter
sp_remoteoption GATEWAY, pogo, bob, trusted,
true

Effects of using the untrusted mode

The effects of the untrusted mode depend on the users client program. isql and
some user applications require that logins have the same password on the
remote server and the local server. You can write Open Client applications to
allow local logins to have different passwords on different servers.

304

Adaptive Server Enterprise

CHAPTER 7

Managing Remote Servers

To change your password in untrusted mode, you must first change it on all
the remote systems you access before you can change it on your local server. If
you change your password on the local server first, when you issue the remote
procedure call to execute sp_password on the remote server, your passwords no
longer match.
The syntax for changing your password on the remote server is:
remote_server...sp_password caller_passwd, new_passwd

On the local server, the syntax is:

sp_password caller_passwd, new_passwd

See Chapter 5, Managing Adaptive Server Logins and Database Users, in the
Security Administration Guide.

Getting information about remote logins

sp_helpremotelogin prints information about the remote logins on a server. The
following example shows the remote login pogo mapped locally to login
name bob, with all other remote logins keeping their remote names:

sp_helpremotelogin
server
--------GATEWAY
GATEWAY

remote_user_name
---------------**mapped locally**
pogo

local_user_name
----------------**use local name**
bob

options
-------untrusted
untrusted

Configuration parameters for remote logins

Table 7-2 shows the configuration parameters that affect RPCs. All these
configuration parameters are set using sp_configure and do not take effect until
Adaptive Server is restarted.

System Administration Guide: Volume 1

305

Configuration parameters for remote logins

Table 7-2: Configuration parameters that affect RPCs

Configuration parameter
allow remote access
number of remote logins
number of remote sites
number of remote connections
remote server pre-read packets

Default
1
20
10
20
3

See the individual configuration parameter descriptions in Chapter 5, Setting

Configuration Parameters.

306

Adaptive Server Enterprise

CH A PTE R

Initializing Database Devices

Topic
Database devices

Page
307

Using the disk init command

disk init syntax

308
308

Getting information about devices

Dropping devices

314
316

Designating default devices

Increasing the size of devices with disk resize

316
318

Database devices
A database device stores the objects that make up databases. The term
device does not necessarily refer to a distinct physical device: it can refer
to any piece of a disk (such as a disk partition) or a file in the file system
that is used to store databases and their objects.
Each database device or file must be prepared and made known to
Adaptive Server before it can be used for database storage. This process
is called initialization.
After a database device has been initialized, it can be:

Allocated to the default pool of devices for the create and alter
database commands

Assigned to the pool of space available to a user database

Assigned to a user database and used to store one or more database

objects

Assigned to store a databases transaction logs

System Administration Guide: Volume 1

307

Using the disk init command

Using the disk init command

A system administrator initializes new database devices with the disk init
command, which:

Maps the specified physical disk device or operating system file to a

database device name

Lists the new device in master..sysdevices

Prepares the device for database storage

Note Before you run disk init, see the installation documentation for your
platform for information about choosing a database device and preparing it for
use with Adaptive Server. You may want to repartition the disks on your
computer to provide maximum performance for your Sybase databases.
disk init divides the database devices into allocation units, groups of 256
logical pages. The size of the allocation unit depends on which logical page
size your server is configured for (2, 4, 8, or 16K). In each allocation unit, the
disk init command initializes the first page as the allocation page, which
contains information about the database (if any) that resides on the allocation
unit.

Warning! After you run the disk init command, dump the master database. This
makes recovery easier and safer in case master is damaged. See Chapter 13,
Restoring the System Databases, in System Administration Guide: Volume 2.

disk init syntax

See the Reference Manual: Commands for the disk init syntax.

Specifying a logical device name

The device_name must be a valid identifier. This name is used in the create
database and alter database commands, and in the system procedures that
manage segments. The logical device name is known only to Adaptive Server,
not to the operating system on which the server runs.

308

Adaptive Server Enterprise

CHAPTER 8

Initializing Database Devices

Specifying a physical device name

The physname of the database device gives the name of a raw disk partition
(UNIX), foreign device, or the name of an operating system file. On PC
platforms, you typically use operating system file names for physname.

Choosing a device number

Adaptive Server accepts, but does not require, the disk init vdevno parameter. If
you specify a vdevno, you may choose any currently unused identifier from 1
to 2,147,483,647 (virtual device ID 0 is used by the master device). For
example, specifying vdevno = 33 assigns virtual device ID 33 to a device. If
you do not specify a vdevno, Adaptive Server chooses a number higher than the
highest vdevno currently listed in sysdevices.
The number of database devices you can create is limited by the number of
devices configuration parameter. Adaptive Server is initially configured for 10
devices. Use sp_configure to change the value for number of devices.
Your operating system may also limit the number of devices your installation
can use concurrently. To the operating system, each Sybase device counts as
one open file.
Adaptive Server automatically specifies the next available identifying number
for the database device. This is the virtual device number (vdevno). You need
not specify this number when you issue the disk init command.
If you manually select the vdevno, it must be unique among the devices used
by Adaptive Server. Device number 0 represents the master device. Legal
numbers are 1 2,147,483,647. You can choose any unused vdevno within that
range.
To see the numbers already in use for vdevno, look in the vdevno column of the
report from sp_helpdevice, or use the following query to list all the device
numbers currently in use:
select vdevno from master..sysdevices
where status & 2=2

Here, status & 2=2 specifies physical disk.

System Administration Guide: Volume 1

309

disk init syntax

Specifying the device size

To indicate the size of the device, use k or K for kilobytes, m or M for
megabytes, g or G for gigabytes, and t or T for terabytes. Sybase
recommends that, to avoid confusion in the actual number of pages allocated,
you always include the unit specifier in both the disk init and create database
commands. Enclose the unit specifier in single or double quotes or in brackets.
Theoretically, you can create as many as 2,147,483,647 disk devices, each of
which can be as large as 2,147,483,648 2K-blocks. The maximum installation
size becomes a function of database size, hardware, and operating system
limits.
The following guidelines apply to the syntax for disk init:

If you do not include a unit specifier for the size argument of disk init or
disk reinit, size is measured, by default, in number of virtual pages. Thus,
if you enter size = 15000, Adaptive Server assumes 15,000 virtual
pages. A virtual page is 2048 bytes.

You can increase, but not decrease, the size of an existing database device
using the disk resize command.

If you are planning to use the new device for the creation of a new
database, the minimum size depends on the logical page size used by the
server, described in Table 8-1:
Table 8-1: Minimum database sizes
Logical page size

Minimum database size

2K
4K

3MB
6MB

8K
16K

12MB
24MB

You cannot have a database smaller than the model database. A model database
larger than the minimums listed above, determines the minimum database size.
Adaptive Server allocates and manages database space in allocation units,
which are groups of 256 logical pages. The smallest database you can create
(using create database) is 1MB; therefore, the size of the smallest usable
database device is the larger of 1MB or 256 logical pages (for a 2K or 4K
logical page size, this is 1MB for a 8K logical page size, this is 2MB, for a 16K
logical page size, this is 4MB.

310

Adaptive Server Enterprise

CHAPTER 8

Initializing Database Devices

Keep this grouping of 256 pages in mind when you decide how large to make
a device to avoid wasting space. For example, if your installation uses a 16k
logical page size, specifying a device as size = '31M' leaves 3MB wasted at
the end of the device, since an allocation unit is 4MB.
If you are initializing a raw device, determine the size of the device from your
operating system, as described in the installation guide for your platform. Use
the total size available, up to the maximum for your platform. After you have
initialized the disk for use by Adaptive Server, you cannot use any space on that
raw device for any other purpose.
disk init uses size to compute the value for the high virtual page number in
sysdevices.high. The values for sysdevices.high and sysdevices.low are virtual

page numbers in 2K-byte blocks, which is the Adaptive Server unit of physical
disk management. This may not be the same as your installation's logical page
size
Note If the physical device does not contain the number of blocks specified by
the size parameter, disk init fails. If you use the optional vstart parameter, the
physical device must contain the sum of the blocks specified by both the vstart
and size parameters, or the command fails.

Specifying the dsync setting (optional)

For devices initialized on UNIX operating system files, the dsync setting
controls whether or not writes to those files are buffered. When the dsync
setting is on, Adaptive Server opens a database device file using the UNIX
dsync flag, which ensures that writes to the device file occur directly to the
physical storage media, and that Adaptive Server can recover data on the
device in the event of a system failure.
When dsync is off, writes to the device file may be buffered by the UNIX file
system, and the recovery of data on the device cannot be ensured. Turn off
dsync only when data integrity is not required.
Note The dsync setting is ignored for devices initialized on raw partitions.

Instead, writes to the database device take place directly to the physical media.

System Administration Guide: Volume 1

311

disk init syntax

Performance implications of dsync

Using the dsync setting with database device files incurs several performance
trade-offs:

Adaptive Server does not support asynchronous I/O on operating system

files for HP-UX.

If database device files use the dsync option, the Adaptive Server engine
that is writing to the device file waits until the write operation completes.
This may cause poor performance during update operations.

When dsync is on, write operations to database device files may be slower
compared to earlier versions of Adaptive Server (where dsync is not
supported). This is because Adaptive Server must write data to disk
instead of simply copying cached data to the UNIX file system buffer.
In cases where highest write performance is required (but data integrity
after a system failure is not required) turning dsync off yields device file
performance similar to earlier Adaptive Server versions. For example, you
may consider storing tempdb on a dedicated device file with dsync
disabled, if performance is not acceptable while using dsync.

Response time for read operations is generally better for devices stored on
UNIX operating system files as compared to devices stored on raw
partitions. Data from device files can benefit from the UNIX file system
cache as well as the Adaptive Server cache, and more reads may take place
without requiring physical disk access.

Limitations and restrictions of dsync

The following limitations and restrictions apply to using dsync:

312

For the master device, dsync is always set to true and you cannot change
the setting.

If you change a device files dsync setting using the sp_deviceattr

procedure, restart Adaptive Server for the change to take effect.

When you upgrade from an Adaptive Server earlier than version 12.x,
dsync is set to true only for the master device file. Use sp_deviceattr to
change the dsync setting for any other device files.

Adaptive Server ignores the dsync setting for database devices stored on
raw partitions. Writes to devices stored on raw partitions are always
directly to the physical media.

Adaptive Server Enterprise

CHAPTER 8

Initializing Database Devices

The directio and dsync parameters are mutually exclusive. If a device has
dsync set to true, you cannot set directio to true for the same device. To
enable directio for a device, you must first reset dsync to false.

Using directio to bypass operating system buffer

The directio parameter for disk init, disk reinit, and sp_deviceattr allows you to
configure Adaptive Server to transfer data directly to disk, bypassing the
operating system buffer cache. directio performs I/O in the same manner as raw
devices and provides the same performance benefit as raw devices, but has the
ease of use and manageability of file system devices. You cannot set directio for
the master device. directio is a static parameter; restart Adaptive Server for it to
take effect.
Note directio is not available on all platforms. If you issue disk init with the
directio parameter on a platform on which it is not supported, Adaptive Server
issues the message No such parameter: 'directio'.

By default, the directio option is set to false (off) for all platforms.
Note Devices used for databases for which recovery is not important (for
example, tempdb), may, by default, have dsync set to false. For these devices,
enabling directio may have an adverse performance effect, so carefully review
device use before you enable directio.

This example creates a device named user_disk that uses directio to write data
directly to disk:
disk init
name = "user_disk",
physname = "/usr/u/sybase/data/userfile1.dat",
size = 5120, directio = true

To initializes 10MB of a disk on a UNIX operating system file, enter:

disk reinit
name = "user_disk",
physname = "/usr/u/sybase/data/userfile1.dat",
size = 5120, directio = true

By default, directio is disabled for all existing devices; enable it using

sp_deviceattr:

System Administration Guide: Volume 1

313

Getting information about devices

sp_deviceattr device_name, directio, [true | false]

For example, the following enables directio disk writes for the user_disk
device:
sp_deviceattr user_disk, directio, true

Other optional parameters for disk init

vstart is the starting virtual address, or the offset, for Adaptive Server to begin
using the database device. vstart accepts the following optional unit specifiers:
k or K (kilobytes), m or M (megabytes), g or G (gigabytes) and t or T
(terabytes). The size of the offset depends on how you enter the value for vstart:

If you do not specify a unit size, vstart uses 2K pages for its starting
address. For example, if you specify vstart = 13, Adaptive Server uses
13 * 2K pages as the offset for the starting address.

If you specify a unit value, vstart uses the value you enter as the starting
address. For example, if you specify vstart = "13M", Adaptive Server
sets the starting address offset at 13MB.

The default, and usually preferred value, of vstart is 0. If the specified device
does not have the sum of vstart + size blocks available, the disk init command
fails.
The optional cntrltype keyword specifies the disk controller. Its default value is
0. Reset it only if instructed to do so by your system administrator.
Note To perform disk initialization, the user who started Adaptive Server must

have the appropriate operating system permissions on the device that is being
initialized.

Getting information about devices

sp_helpdevice provides information about the devices in the sysdevices table.

When used without a device name, sp_helpdevice lists all the devices available
on Adaptive Server. When used with a device name, it lists information about
that device. Here, sp_helpdevice is used to report information about the master
device:

314

Adaptive Server Enterprise

CHAPTER 8

device_name
----------master

status
-----3

Initializing Database Devices

sp_helpdevice master
physical_name description
-------------- -----------------------------------------d_master
special, default disk, physical disk, 30 MB

cntrltype
---------0

vdevno
------0

vpn_low
------0

vpn_high
--------10239

Each row in master..sysdevices describes:

A dump device (tape, disk, or file) to be used for backing up databases, or

A database device to be used for database storage.

The initial contents of sysdevices are operating-system-dependent. sysdevices

entries usually include:

One for the master device

One for the sybsystemprocs database, which you can use to store
additional databases such as pubs2 and sybsyntax, or for user databases
and logs

Two for tape dump devices

If you installed auditing, there is a separate device for sybsecurity.

The vpn_low and vpn_high columns represent the page numbers that have been
assigned to the device. For dump devices, these columns represent the media
capacity of the device.
The status field indicates the type of device, whether a disk device is used as a
default storage device when users issue a create or alter database command
without specifying a database device, disk mirroring information, and dsync
settings.
Table 8-2: Status bits in sysdevices
Bit
1

Meaning
Default disk (may be used by any create or alter database command that does not specify a location)

2
4

Physical disk
Logical disk (not used)

8
16

Skip header (used with tape dump devices)

Dump device

32
64

Serial writes
Device mirrored

System Administration Guide: Volume 1

315

Dropping devices

Bit

Meaning

128
256

Reads mirrored
Secondary mirror side only

512
2048

Mirror enabled
Used internally; set after disk unmirror, side = retain

4096
8192

Primary device needs to be unmirrored (used internally)

Secondary device needs to be unmirrored (used internally)

16384

UNIX file device uses dsync setting (writes occur directly to physical media)

For more information about dump devices and sp_addumpdevice, See Chapter
11, Developing a Backup and Recovery Plan in the System Administration
Guide: Volume 2.

Dropping devices
To drop database and dump devices, use sp_dropdevice:
sp_dropdevice logicalname

You cannot drop a device that is in use by a database. You must drop the
database first.
sp_dropdevice removes the device name from sysdevices. sp_dropdevice does

not remove an operating system file; it only makes the file inaccessible to
Adaptive Server. Use operating system commands to delete a file after using
sp_dropdevice.

Designating default devices

To create a pool of default database devices to be used by all Adaptive Server
users for creating databases, use sp_diskdefault after the devices are initialized.
sp_diskdefault marks these devices in sysdevices as default devices. Whenever
users create (or alter) databases without specifying a database device, new disk
space is allocated from the pool of default disk space.
The syntax for sp_diskdefault is:
sp_diskdefault logicalname, {defaulton | defaultoff}

316

Adaptive Server Enterprise

CHAPTER 8

Initializing Database Devices

After adding user devices, use the defaultoff option to remove the master device
from the pool of default space:
sp_diskdefault master, defaultoff

The following designates sprocdev, the device that holds the sybsystemprocs
database, a default device:
sp_diskdefault sprocdev, defaulton

Adaptive Server can have multiple default devices. They are used in the order
in which they appear in the sysdevices table (that is, alphabetical order). When
the first default device is filled, the second default device is used, and so on.
Note After initializing a set of database devices, you may want to assign them

to specific databases or database objects rather than adding them to the default
pool of devices. For example, you may want to make sure a table never grows
beyond the size of a particular device.

Choosing default and nondefault devices

sp_diskdefault lets you plan space usage for performance and recovery, while
allowing users to create or alter databases.

Do not use these devices as default devices:

The master device

The device used for sybsecurity

Any device intended solely for logs

Devices where high-performance databases reside

You can use the device that holds sybsystemprocs for other user databases.
Note If you are using disk mirroring or segments, exercise caution in deciding
which devices you add to the default list. In most cases, devices that are to be
mirrored, or databases that contain objects placed on segments should
specifically allocate devices, rather than being made part of default storage.

System Administration Guide: Volume 1

317

Increasing the size of devices with disk resize

Increasing the size of devices with disk resize

The disk resize command allows you to increase the size of your database
devices dynamically, rather than initializing a new device. For example, if
/sybase/testdev.dat requires an additional 10MB of space, you can run disk
resize and allocate this amount of space to the device. The create and alter
database commands can use this added space.
Use disk resize to increase the size for both devices on raw partitions and for
file systems. The minimum amount of space by which you can increase a
device is 1MB or an allocation unit, whichever is greater.
Page size

Allocation unit size

Minimum incremental size

2K
4K

0.5MB
1MB

1MB
1MB

8K
16K

2MB
4MB

2MB
4MB

You cannot use disk resize on dump or load devices.

Any properties that are set on the device continue to be set after you increase
its size. That is, if a device has dsync set before you increase its size, it has
dsync set afterwards. Also, any access rights that were set before you increased
the size of the device remain set.
A user with the sa_role can execute the disk resize command, which:

Updates the high value in master....sysdevices, and

Prepares the additional space for database storage.

Use audit trails on disk resize to track the number of times a device is resized.
The device being resized is always online and available for users during the
resize operation.
Resizing a disk requires that:

You have already initialized the device with disk init.

device_name must refer to a valid logical device name.

You disable mirroring while the resize operation is in progress. You can
reestablish mirroring when the resize operation is complete.

In this example, the configuration of the device testdev is:

sp_helpdevice testdev
device_name physical_name
status cntrltype vdevno

318

description
vpn_low

vpn_high

Adaptive Server Enterprise

CHAPTER 8

Initializing Database Devices

----------- ------------------------------------- --------- -------------------------------------testdev

/sybase/dev/testdev.dat
special, dsync on, directio off,
physical disk, 10.00MB
16386
0
1
0
5119

To increase the size of testdev by 4MB using disk resize, enter:

disk resize
name = "test_dev",
size = "4M"

testdev.dat is now 14MB:

sp_helpdevice testdev
device_name physical_name
description
status cntrltype vdevno
vpn_low
vpn_high
----------- ------------------------------------- --------- -------------------------------------testdev
/sybase/dev/testdev.dat
special, dsync on, directio off,
physical disk, 14.00MB
16386
0
1
0
7167

See the Reference Manual: Commands for disk resize syntax.

Insufficient disk space

During the physical initialization of the disk, if an error occurs due to
insufficient disk space, disk resize extends the database device to the largest
size possible before the error occurs.
For example, on a server that uses 4K logical pages, if you try to increase the
size of the device by 40MB, but only 39.5MB is available, the device is
extended only by 39.5MB.
You cannot decrease the size of a device with disk resize.

System Administration Guide: Volume 1

319

Increasing the size of devices with disk resize

320

Adaptive Server Enterprise

CH A PTE R

Setting Database Options

Topic
Using the sp_dboption procedure

Page
321

Database option descriptions

Viewing the options on a database

322
323

Database options control:

Transaction behavior

Table-colum defaults

User access restrictions

Performance of recovery and bcp operations

Log behavior

The system administrator and the database owner can use database options
to configure settings for an entire database. Database options differ from
sp_configure parameters, which affect the entire server, and set options,
which affect only the current session or stored procedure.

Using the sp_dboption procedure

Use sp_dboption to change settings for an entire database. The options
remain in effect until they are changed. sp_dboption:

Displays a complete list of the database options when it is used

without a parameter

Changes a database option when used with parameters

You can change options only for user databases. You cannot change
options for the master database. To change a database option in a user
database (or to display a list of the database options), execute sp_dboption
while using the master database.

System Administration Guide: Volume 1

321

Database option descriptions

The syntax is:

sp_dboption [dbname, optname, {true | false}]
Note Changes to models database options do not affect tempdb or current
user-defined multiple temporary databases when you restart Adaptive
Server. These changes appear only in databases that you create after you
change the model database. Restarting Adaptive Server clears objects and
data contained in the temporary databases, but does not reset database
options.

Database option descriptions

All users with access to the master database can execute sp_dboption with
no parameters to display a list of the database options. The report from
sp_dboption looks like this:
sp_dboption
Settable database options.
-------------------abort tran on log full
allow nulls by default
allow wide dol rows
async log service
auto identity
dbo use only
ddl in tran
delayed commit
enforce dump tran sequence
full logging for all
full logging for alter table
full logging for reorg rebuild
full logging for select into
identity in nonunique index
no chkpt on recovery
no free space acctg
read only
select into/bulkcopy/pllsort
single user
trunc log on chkpt
trunc. log on chkpt.
unique auto_identity index

322

Adaptive Server Enterprise

CHAPTER 9

Setting Database Options

For a report on which options have been set in a particular database,

execute sp_helpdb in that database.
See the Commands Reference: Procedures for information about each
database option in detail.

Viewing the options on a database

Use sp_helpdb to determine the options that are set for a particular
database. sp_helpdb lists each active option in the status column of its
output.
The following example shows that the read only option is turned on in
mydb:

sp_helpdb mydb
name
----mydb

db_size
------20.0 MB

device_fragments
----------------master

owner
----sa

dbid
---5

created
-----------Mar 05, 2005

size
usage
------ -----------10.0 MB data and log

device
-----------------------------master
master
master

status
------------------read only

created
-------Mar 05 2005

free kbytes
----------1792

segment
-----------------------------default
logsegment
system

To display a summary of the options for all databases, use sp_helpdb

without specifying a database:
sp_helpdb
name
db_size owner dbid
created
status
-------------------- ----- --------------------------------master
48.0 MB
sa
1
Apr 12, 2005
mixed log and data
model
8.0 MB
sa
3
Apr 12, 2005
mixed log and data
pubs2
20.0 MB
sa
6
Apr 12, 2005
select into/
bulkcopy/pllsort, trunc log on chkpt, mixed log and data
sybsystemdb
8.0 MB
sa
5
Apr 12, 2005
mixed log and data
sybsystemprocs 112.0 MB
sa
4
Apr 12, 2005
trunc log on chkpt,
mixed log and data
tempdb
8.0 MB
sa
2
Apr 12, 2005
select into/

System Administration Guide: Volume 1

323

Displaying currently set switches with sysoptions

bulkcopy/pllsort, trunc log on chkpt, mixed log and data

Displaying currently set switches with sysoptions

The sysoptions.number column contains the switch ID for currently set
switches. sysoptions shows these switches:

Trace flag set in the runserver file with the -T flag

Trace flag set with dbcc traceon(flag_number) or set switch serverwide
on

Trace flags and switches set for a specific system process ID (spid)
with set switch on

sysoptions shows only the switches that are visible to the user. Users
cannot see switches set privately by other spids. The value for number is
Null for all option categories other than switches.
sysoptions displays this information for switches:

spid spid for the current session.

name name of the switch. If an unnamed switch is set, name

contains the number of the switch converted to a string.

category specifies the string Switch.

currentsetting set to 1 if the switch is set, 0 if the switch is not set.

defaultsetting contains 0.

scope indicates whether the switch is set server-wide or for the

session. Values are:

0 switch is not set.

8 switch is set server-wide.

16 switch is set privately to the current spid.

24 switch is set server-wide and privately.

number contains the switch ID as an integer.

This query displays all sysoptions entries for switches:

select * from sysoptions

324

Adaptive Server Enterprise

CHAPTER 9

Setting Database Options

where category = 'Switch'

This query displays the switches set for the current session:
select * from sysoptions
where category = 'Switch'
and currentsetting = '1'

This shows the sysoptions output after setting trace flag 3604:
spid

name
defaultsetting
scope
----------------------------------------------------------11
print_output_to_client
0
8

System Administration Guide: Volume 1

category
currentsetting
number
--------------- ---------------------------Switch
1
3604

325

Displaying currently set switches with sysoptions

326

Adaptive Server Enterprise

CH A PTE R

1 0

Configuring Character Sets, Sort

Orders, and Languages

This chapter discusses Adaptive Server internationalization and

localization support issues.
Topic
Understanding internationalization and localization
Advantages of internationalized systems

Page
327
328

A sample internationalized system

Elements of an internationalized system

329
331

Selecting the character set for your server

Selecting the sort order

331
341

Selecting a language for system messages

Setting up your server: examples

349
350

Changing the character set, sort order, or message language

352

Installing date strings for unsupported languages

Internationalization and localization files

363
365

Understanding internationalization and localization

Internationalization is the process of enabling an application to support
multiple languages and cultural conventions.
An internationalized application uses external files to provide languagespecific information at execution time. Because it contains no languagespecific code, an internationalized application can be deployed in any
native language environment without code changes. A single version of a
software product can be adapted to different languages or regions,
conforming to local requirements and customs without engineering
changes. This approach to software development saves significant time
and money over the lifetime of an application.

System Administration Guide: Volume 1

327

Advantages of internationalized systems

Localization is the process of adapting an internationalized product to meet the

requirements of one particular language or region, for example Spanish,
including providing translated system messages; translations for the user
interface; and the correct formats for date, time, and currency. One version of
a software product may have many localized versions.
Sybase provides both internationalization and localization support. Adaptive
Server includes the character set definition files and sort order definition files
required for data processing support for the major business languages in
Western Europe, Eastern Europe, the Middle East, Latin America, and Asia.
Sybase Language Modules provide translated system messages and formats for
Chinese (Simplified), French, German, Japanese, Korean, Brazilian
Portuguese, and Spanish. By default, Adaptive Server comes with U.S. English
message files.
This chapter describes the available character sets and language modules and
summarizes the steps necessary to change the default character set, sort order,
or message language for Adaptive Server.

Advantages of internationalized systems

The task of designing an application to work outside its country of origin can
seem daunting. Often, programmers think that internationalizing means hardcoding dependencies based on cultural and linguistic conventions for just one
country.
A better approach is to write an internationalized application: that is, one that
examines the local computing environment to determine what language to use
and loads files containing language-specific information at runtime.
When you use an internationalized application, a single application can be
deployed in all countries. This has several advantages:

328

You write and maintain one application.

The application can be deployed, without change, in new countries as

needed. You need only supply the correct localization files.

All sites can expect standard features and behavior.

Adaptive Server Enterprise

CHAPTER 10

Configuring Character Sets, Sort Orders, and Languages

A sample internationalized system

An internationalized system may include internationalized client applications,
gateways, and servers running on different platforms in different native
language environments.
For example, an international system might include the following components:

Order processing applications in New York City, Mexico City, and Paris
(Client-Library applications)

An inventory control server in Germany (Adaptive Server)

An order fulfillment server in France (Adaptive Server)

A central accounting application in Japan (an Open Server application

working with an Adaptive Server)

In this system, the order processing applications:

Query the inventory control server to determine if requested items are in

stock

Place orders with the order fulfillment server

Send financial information to the accounting application

The inventory control server and the order fulfillment server respond to
queries, and the accounting application collects financial data and generates
reports.
The system looks like this:

System Administration Guide: Volume 1

329

A sample internationalized system

Figure 10-1: Example of an international system

Order processing
New York City (PC)

Order fulfillment
France

English

Order processing
Mexico City (UNIX)

Inventory control
Germany

Spanish

Order processing
Paris (Mac)

Accounting gateway
Japan

Account server
Japan

French

In this example, all applications and servers use local languages and character
sets to accept input and output messages.

330

Adaptive Server Enterprise

CHAPTER 10

Configuring Character Sets, Sort Orders, and Languages

Elements of an internationalized system

There are three elements that you can manipulate to configure your server
language in an internationalized environment. Sybase suggests that you review
these three elements and carefully plan the client/server network you want to
create.

Character set the language in which the server sends and receives data to
and from the client servers. Select the character set after carefully planning
and analyzing the language needs of all client servers.

Sort order sort order options are dependent on the language and character
set you select.

System messages messages display in one of several languages provided

by Sybase. If your server language is not one of the languages provided,
your system messages display in English, the default.

The following sections provide details about each of these elements.

Selecting the character set for your server

All data is encoded in your server in a special code. For example, the letter a
is encoded as 97 in decimal. A character set is a specific collection of
characters (including alphabetic and numeric characters, symbols, and
nonprinting control characters) and their assigned numerical values, or codes.
A character set generally contains the characters for an alphabet, for example,
the Latin alphabet used in the English language, or a script such as Cyrillic used
with languages such as Russian, Serbian, and Bulgarian. Character sets that are
platform-specific and support a subset of languages, for example, the Western
European languages, are called native or national character sets. All
character sets that come with Adaptive Server, except for Unicode UTF-8, are
native character sets.

System Administration Guide: Volume 1

331

Selecting the character set for your server

A script is a writing system, a collection of all the elements that characterize

the written form of a human languagefor example, Latin, Japanese, or
Arabic. Depending on the languages supported by an alphabet or script, a
character set can support one or more languages. For example, the Latin
alphabet supports the languages of Western Europe (see Group 1 in Table 101 on page 333). On the other hand, the Japanese script supports only one
language, Japanese. Therefore, the Group 1 character sets support multiple
languages, while many character sets, such as those in Group 101, support only
one language.
The language or languages that are covered by a character set is called a
language group. A language group can contain many languages or only one
language; a native character set is the platform-specific encoding of the
characters for the language or languages of a particular language group.
Within a client/server network, you can support data processing in multiple
languages if all the languages belong to the same language group (see
Table 10-1 on page 333). For example, if data in the server is encoded in a
Group 1 character set, you could have French, German, and Italian data and any
of the other Group 1 languages in the same database. However, you cannot
store data from another language group in the same database. For example, you
cannot store Japanese data with French or German data.
Unlike the native character sets just described, Unicode is an international
character set that supports over 650 of the worlds languages, such as Japanese,
Chinese, Russian, French, and German. Unicode allows you to mix different
languages from different language groups in the same server, no matter what
the platform. See Unicode on page 334 for more information.
Since all character sets support the Latin script, and therefore English, a
character set always supports at least two languagesEnglish and one other
language.
Many languages are supported by more than one character set. The character
set you install for a language depends on the clients platform and operating
system.
Adaptive Server supports the following languages and character sets:

332

Adaptive Server Enterprise

CHAPTER 10

Configuring Character Sets, Sort Orders, and Languages

Table 10-1: Supported languages and character sets

Language
group
Group 1

Group 2

Group 4
Group 5

Languages

Character sets

Albanian, Catalan,
Danish, Dutch, English, Faeroese, Finnish,
French, Galician, German, Icelandic, Irish,
Italian, Norwegian, Portuguese, Spanish,
Swedish
Eastern European: Croatian, Czech, Estonian,
Hungarian, Latvian, Lithuanian, Polish,
Romanian, Slovak, Slovene (and English)
Baltic (and English)
Cyrillic: Bulgarian, Byelorussian, Macedonian,
Russian, Serbian, Ukrainian (and English)

ASCII 8, CP 437, CP 850, CP 860, CP 863,

CP 1252a , ISO 8859-1, ISO 8859-15,
Macintosh Roman, ROMAN8, ROMAN9, ISO15, CP 858

Western European:

CP 852, CP 1250, ISO 8859-2, Macintosh

Central European
CP 1257
CP 855, CP 866, CP 1251, ISO 8859-5, Koi8,
Macintosh Cyrillic
CP 864, CP 1256, ISO 8859-6

Group 6

Arabic (and English)

Group 7

Greek (and English)

CP 869, CP 1253, GREEK8, ISO 8859-7,

Macintosh Greek

Group 8
Group 9

Hebrew (and English)

Turkish (and English)

CP 1255, ISO 8859-8

CP 857, CP 1254, ISO 8859-9, Macintosh
Turkish, TURKISH8

Group 101
Group 102

Japanese (and English)

Simplified Chinese (PRC) (and English)

CP 932 DEC Kanji, EUC-JIS, Shift-JIS

CP 936, EUC-GB, GB18030

Group 103
Group 104

Traditional Chinese (ROC) (and English)

Korean (and English)

Big 5, CP 950b , EUC-CNS, Big 5 HKSCS

EUC-KSC, cp949

Group 105
Group 106

Thai (and English)

Vietnamese (and English)

CP 874, TIS 620

CP 1258

Unicode

Over 650 languages

UTF-8

a. CP 1252 is identical to ISO 8859-1 except for the 0x800x9F code points which are mapped to characters in CP 1252.
b. CP 950 is identical to Big 5.

Note The English language is supported by all character sets because the first
128 (decimal) characters of any character set include the Latin alphabet
(defined as ASCll-7). The characters beyond the first 128 differ between
character sets and are used to support the characters in different native
languages. For example, code points 0-127 of CP 932 and CP 874 both support
English and the Latin alphabet. However, code points 128-255 support
Japanese characters in CP 932 and code points 128-255 support Thai characters
in CP 874.

System Administration Guide: Volume 1

333

Selecting the character set for your server

The following character sets support the European currency symbol, the
euro: CP 1252 (Western Europe); CP 1250 (Eastern Europe); CP 1251
(Cyrillic); CP 1256 (Arabic); CP 1253 (Greek); CP 1255 (Hebrew); CP 1254
(Turkish); CP 874 (Thai); iso15, roman9 and CP858. Unicode UTF-8 also
supports:

Traditional Chinese on the Windows and Solaris platforms

Arabic, Hebrew, Thai, and Russian on the Linux platform

Note iso_1 and ISO 8859-1 are different names for the same character set.

To mix languages from different language groups you must use Unicode. If
your server character set is Unicode, you can support more than 650 languages
in a single server and mix languages from any language group.

Unicode
Unicode is the first character set that enables all the worlds languages to be
encoded in the same data set. Prior to the introduction of Unicode, if you
wanted to store data in, for example, Chinese, you had to choose a character set
appropriate for that languageto the exclusion of most other languages. It was
either impossible or impractical to mix character sets, and thus diverse
languages, in the same data set.
Sybase supported Unicode in the form of three datatypes: unichar, univarchar,
and unitext. These datatypes store data in the UTF-16 encoding of Unicode.
UTF-16 is an encoding wherein Unicode scalar values are represented by a
single 16-bit value (or, in rare cases, as a pair of 16-bit values). The three
encodings are equivalent insofar as either encoding can be used to represent
any Unicode character. The choice of UTF-16 datatypes, rather than a UTF-16
server default character set, promotes easy, step-wise migration for existing
database applications.
Adaptive Server supports Unicode literals in SQL queries and a wide range of
sort orders for UTF-8.

334

Adaptive Server Enterprise

CHAPTER 10

Configuring Character Sets, Sort Orders, and Languages

The character set model used by Adaptive Server is based on a single,

configurable, server-wide character set. All data stored in Adaptive Server,
using any of the character datatypes (char, varchar, nchar, nvarchar, and text),
is interpreted as being in this character set. Sort orders are defined using this
character set, as are language modulescollections of server messages
translated into local languages.
During the connection dialog, a client application declares its native character
set and language. If properly configured, the server thereafter attempts to
convert any character data between its own character set and that of the client
(character data includes any data stored in the database, as well as server
messages in the clients native language).This works well as long as the
servers and clients character sets are compatible. It does not work well when
characters are not defined in the other character set, as is the case for the
character sets SJIS, used for Japanese, and KOI8, used for Russian and other
Cyrillic languages. Such incompatibilities are the reason for Unicode, which
can be thought of as a character superset, including definitions for characters
in all other character sets.
The Unicode datatypes unichar, univarchar, and unitext are completely
independent of the traditional character set model. Clients send and receive
Unicode data independently of whatever other character data they send and
receive.

Character set installation

Adaptive Server version 12.5.1 and later supports the 4-byte form of UTF-8.
This form is used to represent the same rare Unicode characters that are
represented in UTF-16 by pairs of 16-bit values (surrogate pairs). Prior to
Adaptive Server version 12.5.1, only the 3-byte forms of UTF-8 were
supported. If you have installed the UTF-8 character set in an Adaptive Server
server earlier than version 12.5.1, you should reinstall it to enable the use of the
4-byte form of UTF-8.

Configuration parameters
The UTF-16 encoding of Unicode includes surrogate pairs, which are pairs
of 16-bit values that represent infrequently used characters. Additional
checking is built in to Adaptive Server to ensure the integrity of surrogate pairs.
You can switch this checking off by setting the configuration parameter
enable surrogate processing to 0. This yields slightly higher performance,
although the integrity of surrogate pairs is no longer guaranteed.

System Administration Guide: Volume 1

335

Selecting the character set for your server

Unicode also defines normalization, which is the process by which all

possible representations of a single character are transformed into a single
representation. Many base characters followed by combining diacritical marks
are equivalent to precomposed characters, although their bit patterns are
different. For example, the following two sequences are equivalent:
0x00E9

-- (LATIN SMALL LETTER E WITH ACUTE)

0x00650301

-- e (LATIN SMALL LETTER E), (COMBINING ACUTE ACCENT)

The enable unicode normalization configuration parameter controls whether or

not Adaptive Server normalizes incoming Unicode data.
Significant performance increases are possible when the default Unicode
sortorder is set to binary and the enable Unicode normalization configuration
parameter is set to 1. This combination allows Adaptive Server to make several
assumptions about the nature of the Unicode data, and code has been
implemented to take advantage of these assumptions.

Functions
All functions taking char parameters have been overloaded to accept unichar as
well. Functions with more than one parameter, when called with at least one
unichar parameter, results in implicit conversion of any non-unichar parameters
to unichar.
To guarantee the integrity of surrogate pairs when enable surrogate processing
is set to 1 (the default), the string functions do not allow surrogate pairs to be
split. Positions are modified to fall at the beginning of a surrogate pair.
Several functions have been added to round out the unichar support. Included
are the functions to_unichar() and uscalar(), which are analogous to char() and
ascii(). The functions uhighsurr() and ulowsurr() allow the explicit handling of
surrogate pairs in user code.
There are restrictions when using unitext with functions. For information, see
the restriction description under the Usage section for each function.

Using unichar columns

When using the isql or bcp utilities, Unicode values display in hexadecimal
form unless the -Jutf8 flag is used, indicating the clients character set is UTF-8.
In this case, the utility converts any Unicode data it receives from the server
into UTF-8. For example:
% isql -Usa -P -Jiso_1

336

Adaptive Server Enterprise

CHAPTER 10

Configuring Character Sets, Sort Orders, and Languages

1> select unicode_name from people where unicode_name = 'Jones'

2> go
unicode_name
------------------------------------------------------------------|
0x004a006f006e00650073
(1 row affected)

whereas:
% isql -Usa -P -Jutf8
1> select unicode_name from people where unicode_name = 'Jones'
2> go
unicode_name
-----------------------------------------------------------------Jones
(1 row affected)

This facilitates ad hoc queries. Not all terminal windows are capable of
displaying the full repertoire of Unicode characters, but simple tests involving
ASCII characters are greatly simplified.

Using unitext
The variable-length unitext datatype can hold up to 1,073,741,823 Unicode
characters (2,147,483,646 bytes). You can use unitext anywhere you use the text
datatype, with the same semantics. unitext columns are stored in UTF-16
encoding, regardless of the Adaptive Server default character set.

Open Client interoperability

The Open Client libraries support the datatype cs_unichar, which can be bound
to user variables declared as an array of short integers. This Open Client
datatype interfaces directly with the servers unichar, unitext, and univarchar.

Java interoperability
The internal JDBC driver efficiently transfers unichar data between SQL and
Java contexts.
Going from SQL to Java, the class java.sql.ResultSet provides a number of
get methods to retrieve data from the columns of a result set. Any of these
get methods work with columns defined as unichar, unitext, or univarchar. The
method getString() is particularly efficient since no conversion needs to be
performed.

System Administration Guide: Volume 1

337

Selecting the character set for your server

Use the setString() method of the class java.sql.PreparedStatement to go from

Java to SQL. The internal JDBC driver copies Java string data directly into the
SQL parameter defined as unichar, unitext, or univarchar.
The external JDBC driver (jConnect) has been modified to support the same
seamless interface as the internal driver.

Limitations
Because the earlier releases of Adaptive Server did not include a
Unicode-based language parser, a restriction was imposed on the use of the
new Unicode datatypes. To use the new datatypes, the server required its
default character set to be configured as UTF-8. This restriction has been
removed in Adaptive Server release 12.5.1 and later. Unicode datatypes can be
used regardless of the servers default character set.

Selecting the server default character set

When you configure your server, you must specify a default character set for
the server. The default character set is the character set in which the server
stores and manipulates data. Each server can have only one default character
set.
By default, the installation tool assumes that the native character set of the
platform operating system is the servers default character set. However, you
can select any character set supported by Adaptive Server as the default on
your server (see Table 10-1 on page 333).
For example, if you are installing the server on IBM RS/6000 running AIX, and
you select one of the Western European languages to install, the installation
tool assumes the default character set to be ISO 8859-1.
If you are installing a Unicode server, select UTF8 as your default character
set.
For non-Unicode servers, determine what platform most of your client systems
use and use the character set for this platform as the default character set on the
server.
This has two advantages:

338

The number of unmappable characters between character sets is

minimized.

Adaptive Server Enterprise

CHAPTER 10

Configuring Character Sets, Sort Orders, and Languages

Since there is usually not a complete one-to-one mapping between the

characters in two character sets, there is a potential for some data loss. This
is usually minor because most unconverted characters are special symbols
that are not commonly used or are specific to a platform.

This minimizes the character set conversion that is required.

When the character set on the client system differs from the default
character set on the server, data must be converted in order to ensure data
integrity. Although the measured performance decrease that results from
character set conversion is insignificant, it is good practice to select the
default character set that results in the fewest conversions.

For example, if most of your clients use CP 850, specify CP 850 on your server.
You can do this even if your server is on an HP-UX system (where its native
character set for the Group 1 languages is ROMAN8).
Note Sybase strongly recommends that you decide which character set to use
as your default before you create any databases or make any changes to the
Sybase-supplied databases.

In the example below (Figure 10-2), 175 clients all access the same Adaptive
Server. The clients are on different platforms and use different character sets.
The critical factor that allows these clients to function together is that all of the
character sets in the client/server system belong to the same language group
(see Table 10-1 on page 333). The default language for the Adaptive Server is
CP 850, which is the character set used by the largest number of clients. This
allows the server to operate most efficiently, with the least amount of character
set conversion.

System Administration Guide: Volume 1

339

Selecting the character set for your server

Figure 10-2: Clients using different character sets in the same language
group

CP 850

100 Clients
ASE
CP 850

ISO 8859-1

50 Clients
Macintosh Roman

25 Clients
To help you choose the default character set for your server, the following
tables list the most commonly used character sets by platform and language.

340

Adaptive Server Enterprise

CHAPTER 10

Configuring Character Sets, Sort Orders, and Languages

Table 10-2: Popular Western European client platforms

Platform
Win 95, 98

Language
U.S. English, Western Europe

Character set
CP 1252

Win NT 4.0
Win 2000

U.S. English, Western Europe

U.S. English, Western Europe

CP 1252
CP 1252

Sun Solaris
HP-UX 10,11

U.S. English, Western Europe

U.S. English, Western Europe

ISO 8859-1
ROMAN8

IBM AIX 4.x

U.S. English, Western Europe

ISO 8859-1

Table 10-3: Popular Japanese client platforms

Platform
Win 95, 98

Language
Japanese

Character set
CP 932 for Windows

Win NT 4.0

Japanese

CP 932 for Windows

Win 2000
Sun Solaris

Japanese
Japanese

CP 932 for Windows

EUC-JIS

HP-UX 10,11
IBM AIX 4.x

Japanese
Japanese

EUC-JIS
EUC-JIS

Table 10-4: Popular Chinese client platforms

Platform

Language

Character set

Win 95, 98
Win NT 4.0

Chinese (simplified)
Chinese (simplified)

CP 936 for Windows

CP 936 for Windows

Win 2000
Sun Solaris

Chinese (simplified)
Chinese (simplified)

CP 936 for Windows

EUC-GB

HP-UX 10,11
IBM AIX 4.x

Chinese (simplified)
Chinese (simplified)

EUC-GBS
EUC-GB

Selecting the sort order

Different languages sort the same characters differently. For example, in
English, Cho would be sorted before Co, whereas in Spanish, the opposite is
true. In German, is a single character, however in dictionaries it is treated as
the double character ss and sorted accordingly. Accented characters are sorted
in a particular order so that amnit comes before amne, whereas if you
ignored the accents, the reverse would be true. Therefore, language-specific
sort orders are required so that characters are sorted correctly.

System Administration Guide: Volume 1

341

Selecting the sort order

Each character set comes with one or more sort orders that Adaptive Server
uses to collate data. A sort order is tied to a particular language or set of
languages and to a specific character set. The same sort orders can be used for
English, French, and German because they sort the same characters identically,
for example, A, a, B, b, and so on. Or the characters are specific to one of the
languagesfor example, the accented characters, , , and , are used in
French but not in English or Germanand therefore, there is no conflict in
how those characters are sorted. The same is not true for Spanish however,
where the double letters ch and ll are sorted differently. Therefore, although the
same character sets support all four languages, there is one set of sort orders for
English, French and German, and a different set of sort orders for Spanish.
In addition, a sort order is tied to a particular character set. Therefore, there is
one set of sort orders for English, French, and German in the ISO 8859-1
character set, another set in the CP 850 character set, and so on. The sort orders
available for a particular character set are located in sort order definition files
(*.srt files) in the character set directory. For a list of character sets and their
available sort orders, see Table 10-5 on page 344.

Using sort orders

Sort orders are used to:

Create indexes

Store data into indexed tables

Specify an order by clause

Different types of sort orders

All character sets are offered with a binary sort order at a minimum, which
blindly sorts all data based only on the arithmetic value of the code assigned to
represent each letter (the binary code) in the character set. Binary sort order
works well for the first 128 characters of each character set (ASCII English)
and for Asian languages.When a character set supports more than one language
(for example, Group 1 or Unicode) the binary sort order most likely give
incorrect results, and you should select another sort order.
Character sets may also have one or more of the following dictionary sort
orders:

342

Adaptive Server Enterprise

CHAPTER 10

Configuring Character Sets, Sort Orders, and Languages

Dictionary order, case-sensitive, accent-sensitive sorts uppercase and

lowercase letters separately. Dictionary order recognizes the various
accented forms of a letter and sorts them after the associated unaccented
letter.

Dictionary order, case-insensitive, accent-sensitive sorts data in

dictionary order but does not recognize case differences. Uppercase letters
are equivalent to their lowercase counterparts and are intermingled in
sorting results. Useful for avoiding duplicate entries in tables of names.

Dictionary order, case-insensitive, accent-sensitive, order with preference

does not recognize case difference in determining equivalency of items.
A word in uppercase is equivalent to the same word in lowercase.
Preference is given to uppercase letters (they appear first) if all other
conditions are equal.
Using case-insensitive with preference may cause poor performance in
large tables when the columns specified in an order by clause match the
key of the tables clustered index. Do not select case-insensitive order with
preference unless your installation requires that uppercase letters be sorted
before lowercase letters in otherwise equivalent strings for order by
clauses.

Dictionary order, case-insensitive, accent-insensitive treats accented

forms of a letter as equivalent to the associated unaccented letter. It
intermingles accented letters in sorting results.

Selecting the default sort order

Sybase servers can support only one default sort order at a time. If your users
are using the same language or their languages use the same sort order, then
select the desired sort order. For example, if your users are using French data
and expect French sorting, then you can pick one of the French dictionary sort
orders. Or if your users are using data in multiple languages and the languages
use the same sort order, for example English, French, and German, you can
pick one sort order and it works for all your users in all languages.
However, if you have users using different languages that require different sort
orders, for example French and Spanish, then you must select one of the sort
orders as the default. If you pick, for example, a French sort order, your
Spanish users will not see the ch and ll double characters sorted as they would
expect. The installation procedure, by default, configures the server with the
binary sort order.

System Administration Guide: Volume 1

343

Selecting the sort order

You can use the sortkey function to setup customized alternative sort orders for
your dataone for each language.These sort orders can be selected
dynamically to meet the needs of different users. The sortkey function is
separate from the default sort order, but can coexist in the same server. The
range and depth of sort orders provided by the sortkey function is better than
those provided by the default sort order mechanism. For more information, see
sortkey and compare in the Reference Manual: Building Blocks.
Table 10-5: Available sort orders
Language or script
All languages
Cyrillic: Bulgarian,
Byelorussian,
Macedonian, Russian,
Serbian, Ukrainian

Character sets
UTF-8

Sort orders
Multiple sort orders, see Table 10-7 for list

CP 855, CP 866, CP 1251, ISO

8859-5, Koi8, Macintosh Cyrillic

Dictionary order, case sensitive, accent sensitive

Eastern European:

CP 852, ISO 8859-2, CP 1250

Czech, Slovak

Dictionary order, case sensitive, accent sensitive

Dictionary order, case insensitive, accent sensitive
Dictionary order, case sensitive, accent sensitive,
with preference
Dictionary order, case insensitive, accent insensitive

English, French,
German

ASCII 8, CP 437, CP850, CP 860,

CP 863, CP 1252a, ISO 8859-1,
ISO 8859-15, Macintosh Roman,
ROMAN8, ROMAN9, ISO 15

Dictionary order, case sensitive, accent sensitive

Dictionary order, case insensitive, accent sensitive
Dictionary order, case sensitive, accent sensitive,
with preference
Dictionary order, case insensitive, accent insensitive

English, French,
German

CP 850, CP 858

Alternate dictionary order, case sensitive

Alternate dictionary order, case sensitive, accent
insensitive

Greek

ISO 8859-7

Alternate dictionary order, case sensitive, with

preference
Dictionary order, case sensitive, accent sensitive

Hungarian

ISO 8859-2

Dictionary order, case sensitive, accent sensitive

Dictionary order, case insensitive, accent sensitive
Dictionary order, case insensitive, accent insensitive

Japanese

EUCJIS, SJIS, DECKANJI

General purpose case-insensitive dictionary

ordering

Kazakh
Russian

87
CP 866, CP 1251, ISO 8859-5,
Koi8, Macintosh Cyrillic

50
Dictionary order, case sensitive, accent sensitive

344

Dictionary order, case insensitive, accent sensitive

Adaptive Server Enterprise

CHAPTER 10

Configuring Character Sets, Sort Orders, and Languages

Language or script

Character sets

Sort orders

Scandinavian

CP 850

Dictionary order, case sensitive, accent sensitive

Dictionary order, case insensitive, with preference

Simplified Chinese

EUC-GB, GB-18030, CP936

General purpose case-insensitive dictionary

ordering

Spanish

Dictionary order, case sensitive, accent sensitive

Thai

ASCII 8, CP 437, CP850, CP 860,

CP 863, CP 1252, ISO 8859-1, ISO
8859-15, Macintosh Roman,
ROMAN8
CP 874, TIS 620

Turkish

ISO 8859-9

Dictionary order, case sensitive, accent sensitive

Dictionary order, case insensitive, accent sensitive

Dictionary order, case insensitive, accent insensitive
Dictionary order
Dictionary order, case insensitive, accent insensitive
Dictionary order, case insensitive, accent sensitive

Western European

CP 1252

Dictionary order, case insensitive, case sensitive,

with preference, accent insensitive, Spanish
dictionary, Spanish case insensitive, Spanish accent
insensitive

If your language does not appear here, there is no language-specific sort order
for your language. Select a binary sort order and then investigate whether the
sortkey function meets your needs. As this table illustrates, many languages
have more than one sort order.

Chinese Pinyin sort order

Pinyin, more formally known as Hanyu Pinyin, uses the Roman alphabet to
represent the standard Chinese pronunciation system. Pinyin consists of a
system of transliteration to Roman alphabets for reading and writing Mandarin
without Chinese characters. Pinyin uses accents to represent the four tones of
Mandarin.
Earlier versions of Adaptive Server used the Simplified Chinese (GB) sort
orders, gbpinyin and gbpinyinocs, using the Unilib character set, significantly
impacting the performance of databases using the GB character sets.
Adaptive Server version 15.0.3 automatically uses the gbpinyin and gbpinyinocs
sort orders, eliminating a processing step and significantly improving
performance.
In earlier versions, the default size of unilib cache configuration parameter was
268 KB. In version 15.0.3, the default has been increased to 302 KB.

System Administration Guide: Volume 1

345

Selecting the sort order

Improved performance occurs in queries that access ASCII and gbpinyin data.
However, if the data set has a mixture of other characters, you may not see any
performance improvement.
See Chapter 9, Configuring Character Sets, Sort Orders, and Languages in
the System Administration Guide for information about configuring Adaptive
Server to use the gbpinyin and gbpinyinocs sort orders.

Selecting case-insensitive sort orders for Chinese and Japanese character sets
Use two stored procedures to select case-insensitive sort orders:

sp_helpsort

sp_helpsort

sp_configure

sp_helpsort lists the available case-insensitive sort orders.

sp_helpsort
----------Name
ID
------------------------nocase_eucgb
52
nocase_cp936
52
nocase_gb18030
52
nocase_eucjis
52
nocase_sjis
52
nocase_deckanji
52
sp_configure

To switch to a case-insensitive sort order, enter:

sp_configure 'default sortorder id', 52

Selecting the default Unicode sort order

The default Unicode sort order is distinctly different from the sort order for the
servers default character set. This separate configuration parameter is a static
parameter that requires that you restart your server and reindex the unichar data
if it is changed. This sort order is identified using a string parameter, rather than
a numeric parameter, to guarantee that the sort order is unique.
Table 10-6 lists the available default Unicode sort orders.

346

Adaptive Server Enterprise

CHAPTER 10

Configuring Character Sets, Sort Orders, and Languages

Table 10-6: Default Unicode sort orders

Name
defaultml

ID
20

Description
Default Unicode multi-lingual ordering

thaidict
iso14651

21
22

Thai dictionary ordering

Ordering as per ISO14651 standard

utf8bin
binary

24
25

Ordering for UTF-16 that matches the UTF-8 binary

Binary sort

altnoacc
altdict

39
45

Alternate accent-insensitive
Alternate dictionary ordering

altnocsp
scandict

46
47

Alternate case-insensitive with preference

Scandinavian dictionary ordering

scannocp
bin_utf8

48
50

Scandinavian case-insensitive with preference

UTF-8 binary sort order

dict
nocase

51
52

General-purpose dictionary ordering

General-purpose case-insensitive dictionary ordering

nocasep
noaccent

53
54

General-purpose case-insensitive with preference

General-purpose accent-insensitive dictionary ordering

espdict
espnocs

55
56

Spanish dictionary ordering

Spanish case-insensitive dictionary ordering

espnoac
rusnocs

57
59

Spanish accent-insensitive dictionary ordering

Russian case-insensitive dictionary ordering

cyrnocs
elldict

64
65

Cyrillic case-insensitive dictionary ordering

Greek dictionary ordering

hundict
hunnoac

69
70

Hungarian dictionary ordering

Hungarian accent-insensitive dictionary ordering

hunnocs
turknoac

71
73

Hungarian case-insensitive dictionary ordering

Turkish accent-insensitive dictionary ordering

Table 10-7 lists the loadable sort orders.

Table 10-7: Loadable sort orders
Name
cp932bin

ID
129

Description
Ordering that matches the binary ordering of CP932

dynix
gb3213bn

130
137

Chinese phonetic ordering

Ordering that matches the binary ordering of GB2312

cyrdict
turdict

140
155

Common cyrillic dictionary ordering

Turkish Dictionary ordering

euckscbn
gbpinyin

161
163

Ordering that matches the binary ordering of EUCKSC

Chinese phonetic ordering

System Administration Guide: Volume 1

347

Selecting the sort order

Name

ID

Description

rusdict
sjisbin

165
179

Russian dictionary ordering

Ordering that matches the binary ordering of SJIS

eucjisbn
big5bin

192
194

Ordering that matches the binary ordering of EUCJIS

Ordering that matches the binary ordering of BIG5

To view this sort order list in Adaptive Server, use sp_helpsort. See the
Reference Manual: Procedures.
You can add sort orders using external files in the $SYBASE/collate/Unicode
directory. The names and collation IDs are stored in syscharsets. The names of
external Unicode sort orders do not have to be in syscharsets before you can
set the default Unicode sort order.
Note External Unicode sort orders are provided by Sybase. Do not attempt to

create external Unicode sort orders.

Sort order associated with Unicode data is completely independent of the sort
order associated with traditional character data. All relational expressions
involving the Unicode datatypes are performed using the Unicode sort order.
This includes mixed-mode expressions involving Unicode and non-Unicode
data. For example, in the following query the varchar character constant M
is implicitly cast to unichar and the comparison is performed according to the
Unicode sort order:
select * from authors where unicode_name > 'M'

The same holds true for all other comparison operators, as well as the
concatenation operator +, the operator in, and the operator between.
Once again, the goal is to retain compatibility with existing database
applications.
Tables joins based on equality (equijoins) deserve special mention. These are
generally optimized by the server to take advantage of indexes that defined on
the participating columns. When a unichar column is joined with a char
column, the latter requires a conversion, and since the character sort order and
the Unicode sort order are distinct, the optimizer will ignore the index on the
char column.

348

Adaptive Server Enterprise

CHAPTER 10

Configuring Character Sets, Sort Orders, and Languages

In Adaptive Server version 12.5.1 and later, when the servers default character
set is configured to UTF-8, you can configure the server's default sort order (for
char data) to be any of the above sort orders. Prior to this version, the binary
sort order bin_utf8 (ID=50) was the only well-behaved sort order for UTF-8.
Although not required, the sort order for char data in UTF-8 can be selected so
that it corresponds with the sort order for unichar.
There is a potential confusion regarding choice of binary sort orders for
Unicode. The sort order named binary is the most efficient one for unichar
data (UTF-16), and is thus the default. This order is based on the Unicode
scalar value, meaning that all 32-bit surrogate pairs are placed after all 16-bit
Unicode values. The sort order named utf8bin is designed to match the order
of the default (most efficient) binary order for UTF-8 char data, namely
bin_utf8. The recommended matching combinations are thus binary for
unichar and binary for UTF-8 char, or utf8bin for unichar and bin_utf8
for UTF-8 char. The former favors unichar efficiency, while the latter favors
char efficiency. Avoid using utf8bin for UTF-8 char, since it is equivalent to
bin_utf8 but less efficient.

Selecting a language for system messages

Any installation of Adaptive Server can use Language Modules containing
files of messages in different languages. Adaptive Server provides Language
Modules for messages in the following languages: English, Chinese
(Simplified), French, German, Japanese, Korean, Brazilian Portuguese, and
Spanish. If your client language is not one of these languages, you see system
messages in English, the default language.
Each client can choose to view messages in their own language at the same
time, from the same server; for example, one client views system messages in
French, another in Spanish, and another in German. To do this, however, all
selected languages must be part of the same language group. For example,
French, Spanish and German are all part of language group 1. Japanese, on the
other hand, is part of language group 101, which contains no other languages.
Therefore, if Japanese is your server language, you can display system
messages only in Japanese or English. Remember that all language groups can
display messages in English. There is also a server-wide default language, used
if the user has not selected a specific language. If you use Unicode, you can
view system messages in any of the supported languages.
You can select the language for your system messages in one of two ways:

System Administration Guide: Volume 1

349

Setting up your server: examples

Select a language as part of your user profile

Enter a language in the locales.dat file

Table 10-8 displays the supported system message languages and their
language groups. Each user can select only one language per session for system
messages.
Table 10-8: Supported system messages
Language group
Group 1

System message languages

French, German, Spanish, Brazilian Portuguese

Group 2

Polish

Character sets
ASCII 8, CP 437, CP 850, CP 860,
CP 863, CP 1252, ISO 8859-1, ISO
8859-15, Macintosh Roman, ROMAN8
Cp 1250, CP 852, ISO 8859-2

Group 101
Group 102

Japanese
Simplified Chinese (PRC)

CP 932, DEC Kanji, EUC-JIS, Shift-JIS

CP 936, EUC-GB, GB18030

Group 104
Group 105

Korean
Thai

EUC-KSC, CP 949
CP 874, TIS 620

Unicode

French, German, Spanish, Brazilian Portuguese,

Japanese, Simplified Chinese, Korean

UTF-8

All Other Language

Groups

English

Install Language Modules for all languages in which clients will receive
messages. These Language Modules, located in the locales subdirectory of the
Adaptive Server installation directory, are part of a group of files called
localization files. For information about localization files and the software
message directory structure, see Types of localization files on page 366.

Setting up your server: examples

This section discusses setup options and the steps necessary to implement
them. This is only a sample, and is meant to suggest ideas and methods for your
own setup process.

A Spanish-version server
This examples shows how to set up a new server with all clients using the same
language. To do this:

350

Adaptive Server Enterprise

CHAPTER 10

Configuring Character Sets, Sort Orders, and Languages

Select the server language, in this case, Spanish. By reviewing Table 10-1
on page 333, you see that Spanish is part of language group 1. Based on
your platform, select a character set from language group 1. Sybase
recommends that you select the character set used by the greatest number
of clients. Or, if you think your company might someday expand into other
countries and languages, you might consider installing Unicode (see
Selecting the character set for your server on page 331).

Install the Spanish Language Module in the server. This allows clients to
view system messages in Spanish.

Select the default sort order. By referring to Table 10-5 on page 344, you
see that Spanish has three possible sort orders, in addition to binary sort
order. Select a sort order.

Restart the server.

A U.S.-based company in Japan

This example involves clients in Japan, who want to enter data, sort data, and
receive system messages in Japanese, while submitting data to a server that is
accessed by English-only users:
1

Select the default character set for your server. If you install a character set
from language group 101 (Japanese), you can support both Japanese and
English data in the same server.

Install the Japanese Language Module so that system messages are

available in Japanese.

Select the sort order. By referring to Table 10-5 on page 344, you can see
that a binary sort order is the only sort order available for Japanese.
Therefore, both the English and Japanese clients have a default binary sort
order. Consider using the sortkey function to provide solutions for both
audiences.

Make sure that each Japanese user requests Japanese messages by default.
Since you are using a character set from language group 101, and you have
already installed the Japanese Language Module, your client in Japan sees
messages in Japanese, while clients in the U.S. can choose to see messages
in either English or Japanese.

System Administration Guide: Volume 1

351

Changing the character set, sort order, or message language

A Japan-based company with multinational clients

This company is located in Japan, and has clients in France, Germany, and
Spain. You need to mix European and Asian languages in the same server.
1

Select the default server language and character set. Since your company
is based in Japan and most of your clients are located in Japan, the default
server language should be Japanese. But you also want your clients in
France, Germany, and Spain to be able to send and receive data in their
native languages. By reviewing Table 10-1 on page 333, you can see that
Japanese is part of language group 101, while French, German, and
Spanish are part of language group 1. Since the languages you need are not
part of the same language group, the only way you can have all of these
languages on the same server is to select Unicode as your default character
set.

Install the Language Modules for Japanese, French, German, and Spanish.

Select the binary sort order, since this is the only sort order available for
the Unicode character set. (You can, however, consider using the sortkey
function inside your application code to supply data sorted according to
each users preference.)

Select Japanese as the default language for system messages. Clients in

other countries can select their own native language for messages.

Changing the character set, sort order, or message

language
Even after you have configured your server, a system administrator can change
the default character set, sort order, or message language used by Adaptive
Server. Because a sort order is built on a specific character set, changing
character sets always involves a change in sort order. However, you can change
the sort order without changing character sets, because more than one sort
order may be available for a character set.
To display Adaptive Servers default sort order, character set, and a table of its
primary sort orders, enter:
sp_helpsort

352

Adaptive Server Enterprise

CHAPTER 10

Configuring Character Sets, Sort Orders, and Languages

Changing the default character set

Adaptive Server can have only one default character set, the character set in
which data is stored in its databases. When you install Adaptive Server, you
specify a default character set.
Warning! Read the following carefully, and exercise caution when changing
the default character set in Adaptive Server. Sybase strongly recommends that
you perform backups before you change a default character set.

When you change the default character set in Adaptive Server, you must
convert any existing data to the new default character set. Conversion is
unnecessary only if:

There is no user data in the server.

It is acceptable to destroy user data in the server.

You are absolutely certain that data in the server uses only ASCll-7. In this
case, you can change the default without first copying your data out of the
server.

In all other cases, you must convert the existing data as follows:
1

Copy the data out using bcp.

Change the default character set.

Use bcp with the appropriate flags for data conversion to copy the data
back into the server.

See the Utility Guide for more information about using bcp to copy data.
Warning! After converting data to a different character set (particularly to
UTF-8), the data may be too large for the allocated column size. Re-create the
columns affected with a larger size.

Code conversion between the character set of the existing data and the new
default character set must be supported. If it is not, conversion errors will occur
and the data is not converted correctly. See Chapter 11, Configuring
Client/Server Character Set Conversions, for more information about
supported character set conversions.

System Administration Guide: Volume 1

353

Changing the character set, sort order, or message language

Even if conversions are supported between the character sets, some errors may
occur due to minor differences between the character sets, or because some
characters do not have equivalents in other character sets. Rows containing
problematic data may not get copied back into the database, or data may
contain partial or invalid characters.

Changing the sort order with a resources file

Adaptive Server character sets can be changed using the resource file. The
sample resource file sqlloc.rs is located in $SYBASE/ASE12_5/init/sample_resource_files/.
The resource file from the Adaptive Server 12.5.1 installation looks similar to
the following:
sybinit.release_directory: USE_DEFAULT
sqlsrv.server_name: PUT_YOUR_SERVER_NAME_HERE
sqlsrv.sa_login: sa
sqlsrv.sa_password:
sqlsrv.default_language: USE_DEFAULT
sqlsrv.language_install_list: USE_DEFAULT
sqlsrv.language_remove_list: USE_DEFAULT
sqlsrv.default_characterset: USE_DEFAULT
sqlsrv.characterset_install_list: USE_DEFAULT
sqlsrv.characterset_remove_list: USE_DEFAULT
sqlsrv.sort_order: USE_DEFAULT
# An example sqlloc resource file...
# sybinit.release_directory: USE_DEFAULT
# sqlsrv.server_name: PUT_YOUR_SERVER_NAME_HERE
# sqlsrv.sa_login: sa
# sqlsrv.sa_password:
# sqlsrv.default_language: french
# sqlsrv.language_install_list: spanish,german
# sqlsrv.language_remove_list: USE_DEFAULT
# sqlsrv.default_characterset: cp437
# sqlsrv.characterset_install_list: mac,cp850
# sqlsrv.characterset_remove_list: USE_DEFAULT
# sqlsrv.sort_order: dictionary

354

Adaptive Server Enterprise

CHAPTER 10

Configuring Character Sets, Sort Orders, and Languages

Changing the default sort order

Adaptive Server can have only one default sort order, the collating sequence it
uses to order data. When you consider changing the sort order for character
data on a particular Adaptive Server, keep this in mind: all of your
organizations Adaptive Servers should have the same sort order. A single sort
order enforces consistency and makes distributed processing easier to
administer.
You may have to rebuild your indexes after changing the default sort order. For
more information, see Reconfiguring the character set, sort order, or message
language on page 355.

Reconfiguring the character set, sort order, or message language

This section summarizes the steps to take before and after changing Adaptive
Servers default character set, sort order, or message language. For procedures
on how to configure the character set, sort order, or message language for a new
server, see the configuration documentation for your platform.
Back up all databases in Adaptive Server before and after you change character
sets or sort orders. After you back up your databases, use bcp to copy the data
in and out of your databases if:

A database contains character data and you want to convert the data to a
new character set. Do not load a database dump of the data into a server
that uses the new default character set. Adaptive Server assumes the
loaded data is in the new character set, and corrupts the data.

You are changing the default sort order only and not the default character
set. You cannot load a database from a dump performed prior to changing
the sort orderif you attempt to, an error message appears, and Adaptive
Server aborts the load.

You change the default character set, and either the old or the new sort
order is not binary. You cannot load a database dump that was made before
you changed the character set.

You cannot reload your data from a database dump once you have reconfigured
the default character set and sort order (unless both old and new character sets
use a binary sort order and no conversion is required between the old and new
character sets). See Changing the default character set on page 353 for more
information,

System Administration Guide: Volume 1

355

Changing the character set, sort order, or message language

Unicode examples
In the following example, a fictitious database named xpubs is modified to use
univarchar columns.

Schema
Assume a database was created using the following script on a server that has
all the installation defaults, namely character set iso_1 and default sort order
ID 50, binary_iso_1.
> create database xpubs
> go
> use xpubs
> go
> create table authors (au_id int, au_lname
varchar(255), au_fname varchar(255))
> go
> create index au_idx on authors(au_lname, au_fname)
> go

Then the data was loaded into the server using a series of inserts and updates.

Converting to UTF-8
The first step towards using Unicode is to extract the data and convert it to
UTF-8 form.
% bcp xpubs..authors out authors.utf8.bcp -c -Jutf8 -Usa -P

The next step to install UTF-8 as the default character set in the server:
%
%
>
>
>
>

charset -Usa -P binary.srt utf8

isql -Usa -P
sp_configure 'default sortorder id', 50, 'utf8'
go
shutdown
go

Restart the server to modify the default character set and re-create indexes on
the system tables. Restart the server a second time, then reload the data:
%
>
>
>
>
>

isql -Usa -P
sp_dboption xpubs, 'select into', true
go
use xpubs
go
checkpoint

356

Adaptive Server Enterprise

CHAPTER 10

>
>
>
>

Configuring Character Sets, Sort Orders, and Languages

go
delete from authors
go
quit

% bcp xpubs..authors in authors.utf8.bcp -c -Jutf8 -Usa -P

Migrating selected columns to unichar

With a working database running with UTF-8 as the default character set, it
becomes a simple matter to convert select columns to univarchar:
% isql -Usa -P
> use xpubs
> go
> alter table authors modify au_lname univarchar(255),
au_fname univarchar(255)
> go

The columns are modified to the new datatypes, the data is converted in place,
and the index is re-created.

Migrating to or from unitext

Currently, the alter table modify command does not support text, image, or
unitext columns. To migrate from a text to a unitext column, you must first use
bcp, create a table with unitext columns, and then use bcp again to place data
into the new table. This migration path only works when you invoke bcp with
-Jutf8 option.

Preliminary steps
Before you run the installation program to reconfigure Adaptive Server:
1

Dump all user databases and the master database. If you have made
changes to model or sybsystemprocs, dump them also.

Load the Language Module if it is not already loaded (see the

configuration documentation for your platform for complete instructions).

If you are changing the Adaptive Server default character set, and your
current databases contain non ASCII-7 data, use bcp to copy the existing
data out of your databases.

System Administration Guide: Volume 1

357

Changing the character set, sort order, or message language

Once you have loaded the Language Module, you can run the Adaptive Server
installation program, which allows you to:

Install or remove message languages and character sets included with

Adaptive Server

Change the default message language or character set

Select a different sort order

See the configuration documentation for your platform for instructions on

using the installation program
Note Before you change the character set or sort order, Adaptive Server must
have as many open databases as there are databases managed by the server. If
Adaptive Server does not have a sufficient number of open databases when it
is re-started after a change in sort order, Adaptive Server prints this message to
the error log and the server will revert to the former sort order:
The configuration parameter 'number of open databases'
must be at least as large as the number of databases,
in order to change the character set or sort order." Restart Adaptive Server, use sp_configure to increase
'number of open databases' to at least %d, then reconfigure the character set or sort order

To reconfigure the language, character set, or sort order, use the sqlloc utility,
described in Utility Guide. If you are using Windows, use the Server Config
utility, described in Chapter 3, Default Adaptive Server Configuration, in the
Configuration Guide.
If you installed additional languages but did not change the Adaptive Server
character set or sort order, you have completed the reconfiguration process.
If you changed the Adaptive Server default character set, and your current
databases contain non ASCII-7 data, copy your data back into your databases,
using bcp with the necessary flags to enable conversion.
If you changed the Adaptive Server default sort order or character set, see
Reconfiguring the character set, sort order, or message language on page
355.

358

Adaptive Server Enterprise

CHAPTER 10

Configuring Character Sets, Sort Orders, and Languages

Setting the users default language

If you install an additional language, users running client programs can run
create login to set that language as their default language, or set the LANG
variable on the client machine, with the appropriate entries in locales.dat.

Recovery after reconfiguration

Every time Adaptive Server is stopped and restarted, recovery is performed
automatically on each database. Automatic recovery is discussed in detail in
Chapter 11, Developing a Backup and Recovery Plan, in the System
Administration Guide: Volume 2.
After recovery is complete, the new sort order and character set definitions are
loaded.
If you have changed the sort order, Adaptive Server switches to single-user
mode to allow the necessary updates to system tables and to prevent other users
from using the server. Each table with a character-based index is automatically
checked to see if any indexes have been corrupted by the sort order change.
Character-based indexes in tables are automatically rebuilt, if necessary, using
the new sort order definition.
After the system indexes are rebuilt, character-based user indexes are marked
suspect in the sysindexes system table. User tables with suspect indexes are
marked read-only in sysobjects to prevent updates to these tables and use of
the suspect indexes until they have been checked and, if necessary, rebuilt.
Range-partitioned user tables are checked for character-based partition keys,
and are marked suspect if the sort order change or character set change might
cause partition corruption.
Next, the new sort order information replaces the old information in the area of
the disk that holds configuration information. Adaptive Server then shuts down
so that it starts for the next session with a complete and accurate set of system
information.

Using sp_indsuspect to find corrupt indexes

After Adaptive Server shuts down, restart it, and use sp_indsuspect to find the
user tables that need to be reindexed.
sp_indsuspect [tab_name]

System Administration Guide: Volume 1

359

Changing the character set, sort order, or message language

where tab_name is the name of the table you are investigating. If tab_name is
missing, sp_indsuspect creates a list of all tables in the current database that has
indexes marked suspect when the sort order changes.
This example shows that running sp_indsuspect in mydb database yields one
suspect index:
sp_indsuspect
Suspect indexes in database mydb
Own.Tab.Ind (Obj_ID, Ind_ID) =
dbo.holdings.h_name_ix(160048003, 2)

Rebuilding indexes after changing the sort order

dbcc reindex checks the integrity of indexes on user tables by running a fast
version of dbcc checktable. For details, see Chapter 10, Checking Database
Consistency, in the System Administration Guide: Volume 2. dbcc reindex
drops and rebuilds the indexes where the sort order used is not consistent with
the new sort order. When dbcc reindex discovers the first index-related error, it
displays a message, then rebuilds the inconsistent indexes. The system
administrator or table owner should run dbcc reindex after changing the sort
order in Adaptive Server.

dbcc reindex ({table_name | table_id})

Run this command on all tables listed by sp_indsuspect as containing suspect

indexes. For example:
dbcc reindex(titles)
One or more indexes are corrupt. They will be rebuilt.

In the preceding example, dbcc reindex discovers one or more suspect indexes
in the table titles; it drops and re-creates the appropriate indexes.
If the indexes for a table are already correct, or if there are no indexes for the
table, dbcc reindex does not rebuild any indexes. It displays a message instead.
If a table is suspected of containing corrupt data, the command is aborted. If
that happens, an error message instructs the user to run dbcc checktable.
When dbcc reindex finishes successfully, all suspect marks on the tables
indexes are removed. The read-only mark on the table is also removed, and
the table can be updated. These marks are removed whether or not any indexes
have to be rebuilt.

360

Adaptive Server Enterprise

CHAPTER 10

Configuring Character Sets, Sort Orders, and Languages

dbcc reindex does not reindex system tables. System indexes are checked and
rebuilt, if necessary, as an automatic part of recovery after Adaptive Server is
restarted following a sort order change.

Upgrading text data after changing character sets

If you have changed an Adaptive Server character set to a multibyte character
set, use dbcc fix_text to upgrade text values.
A text value can be large enough to cover several pages; therefore, Adaptive
Server must be able to handle characters that span page boundaries. To do so,
Adaptive Server requires additional information on each of the text pages. The
system administrator or table owner must run dbcc fix_text on each table that
has text data to calculate the new values needed.
To see the names of all tables that contain text data, use:
select sysobjects.name
from sysobjects, syscolumns
where syscolumns.type = 35
and sysobjects.id = syscolumns.id

The system administrator or table owner must run dbcc fix_text to calculate the
new values needed.
The syntax of dbcc fix_text is:
dbcc fix_text (table_name | table_id)

The table named must be in the current database.

dbcc fix_text opens the specified table, calculates the character statistics
required for each text value, and adds the statistics to the appropriate page
header fields. This process can take a long time, depending on the number and
size of the text values in a table. dbcc fix_text can generate a large number of
log records, which may fill up the transaction log. dbcc fix_text performs
updates in a series of small transactions so that if a log becomes full, only a
small amount of work is lost.

If you run out of log space, clear out your log (see Chapter 12, Backing Upa
and Restoring User Databases, in System Administration Guide: Volume 2).
Then restart dbcc fix_text, using the same table that was being upgraded when
the original dbcc fix_text halted. Each multibyte text value contains information
that indicates whether it has been upgraded, so dbcc fix_text upgrades only the
text values that were not processed in earlier passes.

System Administration Guide: Volume 1

361

Changing the character set, sort order, or message language

If your database stores its log on a separate segment, you can use thresholds to
manage clearing the log. See Chapter 16, Managing Free Space with
Thresholds, in System Administration Guide: Volume 2.
If dbcc fix_text cannot acquire a needed lock on a text page, it reports the
problem and continues with the work, like this:
Unable to acquire an exclusive lock on text page 408.
This text value has not been recalculated. In order to
recalculate those TEXT pages you must release the lock
and reissue the dbcc fix_text command.

Retrieving text values after changing character sets

If you attempt to retrieve text values after changing to a multibyte character set,
and you have not run dbcc fix_text, the command fails with this error message:
Adaptive Server is now running a multi-byte character
set, and this TEXT columns character counts have not
been recalculated using this character set. Use dbcc
fix_text before running this query again.

If you have changed the sort order or character set and errors occurred, see
How to Manually Change Sort Order or Default Character Set in the
Adaptive Server Enterprise Troubleshooting and Error Messages Guide.

Handling suspect partitions

Partitions are marked suspect for two reasons:

A sort order or character set change on a range-partitioned table

A cross-platform dump and load with a hash-partitioned table

If the table is marked with suspect partitions:

362

All updates and cursor activities are suspended on this table.

No alter table commands, except partition by, are allowed. create index and
drop index are not allowed on a table with suspect partitions.

The select command is allowed on tables containing suspect partitions.

However, the optimizer treats such tables as round-robin partitioned
tables, to avoid using the possibly corrupt partition condition.

Adaptive Server Enterprise

CHAPTER 10

Configuring Character Sets, Sort Orders, and Languages

Fixing tables with suspect partitions

If the partition condition needs fixing after a sort-order change, you can use
alter table with the partition by option to repartition a table that has suspect
partitions.

If the partition condition does not need fixing, you can use the reorg rebuild
table command to rebuild the table, redistributing only the data rows
among the partitions.
If the indexes as well as the partitions on a table are marked suspect, use
partition by or reorg rebuild to fix both the suspect index and suspect

partitions.

Handling suspect partitions in cross-platform dump and load operations

During the first online database command, after you execute load database
across two platforms with different endian types, the hash partition is
marked suspect.

Any global clustered index on a round-robin partition, which has an

internally generated partition condition with a unichar or varchar partition
key, is marked suspect.

After the database is online, use sp_post_xpload to fix the suspect

partitions and indexes.

Installing date strings for unsupported languages

Use sp_addlanguage to install names for the days of the week and months of
the year for languages that do not have language modules. With
sp_addlanguage, lets you define:

A language name and (optionally) an alias for the name

A list of the full names of months and a list of abbreviations for the month
names

A list of the full names of the days of the week

The date format for entering dates (such as month/day/year)

The number of the first day of the week

For example to add the information for Italian:

System Administration Guide: Volume 1

363

Installing date strings for unsupported languages

sp_addlanguage italian, italiano,

"gennaio,febbraio,marzo,aprile,maggio,giugno,luglio,agosto,settembre,ottobre,
novembre,dicembre",
"genn,feb,mar,apr,mag,giu,lug,ago,sett,ott,nov,dic",
"lunedi,martedi,mercoledi,giovedi,venerdi,sabato,domenica",
dmy, 1
sp_addlanguage enforces strict data entry rules. The lists of month names,

month abbreviations, and days of the week must be comma-separated lists with
no spaces or line feeds (returns). Also, they must contain the correct number of
elements (12 for month strings, 7 for day-of-the-week strings.)
Valid values for the date formats are: mdy, dmy, ymd, ydm, myd, and dym. The
dmy value indicates that the dates are in day/month/year order. This format
affects only data entry; to change output format, you must use the convert
function.

Server versus client date interpretation

Generally, date values are resolved on the client. When a user selects date
values, Adaptive Server sends them to the client in an internal format. The
client uses the common.loc file and other localization files in the default
language subdirectory of the locales directory on the client to convert the
internal format to character data. For example, if the users default language is
Spanish, Adaptive Server looks for the common.loc file in
/locales/spanish/char_set. It uses the information in the file to display, for
example, 12 febrero 2007.
Assume that the users default language is set to Italian, a language for which
Adaptive Server does not provide a language module, and that the date values
in Italian have been added. When the client connects to the server and looks for
the common.loc file for Italian, it does not find the file. The client prints an
error message and connects to the server. If the user then selects date values,
the dates are displayed in U.S. English format.To display the date values added
with sp_addlanguage, use the convert function to force the dates to be
converted to character data at the server.
The following query generates a result set with the dates in U.S. English
format:
select pubdate from titles

The query below, however, returns the date with the month names in Italian:
select convert(char(19),pubdate) from titles

364

Adaptive Server Enterprise

CHAPTER 10

Configuring Character Sets, Sort Orders, and Languages

Internationalization and localization files

Types of internationalization files
The files that support data processing in a particular language are called
internationalization files. Several types of internationalization files come with
Adaptive Server.
Table 10-9: Internationalization files
File
charset.loc

*.srt

*.xlt

Location
In each character set
subdirectory of the
charsets directory
In each character set
subdirectory of the
charsets directory
In each character set
subdirectory of the
charsets directory

Purpose and contents

Character set definition files that define the lexical properties of each
character, such as alphanumeric, punctuation, operand, and uppercase or
lowercase. Used by Adaptive Server to correctly process data.
Defines the sort order for alphanumeric and special characters, including
ligatures, diacritics, and other language-specific considerations.
Terminal-specific character translation files for use with utilities such as bcp
and isql. For more information about how the.xlt files are used, see Chapter
11, Configuring Client/Server Character Set Conversions, and the Utility
Guide.

Warning! Do not alter any of the internationalization files. If you need to

install a new terminal definition or sort order, contact your local Sybase office
or distributor.

Character sets directory structure

Figure 10-3 shows the directory structure for the Western European character
sets that come with Adaptive Server. There is a separate subdirectory for each
character set in the charsets directory. Within the subdirectory for each
character set (for example, cp850) are the character set and sort order definition
files and terminal-specific files.
If you load additional character sets, they also appear in the charsets directory:

System Administration Guide: Volume 1

365

Internationalization and localization files

Figure 10-3: Structure of the charsets directory

charsets

cp850
charset.loc
*.srt

iso_1
charset.loc
*.srt
*.xlt

cp437
charset.loc
*.srt

mac
charset.loc
*.srt

roman8
charset.loc
*.srt

charset_name
charset.loc
*.srt

The following global variables contain information about character sets:

@@char_convert contains 0 if character set conversion is not in effect.

Contains 1 if character set conversion is in effect.

@@client_csname the clients character set name. Set to NULL if client

character set has never been initialized; otherwise, contains the name of
the character set for the connection.

@@client_csid the clients character set ID. Set to -1 if client character

set has never been initialized; otherwise, contains the client character set
ID from syscharsets for the connection.

@@client_csexpansion returns the expansion factor used when

converting from server's character set to client's character set.

@@maxcharlen the maximum length, in bytes, of a character in the

Adaptive Server default character set.

@@ncharsize the maximum length, in bytes, of a character set in the

current server default character set.

@@unicharsize equals 2.

Types of localization files

Adaptive Server includes several localization files for each Language Module:

366

locales.dat in the locales directory. Used by client applications to

identify the default message language and character set.

Adaptive Server Enterprise

CHAPTER 10

Configuring Character Sets, Sort Orders, and Languages

server.loc in the character set subdirectories under each language

subdirectory in the $SYBASE/$SYBASE_ASE/locales directory. Software
messages translated into the local language. Sybase products have
product-specific *.loc files. If an entry is not translated, the software
message or string appears in U.S. English instead of the local language.

common.loc in each language and character set directory of the locales

directory. Contains the local names of the months of the year and their
abbreviations, and information about the local date, time, and money
formats.

All Adaptive Server-related locales files (used by dataserver, sqlloc, syconfig,

and so on) are in $SYBASE/SYBASE_ASE/locales. All Open Client/Serverrelated locales files (ctlib, ctisql, ctbcp, optdiag, installjava, and so on) are located
in $SYBASE/locales.
Warning! Do not alter any of the localization files. If you need to alter any
information in those files, contact your local Sybase office or distributor.

Software messages directory structure

Figure 10-4 shows how localization files are arranged. Within the locales
directory is a subdirectory for each language installed. There is always a
us_english subdirectory. (On PC platforms, this directory is called english.)
During installation, when you are prompted to select the languages you want
installed on Adaptive Server, the installation program lists the supported
software message languages. If you install language modules for additional
languages, you see subdirectories for those languages. Within each language
subdirectory are subdirectories for the supported character sets; for example,
cp850 is a supported character set for us_english. Software message files for
each Sybase product reside in the character set subdirectories.

System Administration Guide: Volume 1

367

Internationalization and localization files

Figure 10-4: Messages directory structure

locales
locales.dat

us_english

cp850
server.loc
common.loc

language2

iso_1

cp437

server.loc
common.loc

server.loc
common.loc

mac
server.loc
common.loc

roman8

char_set1

char_set2

server.loc
server.loc
server.loc
common.loc common.loc common.loc

Message languages and global variables

The following global variables contain information about languages:

368

@@langid contains the local language ID of the language currently in

use (specified in syslanguages.langid)

@@language contains the name of the language currently in use

(specified in syslanguages.name)

Adaptive Server Enterprise

CH A PTE R

11

Configuring Client/Server
Character Set Conversions

Topic
Character set conversion
Supported character set conversions

Page
369
370

Choosing a conversion type

Enabling and disabling character set conversion

374
376

Error handling in character set conversion

Conversions and changes to data lengths

377
378

Specifying the character set for utility programs

Display and file character set command line options

379
379

Character set conversion

In a heterogeneous environment, Adaptive Server may need to
communicate with clients running on different platforms using different
character sets. Although different character sets may support the same
language group (for example, ISO 8858-1 and CP 850 support the group
1 languages), they may encode the same characters differently. For
example, in ISO 8859-1, the character is encoded as 0xE0 in
hexadecimal. However, in CP 850 the same character is encoded as 0x85
in hexadecimal.
To maintain data integrity between your clients and servers, data must be
converted between the character sets. The goal is to ensure that an a
remains an a even when crossing between machine and character set
boundaries. This process is known as character set conversion.

System Administration Guide: Volume 1

369

Supported character set conversions

Supported character set conversions

Character set conversion occurs between a pair of character sets. The supported
conversions in any particular client/server system depend on the character sets
used by the server and its clients. One type of character set conversion occurs
if the server uses a native character set as the default; a different type of
conversion is used if the server default is Unicode UTF-8.

Conversion for native character sets

Adaptive Server supports character set conversion between native character
sets belonging to the same language group. If the server has a native character
set as its default, the clients character sets must belong to the same language
group. Figure 11-1 is an example of a Western European client/server system.
In this example, the clients character sets and the Adaptive Server default
character set all belong to language group 1. Data is correctly converted
between the client character sets and the server default character set. Since they
all belong to the same language group, the clients can view all data on the
server, no matter which client submitted the data.
Figure 11-1: Character set conversion when server and client character
sets belong to the same language group

French client: Roman8

Adaptive
Server
CP 850

German client: ISO 8859-1

Spanish client: Macintosh Roman

For a list of the language groups and supported character sets, see Table 10-1
on page 333.

370

Adaptive Server Enterprise

CHAPTER 11

Configuring Client/Server Character Set Conversions

Conversion in a Unicode system

Adaptive Server also supports character set conversion between UTF-8 and
any native character set that Sybase supports. In a Unicode system, since the
server default character set is UTF-8, the client character set may be a native
character set from any language group. Therefore, a Japanese client (group
101), a French client (group 1), and an Arabic client (group 6) can all send and
receive data from the same server. Data from each client is correctly converted
as it passes between each client and the server.
Figure 11-2: Character set conversion in a Unicode system

French client: CP 1252 (Group 1)

Adaptive
Server
UTF-8

Japanese client: CP 932 (Group 101)

Arabic client: CP 1256 (Group 6)

Each client can view data only in the language supported by its character set.
Therefore, the Japanese client can view any Japanese data on the server, but it
cannot view Arabic or French data. Likewise, the French client can view
French or any other Western European language supported by its character set,
but not Japanese or Arabic.
Figure 11-3: Viewing Unicode data

French data
ASE
Japanese data
UTF-8
Arabic data

French: CP 1252
Japanese: CP 932
Arabic: CP 1256

An additional character set, ASCII-7, is a subset of every character set,

including Unicode, and is therefore compatible with all character sets in all
language groups. If either the Adaptive Server or the clients character set is
ASCII-7, any 7-bit ASCII character can pass between the client and server
unaltered and without conversion.

System Administration Guide: Volume 1

371

Supported character set conversions

Sybase recommends that you do not configure a server for ASCII-7. You can
achieve the same benefits of compatibility by restricting each client to use only
the first 128 characters of each native character set.

Adaptive Server direct conversions

Adaptive Server direct conversions are between two native character sets of the
same language group. For example, Adaptive Server supports conversion
between CP 437 and CP 850, because both belong to language group 1.
Adaptive Server direct conversions exist between many, but not all, native
character sets of a language group (see Table 11-1 on page 375).

Unicode conversions
Unicode conversions exists for all native character sets. When converting
between two native character sets, Unicode conversion uses Unicode as an
intermediate character set. For example, to convert between the server default
character set (CP 437), and the client character set (CP 860), CP 437 is first
converted to Unicode; Unicode is then converted to CP 860.
Unicode conversions may be used either when the default character set of the
server is UTF-8, or a native character set. You must specifically configure your
server to use Unicode conversions (unless the servers default character set is
UTF-8).
Earlier versions of Adaptive Server used direct conversions, and it is the
default method for character set conversions. However, Unicode conversions
allow easier and less complex character set conversion. While Adaptive Server
direct conversions are still supported, Sybase now also uses Unicode
conversions to provide complete conversion support for all character sets and
has no plans to add new direct conversions.

Allowing Unicode noncharacters

In versions of Adaptive Server earlier than 15.7, the unichar, univarchar, unitext,
char, varchar, and text datatypes under the utf-8 default character set did not
accept Unicode noncharacters (code points permanently reserved for internal
use).

372

Adaptive Server Enterprise

CHAPTER 11

Configuring Client/Server Character Set Conversions

Adaptive Server version 15.7 allows you to ignore Unicode noncharacters by

enabling the enable functionaliry group or the enable permissive unicode
configuration parameters.
If you do not enable this feature, Adaptive Server rejects these noncharacters
as in earlier versions.
For more information about the Unicode standard, see the Unicode Consortium
Web site at http://www.unicode.org/.
When you enable this feature, Unicode noncharacters are not detected in:

Parameters:

Presented as univarchar and unitext (UTF-16) datatypes

Presented as varchar and text (UTF-8) datatypes

As parameters to dynamic SQL statements

As input to parameterized language statements

String literals when the servers character set is UTF-8

Escaped string literals (those prefixed with U&), regardless of the servers
character set

Conversion processes between unichar (UTF-16) and varchar (UTF-8) in

either direction

In addition, Unicode noncharacters are acceptable in simple expressions such

as comparisons, where they sort higher than legal Unicode characters.
You can use Unicode noncharacters as parameters to these functions:
ascii()

lower()

sortkey()

char_length()

ltrim()

soundex()

charindex()

patindex()

str_replace()

compare()

replicate()

stuff()

datalength()

reverse()

substring()

difference()

right()

upper()

left()

rtrim()

uscalar()

len()

Note This feature does not affect UTF-16 surrogate handling enabled with the
enable surrogate handling configuration parameter. See Setting Configuration

Parameters in the System Administration Guide.

System Administration Guide: Volume 1

373

Choosing a conversion type

Choosing a conversion type

To determine the conversion options that are available for your client/server
system, see Table 11-1 on page 375.

Non-Unicode client/server systems

In a non-Unicode system, the character sets of the server and clients are native
character sets; therefore, you can use the Adaptive Server direct conversions.
However, there are some character sets for which there is no Adaptive Server
direct conversion; in this situation, you must use Unicode conversions.

If all character sets in your client/server system are column 1 of Table 111, use the Adaptive Server direct conversions. The character sets must all
belong to the same language group.

If the character sets in your client/server system are in column 2 of Table

11-1, or some combination of columns 1 and 2, configure your server to
use Unicode conversions. Again, the character sets must all belong to the
same language group.

For example, assume the server default character set is CP 850 and the clients
character sets are either ISO 8859-1 or ROMAN 8. Table 11-1 shows that direct
conversions exist between CP 850 and the client character sets. Now, suppose
you add a client using CP 1252. Since there is no direct conversion between CP
1252 and CP 850, (the default server character set), you must use Unicode
conversions to convert between CP 1252 and CP 850. When you have a
mixture of character setssome where you can use Adaptive Server direct
conversions and others where you must use Unicode conversionsyou can
specify that a combination of Adaptive Server direct conversion and Unicode
conversion be used.

Unicode client/server systems

If your server default is Unicode UTF-8, then all conversions are between
UTF-8 and the native character set being used on the client systems. In a
Unicode system, Unicode conversions are used exclusively.

374

Adaptive Server Enterprise

CHAPTER 11

Configuring Client/Server Character Set Conversions

Table 11-1: Conversion methods for character sets

Language
group

Column 1 Adaptive Server direct conversions and

Unicode conversions

Column 2 Unicode
conversions only

Group 1

CP 437, CP 850, ISO 8859-1, Macintosh Roman

Group 2

CP 852, CP 1250, CP 8859-1, Macintosh Central European

CP 860, CP 1252, ISO 8859-15,

CP 863
ISO 8859-2

Group 4
Group 5

No conversions needed (only one character set supported)

CP 855, CP 866, CP 1251, ISO 8859-5, Koi8, Macintosh Cyrillic

Group 6
Group 7

CP 869, CP 1253, GREEK8, ISO 8859-7, Macintosh Greek

Group 8
Group 9
Group 101
Group 102

CP 864, CP 1256, ISO 8859-6

CP 1255, ISO 8859-8
CP 857, CP 1254, ISO 8859-9, Macintosh Turkish, TURKISH8
DEC Kanjii, EUC-JIS, Shift-JIS

Group 103
Group 104

CP 932
CP 936, EUG-GB, GB18303
Big 5, CP 950, EUC-CNS
EUCKSC, CP 949

Group 105
Group 106

No conversions needed (only one character set supported)

CP 874, TIS 620

Unicode

No conversions needed (only one character set supported)

Configuring the server

By default, Adaptive Server uses direct conversions to convert data between
different character sets. To use the Unicode conversions, Use sp_configure to
set the enable unicode conversions option to either 1 or 2.

If you set sp_configure enable unicode conversions to 1:

This setting uses Adaptive Server direct conversions or Unicode
conversions. Adaptive Server first checks to see if an Adaptive Server
direct conversion exists for the server and client character set. If a direct
conversion is used; if no direct conversion exists, the Unicode conversion
is used.
Use this setting if the character sets in your client/server system fall into
both columns 1 and 2 in Table 11-1.

If you set sp_configure enable unicode conversions to 2:

This setting uses Unicode conversions only. Adaptive Server uses Unicode
conversions, without attempting to find an Adaptive Server direct
conversion.

System Administration Guide: Volume 1

375

Enabling and disabling character set conversion

Use this setting if the client/server conversions result in a change in the

data length (see Conversions and changes to data lengths on page 378)
If all character sets fall into column 2 in Table 11-1, set enable unicode
conversions to 2 to always use Unicode conversions.

For Adaptive Server version 15.0 and later, the default value for enable unicode
conversions is 1.
If the server default is UTF-8, the server automatically uses Unicode
conversions only.

Enabling and disabling character set conversion

A client that is requesting a connection identifies its character set to Adaptive
Server. Adaptive Server compares the client character set with its default
character set, and if the two names are identical, no conversion is required. If
the names differ, Adaptive Server determines whether it supports conversion
between its default and the clients character set. If it does not, it sends an error
message to the client and continues with the login process. If it does, character
set conversion is automatically enabled. If the default character set of the server
is UTF-8, Unicode conversions are automatically used. If the default is a native
character set, the server uses Adaptive Server direct conversions, unless the
user requests Unicode conversions.
You can disable character set conversion at the server level. You may want to
do this if:

All of your clients are using the same character set as the server default,
and therefore, no conversion is required.

Conversion between the client character set and the server default is not
supported.

You want to store data in the server without changing the encoding.

To disable character set conversion at the server level, set the disable character
set conversion parameter to 1.

You can control character set conversion at the connection level using the set
char_convert command from within a client session. set char_convert off turns
conversion off between a particular client and the server. You may want to set
char_convert off if the client and the server use the same character set, which
makes conversion unnecessary. set char_convert on turns conversion back on.

376

Adaptive Server Enterprise

CHAPTER 11

Configuring Client/Server Character Set Conversions

Characters that cannot be converted

Some characters may not be converted, if:

The character exists (is encoded) in the source character set, but does not
exist in the target character set. For example, the OE ligature is part of the
Macintosh character set (code point 0xCE). This character does not exist
in the ISO 8859-1 character set. If the OE ligature exists in data that is
being converted from the Macintosh to the ISO 8859-1 character set, it
causes a conversion error.

The character exists in both the source and the target character set, but in
the target character set, the character is represented by a different number
of bytes than in the source character set.
For example, 1-byte accented characters (such as , ) are 2-byte
characters in UTF-8; 2-byte Thai characters are 3-byte characters in UTF8. Avoid this limitation by configuring the enable unicode conversion
option to 1 or 2.

Error handling in character set conversion

The Adaptive Server character set conversion reports errors when a character
exists in the clients character set but not in the servers character set, or vice
versa. Adaptive Server must guarantee that data successfully converted on
input to the server can be successfully converted back to the clients character
set when the client retrieves that data. To do this effectively, Adaptive Server
must avoid putting suspect data into the database.
When Adaptive Server encounters a conversion error in the data being entered,
it generates this message:
Msg 2402, Severity 16 (EX_USER):
Error converting client characters into servers
character set. Some character(s) could not be converted.

A conversion error prevents query execution on insert and update statements.

If this occurs, review your data for problem characters and replace them.
When Adaptive Server encounters a conversion error while sending data to the
client, it replaces the bytes of the suspect characters with ASCII question marks
(?). The query batch continues to completion. When the statement is complete,
Adaptive Server sends the following message:

System Administration Guide: Volume 1

377

Conversions and changes to data lengths

Msg 2403, Severity 16 (EX_INFO):

WARNING! Some character(s) could not be converted into
clients character set. Unconverted bytes were changed
to question marks ('?').

Conversions and changes to data lengths

In some cases, converting data between the servers character set and the
clients character set results in a change to the length of the data, for example,
when the character set on one system uses one byte to represent each character
and the character set on the other system requires two bytes per character.
When character set conversion results in a change in data length, there are two
possibilities:

The data length decreases, as in the following examples:

Greek or Russian in multibyte UTF-8 to a single-byte Greek or

Russian character set

Japanese two-byte Hankaku Katakana characters in EUC-JIS to

single-byte characters in Shift-JIS

The data length increases, as in the following examples:

Single-byte Thai to multibyte Thai in UTF-8

Single-byte Japanese characters in Shift-JIS to two-byte Hankaku

Katakana in EUC-JIS

Configuring your system and application

If you are using UTF-8 anywhere in your client/server system, or using a
Japanese character set, you are likely to encounter changes in data length as a
result of character set conversion. You must configure your server to handle
changes in data length. You may also need to set up your client to handle
changes in data length.
1

378

Configure the server to use Unicode conversions. See Configuring the

server on page 375. If the data length increases between the server and
the client, you must also complete steps 2 and 3.

Adaptive Server Enterprise

CHAPTER 11

Configuring Client/Server Character Set Conversions

The client must be using Open Client 11.1 or later. It must inform the
server that it can handle CS_LONGCHAR data at connection time, using
the Open Client ct_capability function.
The capability parameter must be set to CS_DATA_LCHAR and the value
parameter must be set to CS_TRUE, where connection is a pointer to a
CS_CONNECTION structure:
CS_INT capval = CS_TRUE
ct_capability(connection,CS_SET,CS_CAP_RESPONS,
CS_DATA_LCHAR,&capval)

When conversions result in an increase in data length, char and varchar

data are converted to the clients character set and are sent to the client as
CS_LONGCHAR data. The client application must be coded to extract the
data received as CS_LONGCHAR.

Specifying the character set for utility programs

The Sybase utility programs assume that the default character set of the client
platform is the same character set the client is using. However, sometimes the
client character set differs from the character set for the platform. For this
reason, you may need to specify the client character set at the command line.
A command line option for the isql, bcp, and defncopy utilities specifies the
clients character set, and temporarily overrides settings of the LANG variable
or settings in locales.dat.
-J charset_name (UNIX and PC) sets the clients character set to the
charset_name.

If yo omit the client character sets command line flag, the platforms default
character set is used. See the Utility Guide.

Display and file character set command line options

Although the focus of this chapter is on character set conversion between
clients and Adaptive Server, there are two other places where you may need
character set conversion:

Between the client and a terminal

Between the client and a file system

System Administration Guide: Volume 1

379

Specifying the character set for utility programs

Figure 11-4 illustrates the paths and command line options that are available in
the standalone utilities isql, bcp, and defncopy.
Figure 11-4: Where character set conversion may be needed

-a

display_charset

-J client_charset
Terminal
display
Client
File
system

-q datafile_charset
(bcp only)

Adaptive
Server

The -J or /clientcharset command line option specifies the character set used by
the client when it sends and receives character data to and from Adaptive
Server.

Setting the display character set

Use the -a command line option if you are running the client from a terminal
with a character set that differs from the client character set. In Figure 11-4, the
-a option and the -J option are used together to identify the character set
translation file (.xlt file) needed for the conversion.
Use -a without -J only if the client character set is the same as the default
character set.

Setting the file character set

Use the -q command line option if you are running bcp to copy character data
to or from a file system that uses a character set that differs from the client
character set. In Figure 11-4, use the -q or /filecharset option and the -J or
/clientcharset option together to identify the character set translation file (.xlt
file) needed for the conversion.

380

Adaptive Server Enterprise

CH A PTE R

1 2

Diagnosing System Problems

This chapter discusses diagnosing and fixing system problems.

Topic
How Adaptive Server uses error messages

Page
381

Adaptive Server error logging

Backup Server error logging

385
393

Killing processes
Housekeeper functionality

394
398

Shutting down servers

Learning about known problems

401
403

How Adaptive Server uses error messages

When Adaptive Server encounters a problem, it displays an error message
that includes:

A message number, which uniquely identifies the error message

A severity level number between 10 and 24, which indicates the type
and severity of the problem

An error state number, which allows unique identification of the

line of Adaptive Server code at which the error was raised

An error message, which tells you what the problem is, and may
suggest how to fix it

See the Configuration Guide for your platform for a description of the
error log format.
For example, if you try to access a table that does not exist, you see:
select * from publisher
Msg 208, Level 16, State 1:
publisher not found. Specify owner.objectname or use
sp_help to check whether the object exists (sp_help

System Administration Guide: Volume 1

381

How Adaptive Server uses error messages

may produce lots of output).

There may be more than one error message for a single query. If there is more
than one error in a batch or query, Adaptive Server usually reports only the first
one. Subsequent errors are reported the next time you execute the batch or
query.
Error messages are stored in master..sysmessages, which is updated with each
new version of Adaptive Server (and has thousands of rows). Here are the first
few rows (from an Adaptive Server that uses us_english as the default
language):
select error, severity, description
from sysmessages
where error >=101 and error <=106
and langid is null
error severity description
----- -------- -------------------------------------------------101
15 Line %d: SQL syntax error.
102
15 Incorrect syntax near '%.*s'.
103
15 The %S_MSG that starts with %.*s is too long.
Maximum length is %d.
104
15 Order-by items must appear in the select-list if
the statement contains set operators.
105
15 Unclosed quote before the character string '%.*s'.
106
16 Too many table names in the query. The maximum
allowable is %d.
(6 rows affected)

You can query sysmessages to generate a custom list of error messages:

If your server supports more than one language, sysmessages stores each
message in each language. The column langid is NULL for us_english and
matches the syslanguages.langid for other languages installed on the
server.

The sqlstate column stores the SQLSTATE value for error conditions and
exceptions defined in ANSI SQL92.

Message numbers 17000 and higher are system procedure error messages
and message strings.

Error log format

Adaptive Server uses this format for error messages:

382

Adaptive Server Enterprise

CHAPTER 12

Diagnosing System Problems

instance_ID:thread_ID:family_ID:spid date time {server | kernel} message

where:

instance_ID the instance running in a clustered environment. This value

is 00 if Adaptive Server is running in a nonclustered environment.

thread_ID the thread on which the error occured. Always apears as a

four-digit number (for example, 10 is shown as 0010).
Note In process mode, Adaptive Server displays the engine involved for

each log entry. The engine number is indicated by a four-digit number. If

only one engine is online, it is indicated by 00.

family_ID family ID of the originating thread:

In serial processing, you see 00000.

In parallel processing, you see the server process ID number of the

parent of the originating thread.

spid the server process ID of the originating thread:

In serial processing, this is the server process ID number of the thread

that generated the message. If the thread is a system task, the spid
value is 00000.

In parallel processing, you see the server process ID number of the

originating thread.

date appeared in the format yyyy/mm/dd, which allows you to sort error
messages by date.

time shown in 24-hour format, which includes seconds and hundredths

of a second.

server | kernel this entry is for use only by Sybase Technical Support.

message the error message

For example:
00:0024:00000:00000:2010/04/27 10:28:07.82 kernel Thread 24 (LWP 24390) of
Threadpool syb_default_pool online as engine 0

instance_ID = 0 the server is not configured for a clustered environment

thread_ID = 0024 Adaptive Server brought engine 0 online as thread

number 24.

family_ID = 00000 the process is running in serial mode.

System Administration Guide: Volume 1

383

How Adaptive Server uses error messages

spid = 00000 this is a system task.

date and time = 2010/04/27 10:28:07.82.

server | kernel = kernel.

Error messages and message numbers

The combination of message number (error) and language ID (langid) uniquely
identifies each error message. Messages that share the same message number
but have different language IDs indicate translations.
select error, description, langid
from sysmessages
where error = 101
error
----101
101
101

description
langid
-------------------------------------- -----Line %d: SQL syntax error.
NULL
Ligne %1!: erreur de syntaxe SQL.
1
Zeile %1!: SQL Syntaxfehler.
2

(3 rows affected)

The error message text describes the problem. The descriptions often include a
line number, a reference to a type of database object (a table, column, stored
procedure, and so forth), or the name of a particular database object.
In the description field of sysmessages, a percent sign (%) followed by a
character or character string serves as a placeholder for these pieces of data,
which Adaptive Server supplies when it encounters the problem and generates
the error message. %d is a placeholder for a number; %S_MSG is a
placeholder for a type of database object; %.*sall within quotesis a
placeholder for the name of a particular database object. Table 12-1 on
page 385 lists placeholders and what they represent.
For example, the description field for message number 103 is:
The %S_MSG that starts with '%.*s' is too long. Maximum
length is %d.

The actual error message that appears to a user might be:

The column that starts with 'title' is too long. Maximum
length is 80.

For errors that you report to Technical Support, include the numbers, object
types, and object names. (See Reporting errors on page 392.)

384

Adaptive Server Enterprise

CHAPTER 12

Diagnosing System Problems

Variables in error message text

Table 12-1 explains the symbols that appear in the text provided with each
error message explanation:
Table 12-1: Error text symbols key
Symbol

Stands for

%d, %D
%x,%X,%.*x,%lx, %04x, %08lx

Decimal number
Hexadecimal number

%s
%.*s, %*s, %*.s

Null-terminated string
String, usually the name of a particular database object

%S_type
%c

Adaptive Server-defined structure

Single character

%f
%ld

Floating-point number
Long decimal

%lf

Double floating-point number

Adaptive Server error logging

Error messages from Adaptive Server are sent only to the users screen.
The stack trace from fatal error messages (severity levels 19 and higher) and
error messages from the kernel are sent to an error log file. The name of this
file varies; see the configuration documentation for your platform or the Utility
Guide.
Note The error log file is owned by the user who installed Adaptive Server (or
the person who started Adaptive Server after an error log was removed).
Permissions or ownership problems with the error log at the operating system
level can block successful start-up of Adaptive Server.

System Administration Guide: Volume 1

385

Adaptive Server error logging

Adaptive Server creates an error log for you if one does not already exist.
Specify the location of the error log at start-up with the errorlogfile parameter
in the runserver file or at the command line. The Sybase installer utility
configures the runserver file with $SYBASE/$SYBASE_ASEinstall as the
location of the error log if you do not choose an alternate location. If you do
not specify the location in the runserver file or at the command line, the
location of the error log is the directory from which you start Adaptive Server.
For more information about specifying the location of the error log, see
dataserver in the Utility Guide.
Note Always start Adaptive Server from the same directory, or with the

runserver file or the error log flag, so that you can locate your error log.
Each time you start a server, messages in the error log provide information on
the success (or failure) of the start and the recovery of each database on the
server. Subsequent fatal error messages and all kernel error messages are
appended to the error log file. To reduce the size of the error log by deleting old
or unneeded messages, prune the log while Adaptive Server is shut down.

Severity levels
The severity level of a message indicates the type and severity of the problem
that Adaptive Server has encountered. For maximum integrity, when Adaptive
Server responds to error conditions, it displays messages from sysmessages,
but takes action according to an internal table. A few corresponding messages
differ in severity levels, so you may occasionally notice a difference in
expected behavior if you are developing applications or procedures that refer
to Adaptive Server messages and severity levels.
Warning! You can create your own error numbers and messages based on
Adaptive Server error numbers (for example, by adding 20,000 to the Adaptive
Server value). However, you cannot alter the Adaptive Server-supplied system
messages in the sysmessages system table.

You can add user-defined error messages to sysusermessages with

sp_addmessage. See the Reference Manual: Procedures.
Users should inform the system administrator whenever problems that
generate severity levels of 17 and higher occur. The system administrator is
responsible for resolving them and tracking their frequency.

386

Adaptive Server Enterprise

CHAPTER 12

Diagnosing System Problems

If the problem has affected an entire database, the system administrator may
have to use the database consistency checker (dbcc) to determine the extent of
the damage. The dbcc may identify some objects that have to be removed. It
can repair some damage, but you may have to reload the database.
For more information, see the following chapters in System Administration
Guide: Volume 2:

dbcc is discussed in Chapter 10, Checking Database Consistency, in the

System Administration Guide: Volume 2.

Loading a user database is discussed in Chapter 12, Backing Upa and

Restoring User Databases, in System Administration Guide: Volume 2

Loading system databases is discussed in Chapter 13, Restoring the

System Databases, in System Administration Guide: Volume 2.

Severity levels 10 18
Error messages with severity levels 1016 are generated by problems that are
caused by user errors. These problems can be corrected by the user. Severity
levels 17 and 18 do not terminate the users session.
Error messages with severity levels 17 and higher should be reported to the
system administrator or database owner.

Level 10: Status information

Messages with severity level 10 are not errors at all. They provide additional
information after certain commands have been executed and, typically, do not
display the message number or severity level. For example, after a create
database command, Adaptive Server displays a message telling the user how
much of the requested space has been allocated for the new database.

Level 11: Specified database object not found

Messages with severity level 11 indicate that Adaptive Server cannot find an
object that is referenced in a command.
This is often because the user has made a mistake in typing the name of a
database object, because the user did not specify the object owners name, or
because of confusion about which database is current. Check the spelling of
object names, use the owner names if the object is not owned by the user or
dbo, and make sure you are in the correct database.

System Administration Guide: Volume 1

387

Adaptive Server error logging

Level 12: Wrong datatype encountered

Messages with severity level 12 indicate a problem with datatypes. For
example, the user may have tried to enter a value of the wrong datatype in a
column or to compare columns of different and incompatible datatypes.
To correct comparison problems, use the convert function with select. See the
Reference Manual: Building Blocks or the Transact-SQL Users Guide.

Level 13: User transaction syntax error

Messages with severity level 13 indicate that something is wrong with the
current user-defined transaction. For example, the user may have issued a
commit transaction command without having issued a begin transaction, or they
may have tried to roll back a transaction to a savepoint that has not been
defined (sometimes there may be a typing or spelling mistake in the name of
the savepoint).
Severity level 13 can also indicate a deadlock, in which case the deadlock
victims process is rolled back. The user must restart his or her command.

Level 14: Insufficient permission to execute command

Messages with severity level 14 mean that the user does not have the necessary
permission to execute the command or access the database object. they can ask
the owner of the database object, the owner of the database, or the system
administrator to grant them permission to use the command or object in
question.

Level 15: Syntax error in SQL statement

Messages with severity level 15 indicate that the user has made a mistake in the
syntax of the command. The text of these error messages includes the line
numbers on which the mistake occurs and the specific word near which it
occurs.

Level 16: Miscellaneous user error

Most error messages with severity level 16 reflect that the user has made a
nonfatal mistake that does not fall into any of the other categories. Severity
level 16 and higher can also indicate software or hardware errors.

388

Adaptive Server Enterprise

CHAPTER 12

Diagnosing System Problems

For example, the user may have tried to update a view in a way that violates the
restrictions. Another error that falls into this category is unqualified column
names in a command that includes more than one table with that column name.
Adaptive Server has no way to determine which one the user intends. Check
the command syntax and working database context.
Messages that ordinarily have severities greater than 16 show severity 16 when
they are raised by dbcc checktable or dbcc checkalloc so that checks can
continue to the next object. When you are running the dbcc utility, check the
Error Messages and Troubleshooting Guide for information about error
messages between 2500 and 2599 with a severity level of 16.
Note Levels 17 and 18 are usually not reported in the error log. Users should

be instructed to notify the system administrator when level 17 and 18 errors

occur.

Level 17: Insufficient resources

Error messages with severity level 17 mean that the command has caused
Adaptive Server to run out of resources or to exceed some limit set by the
system administrator. The user can continue with their work, although they
may not be able to execute a particular command.
These system limits include the number of databases that can be open at the
same time and the number of connections allowed to Adaptive Server. They are
stored in system tables and can be checked with sp_configure. See Chapter 5,
Setting Configuration Parameters, for more information on changing
configuration variables.
The database owner can correct the level 17 error messages indicating that the
user has run out of space. Other level 17 error messages should be corrected by
the system administrator.

Level 18: Nonfatal internal error detected

Error messages with severity level 18 indicate an internal software bug.
However, the command runs to completion, and the connection to Adaptive
Server is maintained. The user can continue with the work they are doing,
although they may not be able to execute a particular command. An example
of a situation that generates severity level 18 is Adaptive Server detecting that
a decision about the access path for a particular query has been made without
a valid reason.

System Administration Guide: Volume 1

389

Adaptive Server error logging

Since problems that generate such messages do not keep users from their work,
users tend not to report them. However, users should be instructed to inform
the system administrator every time an error message with this severity level
(or higher) occurs so that the system administrator can report them.

Severity levels 19 26
Fatal problems generate error messages with severity levels 19 and higher.
They break the users connection to Adaptive Server (some of the higher
severity levels shut down Adaptive Server). To continue working, the user
must restart the client program.
When a fatal error occurs, the process freezes its state before it stops, recording
information about what has happened. The process is then killed and
disappears.
When the users connection is broken, he or she may or may not be able to
reconnect and resume working. Some problems with severity levels in this
range affect only one user and one process. Others affect all the processes in
the database. In some cases, the system administrator must restart Adaptive
Server. These problems do not necessarily damage a database or its objects, but
they can. They may also result from earlier damage to a database or its objects.
Other problems are caused by hardware malfunctions.
Error messages from the kernel are directed to the error log file.

Level 19: Adaptive Server fatal error in resource

Error messages with severity level 19 indicate that some nonconfigurable
internal limit has been exceeded and that Adaptive Server cannot recover
gracefully. You must reconnect to Adaptive Server.

Level 20: Adaptive Server fatal error in current process

Error messages with severity level 20 indicate that Adaptive Server has
encountered a bug in a command. The problem has affected only the current
process, and the database is unlikely to have been damaged. Run dbcc
diagnostics. The user must reconnect to Adaptive Server.

390

Adaptive Server Enterprise

CHAPTER 12

Diagnosing System Problems

Level 21: Adaptive Server fatal error in database processes

Error messages with severity level 21 indicate that Adaptive Server has
encountered a bug that affects all the processes in the current database.
However, it is unlikely that the database itself has been damaged. Restart
Adaptive Server and run dbcc diagnostics. The user must reconnect to
Adaptive Server.

Level 22: Adaptive Server fatal error: Table integrity suspect

Error messages with severity level 22 indicate that the table or index specified
in the message has been previously damaged by a software or hardware
problem.
The first step is to restart Adaptive Server and run dbcc to determine whether
other objects in the database are also damaged. Whatever the report from dbcc
may be, The problem may be only in the cache, and not on the disk itself. If so,
restarting Adaptive Server fixes the problem.
If restarting does not help, then the problem is on the disk as well. Sometimes,
the problem can be solved by dropping the object specified in the error
message. For example, if the message tells you that Adaptive Server has found
a row with length 0 in a nonclustered index, the table owner can drop the index
and re-create it.
Adaptive Server takes any pages or indexes offline that it finds to be suspect
during recovery. Use sp_setsuspect_granularity to determine whether recovery
marks an entire database or only individual pages as suspect. See
sp_setsuspect_granularity in the Reference Manual: Procedures.
The user must reconnect to Adaptive Server.

Level 23: Fatal error: Database integrity suspect

Error messages with severity level 23 indicate that the integrity of the entire
database is suspect due to previous damage caused by a software or hardware
problem. Restart Adaptive Server and run dbcc diagnostics.
Even when a level 23 error indicates that the entire database is suspect, the
damage may be confined to the cache, and the disk itself may be fine. If so,
restarting Adaptive Server with startserver fixes the problem.

System Administration Guide: Volume 1

391

Adaptive Server error logging

Level 24: Hardware error or system table corruption

Error messages with severity level 24 reflect a media failure or (in rare cases)
the corruption of sysusages. The system administrator may have to reload the
database. You may need to call your hardware vendor.

Level 25: Adaptive Server internal error

Users do not see level 25 errors; this level is used only for Adaptive Server
internal errors.

Level 26: Rule error

Error messages with severity level 26 reflect that an internal locking or
synchronization rule has been broken. You must shut down and restart
Adaptive Server.

Reporting errors
When you report an error to Sybase Technical Support, include:

392

The message number, level number, and state number.

Any numbers, database object types, or database object names that are
included in the error message.

The context in which the message was generated, that is, the command that
was running at the time. You can help by providing a hard copy of the error
log.

Adaptive Server Enterprise

CHAPTER 12

Diagnosing System Problems

Backup Server error logging

Like Adaptive Server, Backup Server creates an error log if one does not
already exist. Specify the location of the error log at start-up with the
error_log_file parameter in the runserver file or at the command line. The
Sybase installer configures the runserver file with $SYBASE/install as the
location of the error log if you do not choose an alternate location during
installation. If you do not specify the location in the runserver file or at the
command line, the location of the error log is the directory from which you start
Backup Server. Use the backupserver -V option (bcksvr -V on Windows NT) to
limit the messages printed to the error log. See the sections describing Backup
Server in the Utility Guide.
Backup Server error messages are in this form:
MMM DD YYY: Backup Server:N.N.N.N: Message Text

Backup Server message numbers consist of four integers separated by periods,

in the form N.N.N.N. Messages in the form N.N.N are sent by Open Server.
The four components of a Backup Server error message are
major.minor.severity.state:

The major component generally indicates the functional area of the

Backup Server code where the error occurred:

1 system errors.

2 Open Server event errors.

3 Backup Server remote procedure call errors.

4 I/O service layer errors.

5 network data transfer errors.

6 volume-handling errors.

7 option-parsing errors.

Major error categories 1 6 may result from Backup Server internal errors
or a variety of system problems. Major errors in category 7 are almost
always due to problems in the options you specified in your dump or load
command.

minor numbers are assigned in order within a major category.

severity is:

1 informational, no user action necessary.

System Administration Guide: Volume 1

393

Killing processes

2, 3 an unexpected condition, possibly fatal to the session, has

occurred. The error may have occurred with usage, environment, or
internal logic, or any combination of these factors.

4 an unexpected condition, fatal to the execution of the Backup

Server, has occurred. The Backup Server must exit immediately.

state codes have a one-to-one mapping to instances of the error report

within the code. If you need to contact Technical Support about Backup
Server errors, the state code helps determine the exact cause of the error.

Killing processes
A process is a unit of execution carried out by Adaptive Server. Each process
is assigned a unique process identification number when it starts. This number
is called a spid. These numbers are stored, along with other information about
each process, in master..sysprocesses. Processes running in a parallelprocesses environment create child processes, each of which has its own spids.
Several processes create and assign spids: starting Adaptive Server, login tasks,
checkpoints, the housekeeper tasks, and so on. You can see most of the
information by running sp_who.
Running sp_who on a single-engine server shows the sp_who process running
and all other processes that are runnable or in one of the sleep states. In multiengine servers, there can be a process running for each engine.
The kill command gets rid of an ongoing process. The most frequent reason for
killing a process is that it interferes with other users, and the person responsible
for running it is not available. The process may hold locks that block access to
database objects, or there may be many sleeping processes occupying the
available user connections. A system administrator can kill most running or
"runnable" processes, including those that are waiting for:

394

An alarm, such as a waitfor command

Network sends or receives

A lock

Synchronization messages from another process in a family

Adaptive Server Enterprise

CHAPTER 12

Diagnosing System Problems

Adaptive Server allows you to kill processes only if it can cleanly roll back any
uncompleted transactions and release all system resources that are used by the
process. For processes that are part of a family, killing any of the child
processes also kills all other processes in the family. However, it is easiest to
kill the parent process. For a family of processes, the kill command is detected
more quickly if the status of the child processes is sync sleep.
Table 12-2 shows the status values that sp_who reports and when the kill
command takes effect.
Table 12-2: Status values reported by sp_who
Status

Indicates
Waiting on a network read.

Effects of kill command

Immediate.

alarm sleep

Waiting on a network send.

Waiting on an alarm such as:

Immediate.
Immediate.

lock sleep

Waiting on a lock acquisition.

Immediate.

sync sleep

Waiting on a synchronization message from

another process in the family.

Immediate. Other processes in the family

must also be brought to state in which they
can be killed.

sleeping

Waiting on a disk I/O, or some other resource.

Probably indicates a process that is running, but
doing extensive disk I/O

Killed when it wakes up, usually

immediate; a few sleeping processes do not
wake up and require a server restart to clear.

runnable

In the queue of runnable processes.

Actively running on one of the server engines.

Immediate.
Immediate.

infected

Server has detected serious error condition;

extremely rare.

kill command not recommended. Server

background

A process, such as a threshold procedure, run by

Adaptive Server rather than by a user process.

log suspend

Processes suspended by reaching the last-chance

threshold on the log.

Immediate; use kill with extreme care.

Recommend a careful check of
sysprocesses before killing a background
process.
Immediate.

recv sleep
send sleep

waitfor delay "10:00"

running

restart probably required to clear process.

Only system administrators can issue the kill command; permission to use it
cannot be transferred.
The syntax is:
kill spid

You can kill only one process at a time, but you can perform a series of kill
commands in a batch. For example:
1> kill 7

System Administration Guide: Volume 1

395

Killing processes

2> kill 8
3> kill 9
4> go

A kill command is irreversible and cannot be included in a user-defined

transaction. spid must be a numeric constant; you cannot use a variable. Here
is some sample output from sp_who:
sp_who
fid spid status
loginame origname hostname
blk_spid dbname
tempdbname
cmd
block_xloid
threadpool
--- ---- ---------- -------- -------- ---------------- -------- ------------------------------- ----------------------------0
1 recv sleep
howard
howard
svr30eng
0 master
tempdb
AWAITING COMMAND
0
syb_default_pool
0
2 sleeping
NULL
NULL
0 master
tempdb
NETWORK HANDLER
0
syb_default_pool
0
3 sleeping
NULL
NULL
0 master
tempdb
DEADLOCK TUNE
0
syb_default_pool
0
4 sleeping
NULL
NULL
0 master
tempdb
MIRROR HANDLER
0
syb_default_pool
0
5 sleeping
NULL
NULL
0 master
tempdb
CHECKPOINT SLEEP
0
syb_default_pool
0
6 sleeping
NULL
NULL
0 master
tempdb
HOUSEKEEPER
0
syb_default_pool
0
7 recv sleep
bill
bill
bigblue
0 master
tempdb
AWAITING COMMAND
0
syb_default_pool
0
8 recv sleep
wilbur
wilbur
hazel
0 master
tempdb
AWAITING COMMAND
0
syb_default_pool
0
9 recv sleep
joan
joan
luv2work
0 master
tempdb
AWAITING COMMAND
0
syb_default_pool
0
10 running
foote
foote
svr47hum
0 master
tempdb
SELECT
0
syb_default_master
(10 rows affected, return status = 0)

In the example above, processes 26 cannot be killed: they are system

processes. The login name NULL and the lack of a host name identify
processes them as system processes. NETWORK HANDLER, MIRROR
HANDLER, HOUSEKEEPER, and CHECKPOINT SLEEP (or, rarely,
CHECKPOINT) always appear in sp_who output. AUDIT PROCESS appears
if auditing is available.
Processes 1, 8, 9, and 10 can be killed, since they have the status values recv
sleep, send sleep, alarm sleep, and lock sleep.

396

Adaptive Server Enterprise

CHAPTER 12

Diagnosing System Problems

In sp_who output, you cannot tell whether a is recv sleep belongs to a user
who is using Adaptive Server and may be pausing to examine the results of a
command, or whether a user has restarted a PC or other terminal, and left a
stranded process. Query the sysprocesses table to learn more about
questionable processes. For example, this query shows the host process ID and
client software used by process 8:
select hostprocess, program_name
from sysprocesses
where spid = 8
hostprocess program_name
----------- ---------------3993
isql

This query, plus the information about the user and host from the sp_who
results, provides additional information for tracking down the process from the
operating system level.

Using kill with statusonly

The kill ...statusonly command reports on the progress of a server process ID
(spid) in rollback status. It does not terminate the spid. The statusonly report
displays the percent of rollback completed and the estimated length of time in
seconds before the rollback completes. To track the progress of a rollback, you
must run kill...with statusonly multiple times:
kill spid with statusonly

Where spid is the number of the process you are terminating.

For example, the following reports on the process of the rollback of spid
number 13:
kill 13 with statusonly
spid: 13 Transaction rollback in progress. Estimated rollback completion: 17%
Estimated time left: 13 seconds

If the rollback of the spid has completed when you issue kill...statusonly or if
Adaptive Server cannot roll back the specified spid, kill...statusonly returns the
following message:
Status report cannot be obtained. KILL spid:nn is not
in progress.

System Administration Guide: Volume 1

397

Housekeeper functionality

Using sp_lock to examine blocking processes

In addition to sp_who, sp_lock can help identify processes that are blocking
other processes. If the blk_spid column in the sp_who report indicates that
another process has been blocked while waiting to acquire locks, sp_lock can
display information about the blocking process. For example, process 10 in the
sp_who output above is blocked by process 7. To see information about process
7, execute:
sp_lock 7

For more information about locking in Adaptive Server, see the Performance
and Tuning Series: Locking and Concurrency Control.

Housekeeper functionality
The housekeeper task provides important functionalities:

The housekeeper consists of three tasks: housekeeper wash, housekeeper

garbage collection, and housekeeper chores. sp_who displays all three
tasks in its output as HK WASH, HK GC, and HK CHORES:

sp_who
fid
spid
status
loginame
origname
hostname
blk_spid dbname tempdbname cmd
block_xloid threadpool
---- ----- ------------------------------------------- ------- --------- ------------- --------- ------------0
8
sleeping
NULL
NULL
luv2work
0
master
tempdb
HK WASH
0 syb_default_pool
0
9
sleeping
NULL
NULL
NULL
0
master
tempdb
HK GC
0 syb_default_pool
0
10
sleeping
NULL
NULL
NULL
0
master
tempdb
HK CHORES
0 syb_default_pool

The general automatic restart of housekeeper-related system tasks: you

need not restart the server if these system tasks quit unexpectedly.

A system administrator can change all housekeeper task priorities.

sp_showpsexe, as well as sp_who, recognizes all three housekeeper

names.
For more information about sp_who and sp_showpsexe, see the Reference
Manual: Procedures.

398

Adaptive Server Enterprise

CHAPTER 12

Diagnosing System Problems

Housekeeper wash
Washing buffers is an optional task that, if enabled, runs only during idle times.
You can turn off this task using the configuration parameter housekeeper free
write percent. The housekeeper wash task is the only housekeeper task for
which you use this configuration parameter.

Housekeeper chores
The housekeeper chores task th runs only at idle times, and does not use a
common configuration parameter. It manages miscellaneous chores, such as:

Flushing table statistics.

Flushing account statistics.

Handling timeout of detached transactions. You can turn off this task using
the configuration parameter dtm detach timeout period.

Checking licence usage. You can turn off this task using the configuration
parameter license information.

Housekeeper garbage collection

There are two forms of garbage collection, lazy and aggressive. These terms
describe two distinct tests for finding empty pages.

Lazy garbage collection refers to an inexpensive test to find empty pages.

This test may not be effective during long-running transactions, and empty
pages may accumulate. Lazy garbage collection is inexpensive to use, but
can lower performance, which is affected by the fragmentation of
allocated table space, and by the accumulation of empty pages that must
be evaluated during queries.

Aggressive garbage collection refers to a sophisticated test for empty

pages. This test is more expensive than the lazy garbage collection test,
because it checks each deleted row in a page to determine whether the
deleted transactions are committed.
Use the enable housekeeper GC configuration parameter to configure the
delete command and the housekeeper garbage collection task for
aggressive or lazy garbage collection.

System Administration Guide: Volume 1

399

Housekeeper functionality

The aggressive housekeeper garbage collection self-tunes the frequency

with which the housekeeper garbage collection task examines the
housekeeper list, so that the frequency of examination matches the rate at
which the application generates empty pages.

Running at user priority

The housekeeper garbage collection task typically operates at the priority level
of an ordinary user, competing for CPU time with ordinary user tasks. This
prevents the list of empty pages from growing faster than the housekeeper can
delete them. However, if Adaptive Server is configured for threaded mode, use
the sp_bindexeclass 'sv' object type to change the server-wide priority setting
for the housekeeper wash task. Set the priority to the EC1, EC2, EC3 level or
define a new user-created execution class. This example sets the housekeeper
wash task to the highest, EC1, priority:
sp_bindexeclass 'HK WASH', 'sv', NULL, 'EC1'

Use sp_setpsexe to set the priority level for the housekeeper task for the
session if Adaptive Server is configured for threaded or process mode.
This example sets the priority level for the housekeeper wash task (with a spid
of 8) to HIGH for the current session:
sp_setpsexe 8, 'priority', 'HIGH'

See the Reference Manual: Procedures.

Configuring enable housekeeper GC

To configure Adaptive Server for garbage collection task, use:
sp_configure "enable housekeeper GC", value

The valid values for the enable housekeeper GC configuration parameter are:

400

0 disables the housekeeper garbage collection task, but enables lazy

garbage collection by the delete command. You must use reorg
reclaim_space to deallocate empty pages. This is the cheapest option with
the lowest performance impact, but it may cause performance problems if
many empty pages accumulate. Sybase recommends that you do not use
this value.

Adaptive Server Enterprise

CHAPTER 12

Diagnosing System Problems

1 enables lazy garbage collection, by both the housekeeper garbage

collection task and the delete command. This is the default value. If more
empty pages accumulate than your application allows, consider options 4
or 5. You can use the optdiag utility to obtain statistics of empty pages.

2 reserved for future use.

3 reserved for future use.

4 enables aggressive garbage collection for both the housekeeper

garbage collection task and the delete command. This option is the most
effective, but the delete command is the most expensive. This option is
ideal if the deletes on your dataonly locked tables are in a batch.

5 enables aggressive garbage collection for the housekeeper, and lazy

garbage collection by delete. This option is less expensive for deletes than
option 4. This option is suitable when deletes are caused by concurrent
transactions.

For example, enter:

sp_configure "enable housekeeper GC", 4

Using the reorg command

Garbage collection is most effective when you set enable housekeeper GC to 4
or 5. Sybase recommends that you set the parameter value to 5. However, if
performance considerations prevent setting this parameter to 4 or 5, and you
have an accumulation of empty pages, run reorg on the affected tables. You can
obtain statistics on empty pages through the optdiag utility.
When the server is shut down or crashes, requests to deallocate pages that the
housekeeper garbage collection task has not yet serviced are lost. These pages,
empty but not deallocated by the housekeeper garbage collection task, remain
allocated until you remove them by running reorg.
See Chapter 9, Using the reorg Command, in the System Administration
Guide: Volume 2.

Shutting down servers

A system administrator can shut down Adaptive Server or Backup Server
using:

System Administration Guide: Volume 1

401

Shutting down servers

shutdown [backup_server_name] [with {wait|nowait}]

The default for the shutdown command is with wait. That is, shutdown and
shutdown with wait do exactly the same thing.

Shutting down Adaptive Server

If you do not provide a server name, shutdown shuts down the Adaptive Server
you are using. When you issue a shutdown command, Adaptive Server:
1

Disables logins, except for system administrators

Performs a checkpoint in each database, flushing pages that have changed

from memory to disk

Waits for currently executing SQL statements or procedures to finish

In this way, shutdown minimizes the amount of work that automatic recovery
must do when you restart Adaptive Server.
The with nowait option shuts down Adaptive Server immediately. User
processes are aborted, and recovery may take longer after a shutdown with
nowait. You can help minimize recovery time by issuing a checkpoint command
before you issue a shutdown with nowait command.

Shutting down a Backup Server

To shut down a Backup Server, include the Backup Server name:
shutdown SYB_BACKUP

The default is with wait, so any dumps or loads in progress complete before the
Backup Server process halts. After you issue a shutdown command, no new
dump or load sessions can be started on the Backup Server.
To see the names of the Backup Servers that are accessible from your Adaptive
Server, execute sp_helpserver. Use the value in the name column in the
shutdown command. You can shut down a Backup Server only if it is:

Listed in sysservers on your Adaptive Server, and

Listed in your local interfaces file.

Use sp_addserver to add a Backup Server to sysservers.

402

Adaptive Server Enterprise

CHAPTER 12

Diagnosing System Problems

Checking for active dumps and loads

To see the activity on your Backup Server before executing a shutdown
command, run sp_who on the Backup Server:
SYB_BACKUP...sp_who
spid
status
loginame
------------------1
sleeping
NULL
2
sleeping
NULL
3
runnable
NULL
4
runnable
NULL
5
running
sa

hostname
---------NULL
NULL
NULL
NULL
heliotrope

blk
--0
0
0
0
0

cmd
-------------CONNECT HANDLER
DEFERRED HANDLER
SCHEDULER
SITE HANDLER
NULL

Using nowait on a Backup Server

The shutdown backup_server with nowait command shuts down the Backup
Server, regardless of current activity. Use it only in severe circumstances. It can
leave your dumps or loads in incomplete or inconsistent states.
If you use shutdown with nowait during a log or database dump, check for the
message indicating that the dump completed. If you did not receive this
message, or if you are not sure whether the dump completed, your next dump
should be a dump database, not a transaction dump. This guarantees that you
are not relying on possibly inconsistent dumps.
If you use shutdown with nowait during a load of any kind, and you did not
receive the message indicating that the load completed, you may not be able to
issue further load transaction commands on the database. Run a full database
consistency check (dbcc) on the database before you use it. You may have to
reissue the full set of load commands, starting with load database.

Learning about known problems

The release bulletin is a valuable resource for learning about known problems
or incompatibilities with Adaptive Server and Backup Server. Reading the
release bulletin in advance can save you the time and guesswork of
troubleshooting known problems.

System Administration Guide: Volume 1

403

Learning about known problems

404

Adaptive Server Enterprise

Index

Symbols
% (percent sign)
error message placeholder 384
?? (question marks)
for suspect characters 377
(quotation marks)
enclosing parameter values 10
@@client_csexpansion global variable

366

Numerics
7-bit ASCII character data, character set conversion for
371

A
abstract plan cache configuration parameter 81
abstract plan dump configuration parameter 81
abstract plan load configuration parameter 81
abstract plan replace configuration parameter 82

adding
database devices 219, 308314
date strings 363
months of the year 363
remote logins 300302
remote servers 294299
users to a database 219
additional network memory configuration parameter
82
address, server 14
aggressive garbage collection 399
priority level 400
aggressive housekeeper 399
aliases
server 296
allocation
pages 308

System Administration Guide: Volume 1

units 308
units See also size; space allocation
allow backward scans configuration parameter 85
allow nested triggers configuration parameter 85
allow procedure grouping configuration parameter
86
allow remote access configuration parameter 86
allow resource limits configuration parameter 87
allow sendmsg configuration parameter 87
allow sql server async i/o configuration parameter 87
allow updates configuration parameter (now called
allow updates to system tables) 12
allow updates to system tables configuration parameter
12, 88
alter database command
omitting database device and 315, 316
system tables and 290
alternate languages. See languages, alternate
application design 219
Arabic character set support 333
ASCII characters
character set conversion and 371
asynchronous I/O
limiting Server requests for 166
asynchronous prefetch
configuring 146
@@char_convert global variable 366
@@client_csid global variable 366
@@client_csname global variable 366
@@langid global variable 368
@@language global variable 368
@@max_connections global variable 218
@@maxcharlen global variable 366
@@ncharsize global variable 366
audit queue size configuration parameter 90
audit trail 28
stacktrace of error messages 385
auditing
queue, size of 90
sybsecurity database 28

405

Index

auditing configuration parameter 91

automatic cluster takeover configuration parameter
average cap size configuration parameter 89

91

B
Backing up
master database 53
backtracing errors. See error logs
Backup Server
error messages 393
shutting down 402
tape retention in days configuration parameter 269
backups 3942
hints 3942
Baltic character set support 333
bcp (bulk copy utility)
character set conversion and 379, 380
sort order changes and 355
Big 5
similarities to CP 950 333
binary sort order of character sets
character set changes and database dumps 355
bytes
character 377

C
cache partitions
configuring 145, 146
cache, procedure 236
caches, data
database integrity errors and 391
calls, remote procedure 293306
timeouts 297
caps per ccb configuration parameter 92
changing
See also updating
configuration parameters 76, 305
database options 321323
system tables, dangers of 9, 12
@@char_convert global variable 366
character set conversions 369, 378379
character sets 109

406

See also Japanese character sets

Arabic 333
Baltic 333
changing 352
conversion between client and file system 379
conversion between client and server 370372
conversion between client and terminal 379
conversion errors 377
conversion paths supported 370377
Cyrillic-script 333
default 338
definition 331
definition files 365
Eastern European 333
encoding in different 369
European currency symbol and 334
for language groups 333
Greek 333
Hebrew 333
ID number 109
Japanese 333
Korean 333
multibyte 361
multibyte, changing to 362
reindexing after configuring 359362
Russian 333
Simplified Chinese 333
Thai 333
Traditional Chinese 333
translation files, terminal-specific 365, 380
Turkish 333
Unicode 333
upgrading text values after changing 361
Vietnamese 333
Western European 333
characters
that cannot be converted 377
charset.loc file 365
charsets directory 365
checkpoint process 240
recovery interval parameter and 241
trunc log on chkpt database option 240
checktable option, dbcc 360
CIPC large message pool size configuration parameter
93

Adaptive Server Enterprise

Index
CIPC regular message pool size configuration

parameter

94

cis bulk insert batch size configuration parameter

94
cis connect timeout configuration parameter 95
cis cursor rows configuration parameter 95
cis idle connectin timeout configuration parameter

96
cis packet size configuration parameter 96
cis rpc handling configuration parameter 97

@@client_csid global variable 366

@@client_csname global variable 366
clients
character set conversion 379
Closed Problem Reports 403
cluster heartbeat interval configuration parameter
97
cluster heartbeat retries configuration parameter 98
cluster vote timeout configuration parameter 98, 99
cntrltype option
disk init 314
column default cache size configuration parameter
100
column name
unqualified 389
command
delete 399
disk resize 318
reorg reclaim_space 400
common.loc file 367
comparing values
datatype problems 388
compression info pool size configuration parameter
100
compression memory size configuration parameter
99, 100
configuration (server)
character sets 352
message language 352358
sort orders 352361
configuration file
default name and location 62
specifying at start-up 68
storage of configured value 62
configuration file configuration parameter 101, 122,
253, 254

System Administration Guide: Volume 1

configuration parameter
max native threads per engine 171
rtm thread idle wait period 246
configuration parameter, max transfer history

181

configuration parameters 80279

changing 305
default settings of 61
dtm detach timeout period 399
help information on 65
399
listing values of 66
remote logins and 86, 305
connecting to Adaptive Server 14
connections
directory services 15
interfaces files 14
maximum user number 218
consistency
checking databases 41
copying selected data
See insert command; select command
cost of a cpu unit configuration parameter 102
cost of a logical io configuration parameter 101
cost of a physical io configuration parameter 102
CP 1252
similarities to ISO 8859-1 333
CP 950
similarities to Big 5 333
cp437 character set 109
cp850 character set 109
CPR files 403
cpu accounting flush interval configuration parameter
102
cpu grace time configuration parameter 104
create database command
default database size configuration parameter and
109
model database and 25
omitting database device and 315, 316
system tables and 7
create index command 285, 291
create procedure command 12
create table command 285
creating
database objects 285
databases 52
housekeeper free write percent

407

Index

guest users 52
master database 287
model database 287
segments 287
stored procedures 12
system procedures 12
system tables 7
tempdb database 287
user-defined error messages 386
users 53
cross-platform dump and load, handling suspect partitions
363
cs_connection command, number of user connections
and 219
current audit table configuration parameter 104
current database 387
cursors
row count, setting 95
cyrillic character set support 333

D
data caches
configuring partitions 145, 146
database integrity errors and 391
data dictionary. See system tables
database administration 15
database device
space See segments; space allocation
database devices 307
See also disk mirroring; dump devices; master device
adding 308314
default 316317
dropping 316
fragments 290
information about 315
initializing 307314
names of 288, 309
number of server-usable 196
placing objects on 287
database object owners 4
See also database owners
permissions 5
tasks of 4
database objects

408

See also individual object names

access permissions for 5
assigning to devices 286
controlling user creation of 24
creating 24, 285
errors affecting 391
finding 387
maximum number of open 210
ownership 4
database options 321323
listing 322
setting 323
showing settings 323
database owners 4
error responsibilities of 387, 389
login name 2, 4
permissions of 4
tasks of 4
database size configuration variable 52
databases
See also database objects; user databases
backing up 25, 40, 53
creating 52
default 24
default storage for 22, 316
dumping 40
errors affecting 391
guest users 52
integrity concerns 391
loading after character set change 355
loading after sort order change 355
new 25
number of open 207
options 321323
size 25
system 21
date parts
alternate language 363
dates
adding date parts 363
alternate language 363
display formats 367
format in error messages 383
days
alternate language 363
dbcc (database consistency checker) 41

Adaptive Server Enterprise

Index
database damage and
when to use 391
DB-Library programs

387, 391

number of user connections and 219

dbo user name 2, 4
dbprocess command, number of user connections
and 219
deadlock checking period configuration parameter
105
deadlock pipe active configuration parameter 106
deadlock pipe max messages configuration
parameter 107
deadlock retries configuration parameter 107
deadlocks 388
descending scans and 85
deckanji character set 109
default character set id configuration parameter 109
default database devices
designating 316
default database size configuration parameter 109
default exp_row_size percent configuration
parameter 110
default fill factor percent configuration parameter
110
default language id configuration parameter 111
default network packet size configuration parameter
111
default segment 287
default settings
changing character set 353362
changing sort order 355361
character set ID number 109
configuration parameters 61
databases 24
language 111
permissions 25
sort order 113
system databases at installation 288
default sortorder id configuration parameter 113
default XML sortorder configuration parameter 113
defaulton | defaultoff option, sp_diskdefault 317
defaults
See also database objects
defncopy utility command
See also Utility Programs manual
character set conversion and 379, 380

System Administration Guide: Volume 1

delete command 399

deleting
files 316
users 54
descending scans
deadlocks and 85
detached transactions 117
development server 34
device fragments 290
device shrinkage, disk resize 319
devices 307
See also database devices; dump devices; master
device
adding 308314
dropping 316
information listings on 315
initializing 307314
names for physical 309
number of user connections and 218, 219
using separate 286
direct updates
to system tables 88
directory services in libtcl.cfg file 15
directory structure
character sets 365
internationalization files 365
localization files 368
*.loc files 368
dirty pages 240
disable character set conversions configuration
parameter 114
disable disk mirroring configuration parameter 115
disable varbinary truncation configuration parameter
116
disk controllers 314
disk devices
See also database devices; dump devices; space
allocation
disk I/O
configuration parameters for 200
database loads and 169, 196, 203
disk i/o structures configuration parameter 116
disk init command 284, 290, 308314
disk mirror command 285
disk mirroring
disabling 115

409

Index

enabling 115
recovery and 286
status in sysdevices table 315
disk reinit command
See also disk init command
disk resize 284, 318
device shrinkage 319
insufficient disk space 319
minimum size 318
mirroring 318
syntax 318
using 318
disks See database devices; devices; dump devices
Distributed Transaction Management (DTM) 28
Distributed Transaction Processing (DTP) 28
DMA object pool size configuration parameter 116
drop logins option, sp_dropserver 299
dropping
database devices 316
dump devices 316
master device from default space pool 317
remote logins 299, 300
servers 299
dsync option
disk init 316
dtm detach timeout period configuration parameter 117,
399
dtm lock timeout period configuration parameter 117
dump database command
disk init and 308
master database and 40
model database and 25
dump devices
dropping 316
information about 315
sysdevices table and 289
dump on conditions configuration parameter 119
dump, database 40
dynamic allocation on demand configuration parameter
119
dynamic configuration parameters 62

E
Eastern Europe

410

character set support 333

empty pages, accumulating 400
enable backupserver HA configuration parameter
120
enable cis configuration parameter 120, 123, 136
enable DTM configuration parameter 121
enable encrypted columns configuration parameter
122
enable functionality group configuration parameter
124
enable HA configuration parameter for high availability
125
enable housekeeper GC configuration parameter
126, 400
enable hp posix async i/o configuration parameter
127
enable i/o fencing configuration parameter 128
enable inline default sharing configuration parameter
125, 133
enable java configuration parameter 122, 128
enable job scheduler configuration parameter 129
enable ldap user auth configuration parameter 129
enable literal autoparam configuration parameter 130
enable logins during recovery configuration parameter
130
enable merge join configuration parameter 131
enable metrics capture configuration parameter 131
enable monitoring configuration parameter 131
enable pam user auth configuration parameter 132
enable query tuning mem limit configuration parameter
133
enable real time messaging configuration parameter
134
enable rep agent threads configuration parameter
134
enable row level access control configuration
parameter 135
enable semantic partitioning configuration parameter
135
enable sort-merge join and JTC configuration
parameter 136
enable SQL debugger configuration parameter 136
enable stmt cache monitoring configuration parameter
137
enable surrogate processing configuration parameter
137

Adaptive Server Enterprise

Index
enable unicode conversion configuration parameter

137
enable unicode conversions configuration parameter

375
enable unicode normalization configuration

parameter

138

enable webservices configuration parameter 138

enable xact coordination configuration parameter

139
enable xml configuration parameter

139
encoding characters 369
encryption, password 145
engine memory log size configuration parameter
140
engines
number of 175
error logs 43, 390
creation and ownership 385
format 382
location 13
purging 386
error messages 384392
altering server-provided 367, 386
character conversion 377
creating user-defined 386
for fatal errors 390392
numbering of 384
severity levels of 386392
user-defined 386
errorlog pipe active configuration parameter 140
errorlog pipe max messages configuration parameter
141
errors
See also error logs; error messages
character conversion 377
fatal 390392
logging 385
multiple 382
reporting of 392
server responses to 381392
state numbers 381
types of information logged 13
user 387, 387390
esp execution priority configuration parameter 141
esp execution stacksize configuration parameter
142

System Administration Guide: Volume 1

esp unload dll configuration parameter

142
eucjis character set 109
European currency symbol
character sets 334
event buffers per engine configuration parameter
143
event log computer name configuration parameter
143
event logging configuration parameter 144
executable code size + overhead configuration
parameter 145
execution
ESPs and XP Server priority 141
expired passwords 269
extended cache size configuration parameter 145
extended stored procedures
configuration parameters 141282
extended UNIX character set 109

F
failures, media 392
fatal errors
backtrace from kernel 385, 390
error messages for 390392
severity levels 19 and up 390392
file descriptors 218
maximum per-process configured for your operating
system 221
files
character set translation (.xlt) 365
Closed Problem Reports (CPRs) 403
deleting 316
error log 13, 385
interfaces 14
internationalization 365
libtcl.cfg file 15
localization 366367
System Problem Reports (SPRs) 403
fillfactor
default fill factor percent configuration parameter
110
finding
database objects 387

411

Index

FIPS login password encryption configuration parameter

145
fix_text option, dbcc 361362
For load 52
formats
date, time, and money 367
locale, unsupported 363364
formulas
user requirements and 219
forwarded rows
reducing with default exp_row_size configuration
parameter 110
fragments, device space 290
french
character set support 333

G
garbage collection
aggressive test 399
lazy test 399
garbage collector
configuring aggressive 400
housekeeper utility 399
German
character set support 333
global async prefetch limit configuration parameter 146
global cache partition number configuration parameter
146
granting
access permissions 4
object creation permissions 4
Greek
character set support 333
groups
language 333
public 53
Guest users
creating 52
guest users
databases 52
sample databases and 29

H
handling suspect partitions 362
hardware
errors 392
hash buckets (lock) 163
heap memory per user configuration parameter 147
Hebrew
character set support 333
high availability
installhasvss script 125
insthasv script 125
setting enable HA 125
histogram tuning factor configuration parameter 147
housekeeper chores 399
configuration parameter license information 399
housekeeper free write percent configuration
parameter 149, 399
housekeeper garbage collector 399
housekeeper task
configuring 149
space reclamation and 126
statistics flushing 149
housekeeper utility
functionality 398
housekeeper wash, housekeeper garbage collection,
housekeeper chores 398
three tasks 398
wash 399
wash task 149

I
i/o accounting flush interval configuration parameter

151
i/o batch sizet configuration parameter 151
i/o polling process count configuration parameter

152
IBM character set
Icons 49

109

identity burning set factor configuration parameter

153
identity grab size configuration parameter 154
identity reservation size configuration parameter 155
idle migration timeout configuration parameter 155

IDs, user

412

Adaptive Server Enterprise

Index
system procedures and 12
impersonating a user. See setuser command
index descriptors
maximum number open 208
indexes
character set changes 361
character-based 359
default fill factor percent percentage for 110
object allocation maps of 206
rebuilding 359
sort order changes 360
suspect 359, 391
information (server)
configuration parameters 66
database devices 315
database options 322323
devices 315
dump devices 315
error messages 384392
problems 385
remote server logins 305
remote servers 299
information messages (server). See error messages;
severity levels
initializing
database devices 307314
installation, server
interfaces file 15
status after 287
installhasvss script 125
installing
sample databases 29
insthasv script 125
insufficient disk space
disk resize 319
insufficient permission 388
insufficient resource errors (Level 17) 389
interfaces file 14
internal error, nonfatal 389390
international language support. See character sets;
languages
internationalization
a sample system 329
advantages 328
definition 327
directory structure for character sets 365

System Administration Guide: Volume 1

files 365
ISO 8859-1
similarities to CP 1252 333
iso_1 character set 109
isql utility command
character set conversion and 379, 380
number of user connections and 219
passwords and 304
status and informational messages 387
system administration and 5

J
Japanese character sets 109
sjis (Shift-JIS) 109
support 333
See also languages, alternate
Java configuration parameters 254
job scheduler interval configuration parameter 156
job scheduler tasks configuration parameter 156
js job output width configuration parameter 157

K
kanji. See Japanese character sets
kernel
error messages 385, 390
kernel mode configuration parameter 157
kernel resource memory configuration parameter
158
keys, table
on system tables 9
kill command 394398
kill command, changes 397
kill statusonly parameter 397
known problems 403
Korean
character set support 333

L
@@langid global variable
language defaults 111

368

413

Index

changing users 359

us_english 111
@@language global variable 368
language groups 332, 333
languages
on server 332
supported by a character set 332
languages, alternate 365
See also character sets; charset.loc file; Japanese
character sets
date formats in unsupported 363
localization files 349368
supported languages 328
Latin alphabet 333
lazy garbage collection 399
LDAP
access restrictions 16
defined 16
multiple directory services 17
versus the interfaces file 17
levels, severity.
See severity levels, error
libtcl.cfg file 15
license information configuration parameter 159
license information, configuration parameter 399
license Security and Directory Services required 145
linkage, page
See also pages, data
listing
database options 322
load, database
number of large i/o buffers configuration parameter
115, 169, 196, 203
local and remote servers. See remote servers
local option, sp_addserver 296
local servers 296
locales directory 350
locales.dat file 366
localization 328
See also languages, alternate
files for 366367
lock address spinlock ratio configuration parameter 160
lock hash buckets 163
lock hash table
configuring size of 160
lock hashtable size configuration parameter 160

414

lock promotion thresholds

setting with sp_configure 225246
lock scheme
default 161
lock scheme configuration parameter 161
lock shared memory configuration parameter 162
lock spinlock ratio configuration parameter 162, 163
lock table spinlock ratio configuration parameter 163
lock timeout pipe active configuration parameter 164
lock timeouts
configuring server-wide 165
lock wait period configuration parameter 165
locking 54
by dbcc commands 362
locking logins 54
locking scheme
server-wide default 161
locks
quantity of 204
log audit logon failure configuration parameter 165
log audit logon success configuration parameter 166
log file. See error logs
log on option
create database 290
logging
login failures 165
successful logins 166
Windows NT event log in 144
logical
page sizes 33
login password encryption 145
logins
database object owner 4
dbo user name 2, 4
locking 54
logsegment log storage 287

M
Macintosh character set 109, 377
mail session, starting 262
management, space. See space allocation; storage
management
mapping
device name to physical name 308

Adaptive Server Enterprise

Index
remote users 300304
master database 7, 2324, 40
See also disk mirroring;system tables
backing up 40, 53
changing option settings 321
creating 287
keys for system tables in 9
sysdevices table 315
master device 22, 309, 315
See also database devices
removing from default space pool 316, 317
sp_diskdefault and 317
max async i/os per engine configuration parameter
166
max async i/os per server configuration parameter
167
max cis remote connections configuration parameter
169
max concurrently recovered db configuration
parameter 169, 196
max memory configuration parameter 170
max native threads per engine configuration
parameter 171
max nesting level configuration parameter 171
max network packet size configuration parameter
171
max number network listeners configuration
parameter 174
max online engines configuration parameter 175
max online Q engines 175
max parallel degree configuration parameter 176
max pci slots configuration parameter 177
max repartition degree configuration parameter 179
max resource granularity configuration parameter
179
max roles enabled per user configuration parameter
189
max scan parallel degree configuration parameter
179
max SQL text monitored configuration parameter
181
max transfer history, configuration parameter 181
@@max_connections global variable 218
@@maxcharlen global variable 366
maximum buffers per lava operator configuration
parameter 182

System Administration Guide: Volume 1

maximum dump conditions configuration parameter

182
mci memory size configuration parameter

230
memory
See also space allocation
audit records 90
freeing from XP Server 142
number of open databases and 207
memory alignment boundary configuration parameter
184
memory per worker process configuration parameter
185
messages
error 13, 384392
fatal error 13
language setting for 328
start-up 13
system 384392
user-defined 386
messaging memory configuration parameter 185
metadata caches configuration parameters 73223
metrics elap max configuration parameter 186
metrics exec max configuration parameter 186
metrics lio max configuration parameter 186
metrics pio max configuration parameter 187
Microsoft character set 109
min pages for parallel scan 187
minimum
size, disk resize 318
minimum pages for a parallel scan configuration
parameter 187
minimum password length configuration parameter
188
miscellaneous user error 388
mistakes, user See errors; severity levels, error
mnc_full_index_filter configuration parameter 189
model database 52
model database 25
creating 287
keys for system tables in 9
size 109, 310
money
local formats 367
monitoring
spt_monitor table 11
SQL text 181

415

Index

Windows NT Performance Monitor 256

monitoring tables
configuration options 73
month values
alternate language 363
MSDTC 122
msg confidentiality reqd configuration parameter 189
msg integrity reqd configuration parameter 190
multibyte character sets 361
changing to 362
default character set id configuration parameter 109
incompatible 377
multilingual character set 109
multiple directory services
LDAP 17

N
name of device 308
sysdevices listing 290
names
column, in commands 389
mapping remote user 301
remote server 295
remote user 301
server 296
system extended stored procedures 12
system procedures 10
naming
servers 296
Navigating
to objects 49
@@ncharsize global variable 366
nested trigger configuration parameter (now called allow
nested triggers) 85
net password encryption option 298
net password encryption reqd configuration parameter
190
networks
connections 14
directory services 15
interfaces files 14
software 36
nonstop recovery 286
number (quantity of)

416

database devices 196

engines 175
locks 204
open databases on Server 207
open objects 210
seconds for acquiring locks 165
user connections (@@max_connections) 218
number of alarms configuration parameter 191
number of aux scan descriptors configuration
parameter 192
number of backup connections configuration
parameter 194
number of ccbs configuration parameter 195
number of devices configuration parameter 196
number of disk tasks configuration parameter 198
number of dtx participants configuration parameter
198
number of histogram steps configuration parameter
201
number of index trips configuration parameter 201
number of large i/o buffers configuration parameter
203
number of locks configuration parameter 203
number of mailboxes configuration parameter 204
number of messages configuration parameter 205
number of network tasks configuration parameter
205
number of oam trips configuration parameter 206
number of open databases configuration parameter
206
number of open indexes configuration parameter 208
number of open objects configuration parameter 210
number of pre-allocated extents configuration
parameter 213
number of Q engines at startup 214
number of remote connections configuration
parameter 215
number of remote logins configuration parameter
199, 215
number of remote sites configuration parameter 216
number of sort buffers configuration parameter 216
number of threads for memory dumps, determining
199
number of user connections configuration parameter
77, 217219

Adaptive Server Enterprise

Index
number of worker processes configuration parameter

220

225

numbers
error message 384
sort order 113
status bit (sysdevices)

page lock promotion LWM configuration parameter

226, 245
page lock promotion PCT configuration parameter

315

O
o/s file descriptors configuration parameter 221
object lockwait timing configuration parameter 221

object owners. See database object owners

objects
icons 49
navigating to 49
See database objects
open index hash spinlock ratio configuration
parameter 221
open index spinlock ratio configuration parameter
222
open object spinlock ratio configuration parameter
223
operating system commands
executing 12
operator role 3
optimization goals and configuration parameters 224
optimization timeout limit configuration parameter
224, 225
options
database 321323
remote logins 304
remote servers 297
server 297
overflow errors
server stack 260
overflow stack (stack guard size configuration
parameter) 258
overriding user permissions 53

P
packets, network
size, configuring

page lock promotion HWM configuration parameter

172173

System Administration Guide: Volume 1

227
pages, data 308
dirty 240
partition groups configuration parameter 229
partition spinlock ratio configuration parameter 229
partitions
disk 309
partitions, suspect, fixing tables 363
partitions, suspect, handling 362
passwords
encryption over network 298
remote users 298, 304
per object statistics active configuration parameter
230
per opject statistics active configuration parameter
230
percent sign (%)
error message placeholder 384
performance
audit queue size 90
default fill factor percent effect on 111
disk mirroring and 286
ESPs and XP Server priority 141
space allocation and 287
speed and 287
performance monitoring option configuration
parameter 232
permission cache entries configuration parameter
233
permissions
database object owners 5
database owners 4
default 25
denying 388
disk init 314
group versus user 53
insufficient (Level 14) 388
master database 24
model database 25
object 5
overriding 53

417

Index

remote users 304

stored procedures 304
tempdb database 27
physical resources, managing.
See storage management
pinyin
gbpinyin, gbpinyinocs sort orders 345
size of unilib cache configuration parameter 345
size of unilib cache configuration parameter, 345
transliterates Mandarin pronunciation 345
use of accents 345
placeholders
error message percent sign (%) 384
plan text pipe active configuration parameter 233
plan text pipe max messages configuration parameter
234
preventing garbage collection
accumulating empty pages 400
print deadlock information configuration parameter 234
print recovery information configuration parameter 235
priority
XP Server 141
procedure cache 236, 391
procedure calls.
See remote procedure calls
procedures. See stored procedures; system procedures
process ID, status of 397
process wait events configuration parameter 237
processes (server tasks) 394, 398
killing 394398
prod-consumer overlap factor 237
production server 34
public membership 53
pubs2 database
administering 29
image information in 29
pubs3 database
administering 29

Q
queries
conversion errors, preventing
question marks (??)
for suspect characters 377

418

377

quorum hearbeat interval configuration parameter

237
quorum heartbeat retries configuration parameter

238
quoted identifier enhancement configuration parameter

238

R
read committed with lock configuration parameter

239
read only database option

359
reads
physical 287
rebooting the server
See restarts, server
record keeping 4345
configuration 44
contacts 44
maintenance 44
system 45
recovery
configuration parameters for 239241
loading databases 355
master database 40, 308
nonstop 286
planning backups for 25
after reconfiguration 355
sort order changes and 355
space allocation and 286
recovery interval in minutes configuration parameter
239241
long-running transactions and 240
remote logins
adding 300302
configuration parameters for 86, 305
dropping 299, 300
options for 304
timing out 297
trusted or untrusted mode 302
remote procedure calls 293306
configuration parameters for 305
remote server pre-read packets configuration
parameter 242
remote servers 294299

Adaptive Server Enterprise

Index
adding 294299
dropping 299
information on 299
names of 295
options for 297
reorg command
running manually 401
reorg reclaim_space command 400
reporting
errors 387, 389, 392
reports
See also information (server)
reset configuration.
See configuration parameters;reconfigure
command
resource limits
configuring 87
response time 271
restarts, server
after reconfiguration 359
reindexing after 359
from same directory 386
system tables and 359
temporary tables and 27
restricted decrypt permission configuration
parameter 242
retaindays option
dump database 270
dump transaction 270
return status
system procedures 11
roles, system
operator 3
system administrator 2
system security officer 3
rolling back processes
recovery interval and 239
server stack capacity and 261
roman8 character set 109
row lock promotion HWM configuration parameter
244
row lock promotion LWM configuration parameter
245
row lock promotion PCT configuration parameter
246
row lock promotion thresholds

System Administration Guide: Volume 1

setting with sp_configure 244, 246

rows, table
sysindexes 291
RPCs. See remote procedure calls
rtm thread idle wait period configuration parameter
246
rules
See also database objects
runnable process search count configuration
parameter 247
running out of space. See space
running reorg command manually 401
russian
character set support 333

S
savepoints
error (Level 13) 388
scan descriptors 192194
scripts 332
secure default login configuration parameter 249
Security and Directory Services, requirement of 145
segmap column, sysusages table
procedures that change 290
segments 291
See also database devices; space allocation
creating 287
default 287
logsegment 287
syssegments table 291
system segment 287
select for update configuration parameter 250
select into/bulkcopy/pllsort database option
model database and 25
select on syscomments.text column configuration
parameter 250
send doneinproc tokens 250
server aliases 296
server information options. See information (server)
server.loc file 367
server_name.cfg, default name of configuration file 62
servers
See also processes (server tasks); remote servers
connecting 14

419

Index

error message severity levels 386392

error messages 385
fatal errors and 390392
installing 35, 287
interfaces files 14
local 296
monitoring performance 76
names of 296
nonfatal internal errors 389390
passwords on 298, 304
remote 295301
scheduler 271
shutting down 401
single-user mode 89
sort order consistency among 355
stopping 401
syntax errors 388
user connections to 219
values for configuration parameters 61
session migration timeout configuration parameter 251
set switches
showing current settings 324
7-bit ASCII character data, character set conversion for
371
severity levels, error 381, 386
Backup Server 393
levels 10-18 (user errors) 387
levels 19-24 (fatal) 390
shared memory starting address configuration parameter
252
shutdown command 401403
shutting down servers 401
simplified Chinese
character set support 333
single-user mode 89, 359
size
See also space
dbcc fix_text transaction 361
error log 13
model database 109, 310
new database 25
tempdb database 26
size of auto identity column configuration parameter 253
size of global fixed heap configuration parameter 253
size of process object fixed heap configuration parameter
254

420

size of shared class heap configuration parameter

254
size of unilib cache configuration parameter

255
sjis (Shift-JIS) character set. See Japanese character sets
sort order
changing 355359
consistency among servers 355
default sortorder id 113
default XML sortorder 113
definition files 365
installing new 365
numbers 113
rebuilding indexes after changing 360
sp_addlanguage system procedure 363
sp_addremotelogin system procedure 300302
sp_addsegment system procedure
sysusages and 290
sp_addserver system procedure 295297
sp_adduser system procedure 25
sp_configure system procedure 66
See also individual configuration parameter names
remote logins and 305
sp_countmetadata system procedure 207, 209, 210,
212
sp_dboption system procedure 321323
sp_deviceattr system procedure 284, 312
sp_diskdefault system procedure 284, 316317
sp_dropdevice system procedure 316
sp_dropremotelogin system procedure 300
sp_dropsegment system procedure
sysusages and 290
sp_dropserver system procedure 299
sp_extendsegment system procedure
sysusages and 290
sp_helpconfig system procedure 207, 208, 210
sp_helpdb system procedure 11
database option information 323
sp_helpdevice system procedure 11, 314
sp_helpindex system procedure 11
sp_helpjoins system procedure 9
sp_helpkey system procedure 9
sp_helpremotelogin system procedure 305
sp_helpserver system procedure 299
sp_helptext system procedure 11
sp_indsuspect system procedure 359, 360
sp_modifylogin system procedure 359

Adaptive Server Enterprise

Index
sp_monitorconfig system procedure
configuring number of open databases and 207
configuring number of open indexes and 209
configuring number of open objects and 211,

212
sp_remoteoption system procedure 304305
sp_serveroption system procedure 297
sp_showpsexe system command, housekeeper output

398
sp_who, housekeeper output

398
space
See also size; space allocation
running out of 389
space allocation
See also database devices; segments; storage
management
commands summary 284
recovery/performance and 286
sysusages table 290
space reclamation
enable housekeeper GC configuration parameter
126
Spanish
character set support 333
#spdevtab temporary table 11
speed (server)
system performance and 287
#spindtab temporary table 11
spinlocks
lock hash table 163
splitting
tables across two disks 287
SPR files 403
sproc optimize timeout limit configuration parameter
255
spt_committab table 11
spt_monitor table 11
spt_values table 11
SQL batch capture configuration parameter 256
sql server clock tick length configuration parameter
257
sql text pipe active configuration parameter 258
sql text pipe max messages configuration parameter
258
.srt files 365
srvname column, sysservers table 297

System Administration Guide: Volume 1

srvnetname column, sysservers table 297

stack guard size configuration parameter 258
stack size configuration parameter 261
standalone utilities and character sets 379
start mail session configuration parameter 262
start xp server during reboot configuration parameter
263
startup delay configuration parameter 263
statement pipe active configuration parameter 264
statement pipe max messages configuration parameter
264
statement statistic active configuration parameter
265
statement statistics active configuration parameter
265
static configuration parameters 62
statistics
flushing with housekeeper task 149
housekeeper flushing and 150
status
information messages (Level 10) 387
status bits in sysdevices 315
stopping
Backup Server 402
Servers 402
space allocation
storage management 283
commands summary 284
database device initialization 307315
default database devices 316317
defaults at installation 287
issues 3739, 286
See also space 283
system tables and 288291
stored procedure
triggers. See triggers
stored procedures
See also database objects; system procedures
creating 12
permissions on 304
procedure cache and 236
remote user access to 304
system tables changes and 12
streamlined dynamic SQL configuration parameter
265

421

Index

strict dtm enforcement configuration parameter 265, 266

structure
internationalization files directory 365
localization files directory 367
suffix names, temporary table 27
sun character set 109
suspect partitions, in cross-platform dump and load 363
suspend audit when device full configuration parameter
266
switches, viewing current set settings 324
syb_sendmsg port number configuration parameter 267
Sybase Central, using for system administration tasks 6
sybsecurity database 28
sybsystemdb database 28
sybsystemprocs database 10, 12, 26
See also databases
syntax
disk resize 318
errors in 388
sysconfigures table 7980
syscurconfigs table 7980
sysdevices table 289, 314
disk init and 290
sp_dropdevice and 316
sp_helpdevice and 314
status bits 315
sysindexes table 291, 359
syslogs table
modification of 9
sysmessages table 382, 384
sysobjects table 359
sysremotelogins table 302
syssegments table 291
sysservers table 293, 294, 295, 299
sp_helpserver and 299
srvname column 297
srvnetname column 297
system administration tasks
accomplishing with Sybase Central 6
System Administrator
error responsibilities of 386, 389392
resolving system problems 386, 389
tasks for beginners 3345
system administrator 15
system catalogs. See system tables
system databases 2128

422

system extended stored procedures 12

system messages. See error messages 381
system problems
See also errors
Server responses to 381392
severity levels 10 to 18 387390
severity levels 19 to 24 390392
System Problem Reports (SPRs) 403
system procedure
tables 11
system procedures 1012
See also information (server); stored procedures;
individual procedure names
creating 12
for managing remote servers 294299
on temporary tables 27
using 10
system security officer 3
system segment 287
system tables 79
See also individual table names
changes dangerous to 12
corruption 392
create database and 7, 290
creation of 7
dbcc reindex and 361
keys for 9
querying 8, 12
reindexing and 361
server restarts and 359
storage management relationships 288291
stored procedures and 8, 12
updating 9, 12
for user databases 25
systemwide password expiration configuration
parameter 268
sysusages table 290
corruption 392

T
Table editor 54
table owners. See database object owners
tables
See also database objects; system tables

Adaptive Server Enterprise

Index
dbcc checktable and 360
integrity damage to 391
object allocation maps of 206
read-only 359
splitting across two disks 287
suspect, fixing 363
system procedure 11
temporary 26
with suspect partitions, fixing 363
without indexes 360
tape retention in days configuration parameter 269
tcp no delay configuration parameter 270
tempdb database 2627
See also databases
creating 287
size of 26
temporary tables 26
terminals
character set conversion for 379
installing new definitions 365
test servers 3435
text datatype
changing character sets and 361
multibyte character sets and 361
text prefetch size configuration parameter 271
text values, dbcc fix_text upgrade of 361
Thai
character set support 333
time
for acquiring locks 165
time slice configuration parameter 271
time values
display format 367
timeouts option, sp_serveroption 297
total data cache size configuration parameter 272
traditional Chinese
character set support 333
transaction logs
alter database and 290
create database and 290
device placement 286, 290
purging 361
trunc log on chkpt option and 240
transactions
error within 388
long-running 240

System Administration Guide: Volume 1

recovery and 240

two-phase commit 28
transfer utility memory size configuration parameter

274
transferring ownership.
See database objects, ownership
translation.
See character sets
triggers
See also database objects; stored procedures
nested 85
trunc log on chkpt database option
recovery interval in minutes and 240
trusted mode
remote logins and 304
tuning
monitoring performance 76
turkish
character set support 333
two-phase commit
transactions 28
txn to pss ratio configuration parameter 274

U
unichar datatype 334
Unicode 332, 334338
character sets 333
unichar datatype 334
univarchar datatype 334
UTF-16 334
unified login required 276
univarchar datatype 334
UNIX platforms, raw disk partition 309
untrusted mode, remote logins and 304
updating
See also changing
allow updates to system tables configuration
parameter and 12
text after character set change 361
upgrade version configuration parameter 276
us_english language 111
usage
disk resize 318
use security services configuration parameter 277

423

Index

user connections
memory allocated per 217219
user databases
master database control of 23
system tables for 25
user-defined messages 386
user errors 387, 387390
user IDs
number 1, Database Owner 12
user log cache size configuration parameter 277
user log cache spinlock ratio configuration parameter
279
user mistakes. See errors; severity levels, error
user objects. See database objects
users
creating 53
deleting 54
errors by 387, 387390
guest 52
number of user connections and 219
remote 300304
single-user mode 89
users, object.See database object owners
UTF-16 334
utility commands
See also Utility Programs manual
character sets and 379
utility, housekeeper, aggressive 399

W
wait event timing configuration parameter 279
wash, housekeeper task 149
Western Europe
character set support 333
window of vulnerability 89
with nowait option, shutdown 402, 403
With override, database option 52
workload manager size configuration parameter
write operations
physical 287

279

X
X/Open XA 122
xact 280
.xlt files 365
XP Server
freeing memory from 142
priority 141
xp_cmdshell context configuration parameter 281
xp_cmdshell system extended stored procedure 12

V
variables in error messages 385
verification, user-access 298, 302
Vietnamese
character set support 333
views
See also database objects
virtual
address 314
page numbers 311
vstart option
disk init 314

424

Adaptive Server Enterprise