WebMethods Deployer User's Guide 7.0 - Software AG Documentation
WebMethods Deployer User's Guide 7.0 - Software AG Documentation
Page
webMethods Deployer
User’s Guide
VERSION 7.0
APRIL 2007
webMethods, Inc.
South Tower
3877 Fairfax Ridge Road
Fairfax, VA 22030
USA
703.460.2500
http://www.webmethods.com
Copyright
& Docu‐
ment ID
B2B Integration Server, Business Process Integration, Cerebra, Get There Faster, Glue, Glueprints, Glueware, Guided SOA Governance, Infravio X‐Broker,
Infravio X‐Registry, Infravio, My webMethods Server, My webMethods, Process Improvement Lifecycle, Process Improvement Lifecycle Methodology,
webMethods Access, webMethods Administrator, webMethods Broker, webMethods Central Configuration, webMethods Dashboard, webMethods Designer,
webMethods Developer, webMethods Fabric, webMethods Glue, webMethods Infrastructure Data Collector, webMethods Infravio X‐Broker, webMethods
Infravio X‐Registry, webMethods Installer, webMethods Integration Server, webMethods logo, webMethods Mainframe, webMethods Manager, webMethods
Modeler, webMethods Monitor, webMethods Optimize for Infrastructure, webMethods Optimize for Process, webMethods Optimize, webMethods Portal,
webMethods Process Engine, webMethods Servicenet, webMethods SOA Governance, webMethods SOA Management, webMethods Task Engine,
webMethods Trading Networks, webMethods Workflow, and webMethods are either registered trademarks or trademarks of webMethods, Inc. in the United
States and/or other countries.
Acrobat and Adobe are registered trademarks, and Reader is a trademark of Adobe Systems Incorporated. Amdocs is a registered trademark, and ClarifyCRM
is a trademark of Amdocs. Ariba is a registered trademark of Ariba, Inc. BEA, BEA WebLogic Server, Jolt, and Tuxedo are registered trademarks, and BEA
WebLogic Platform is a trademark of BEA Systems, Inc. Action Request System, BMC Software, PATROL, and Remedy are registered trademarks of BMC
Software, Inc. BroadVision is a registered trademark of BroadVision, Inc. ChemeStandards and CIDX are trademarks of Chemical Industry Data Exchange.
Unicenter is a registered trademark of Computer Associates International, Inc. PopChart is a registered trademark of CORDA Technologies, Inc. Kenan and
Arbor are registered trademarks of CSG Systems, Inc. Data Connection and SNAP‐IX are registered trademarks of Data Connection Corporation. DataDirect,
DataDirect Connect, and SequeLink are registered trademarks of DataDirect Technologies. D&B and D‐U‐N‐S are registered trademarks of Dun & Bradstreet
Corporation. Entrust is a registered trademark of Entrust, Inc. papiNet is a registered trademark of the European Union and the United States. Financial
Information eXchange, F.I.X, and F.I.X Protocol are trademarks of FIX Protocol Ltd. UCCnet and eBusinessReady are registered trademarks, and 1SYNC and
Transora are trademarks of GS1 US. Hewlett‐Packard, HP, HP‐UX, OpenView, PA‐RISC, and SNAplus2 are trademarks of Hewlett‐Packard Company. i2 is a
registered trademark of i2 Technologies, Inc. AIX, AS/400, CICS, DB2, Domino, IBM, Informix, Infoprint, Lotus, Lotus Notes, MQSeries, OS/390, OS/400,
RACF, RS/6000, SQL/400, S/390, System/390, VTAM, z/OS, and WebSphere are registered trademarks; and Communications System for Windows NT, DB2
Universal Database, IMS, MVS, and SQL/DS are trademarks of IBM Corporation. InnoDB is a trademark of Innobase Oy. Itanium is a registered trademark of
Intel Corporation. JBoss is a registered trademark, and JBoss Group is a trademark of Jboss, Inc. Linux is a registered trademark of Linus Torvalds. W3C is a
registered trademark, and X Window System is a trademark of the Massachusetts Institute of Technology. MetaSolv is a registered trademark of Metasolv
Software, Inc. ActiveX, Microsoft, Outlook, Visual Basic, Windows, and Windows NT are registered trademarks; and Windows Server is a trademark of
Microsoft Corporation. Six Sigma is a registered trademark of Motorola, Inc. Firefox is a registered trademark, and Mozilla is a trademark of the Mozilla
Foundation. MySQL is a registered trademark of MySQL AB. nCipher is a trademark of nCipher Corporation Ltd. Teradata is a registered trademark of NCR
International, Inc. Netscape is a registered trademark of Netscape Communications Corporation. ServletExec is a registered trademark, and New Atlanta is a
trademark of New Atlanta Communications, LLC. SUSE is a registered trademark of Novell, Inc. Appia is a registered trademark and Javelin Technologies is
a trademark of NYFIX, Inc. CORBA is a registered trademark of Object Management Group, Inc. JD Edwards, OneWorld, Oracle, PeopleSoft, Siebel, and
Vantive are registered trademarks, and PeopleSoft Pure Internet Architecture and WorldSoftware are trademarks of Oracle Corporation. Infranet and Portal
are trademarks of Portal Software, Inc. Red Hat is a registered trademark of Red Hat, Inc. PIP and RosettaNet are trademarks of RosettaNet, a non‐profit
organization. SAP and R/3 are registered trademarks of SAP AG. SWIFT and SWIFTNet are registered trademarks of Society for Worldwide Interbank
Financial Telecommunication SCRL. SPARC and SPARCStation are registered trademarks of SPARC International, Inc. SSA is a registered trademark, and
Baan and SSA Global are trademarks of SSA Global Technologies, Inc. EJB, Enterprise JavaBeans, Java, JavaServer, JDBC, JSP, J2EE, Solaris, Sun, and Sun
Microsystems are registered trademarks; and Java Naming and Directory Interface, SOAP with Attachments API for Java, JavaServer Pages, and SunSoft are
trademarks of Sun Microsystems, Inc. Sybase is a registered trademark of Sybase, Inc. VERITAS is a registered trademark, and VERITAS Cluster Server is a
trademark of Symantec Corporation. UNIX is a registered trademark of The Open Group. Unicode is a trademark of Unicode, Inc. VeriSign is a registered
trademark of Verisign, Inc.
All other marks are the property of their respective owners.
Copyright © 2007 by webMethods, Inc. All rights reserved, including the right of reproduction in whole or in part in any form.
Document ID: DEP-UG-70-20070420
Contents
Contents
Chapter 1. Concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Deployment Steps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Create the Project . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Define the Project . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Build the Project . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Map the Project . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Deploy the Project . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Checkpoint and Roll Back . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
This guide explains how to use webMethods Deployer. The Deployer is a tool you use to
deploy items that reside on source webMethods servers to target webMethods servers.
You can deploy items from the products listed below.
Product Versions
Blaze Advisor 6.1.7
webMethods Broker 6.5.x
webMethods Designer 7.0 SP2
webMethods Integration Server 6.5.x
My webMethods Server 7.0 SP2
webMethods Optimize 7.0 SP2
webMethods Trading Networks 6.5.x
Document Conventions
Convention Description
Bold Identifies elements on a screen.
Italic Identifies variable information that you must supply or change
based on your specific situation or environment. Identifies
terms the first time they are defined in text. Also identifies
service input and output variables.
Narrow font Identifies storage locations for services on the webMethods
Integration Server using the convention folder.subfolder:service.
Typewriter Identifies characters and values that you must type exactly or
font messages that the system displays on the console.
UPPERCASE Identifies keyboard keys. Keys that you must press
simultaneously are joined with the “+” symbol.
\ Directory paths use the “\” directory delimiter unless the
subject is UNIX‐specific.
[ ] Optional keywords or values are enclosed in [ ]. Do not type
the [ ] symbols in your own code.
Additional Information
The webMethods Advantage Web site at http://advantage.webmethods.com provides you
with important sources of information about the webMethods Integration Platform:
Troubleshooting Information. webMethods provides troubleshooting information for
many webMethods components in the webMethods Knowledge Base.
Documentation Feedback. To provide documentation feedback to webMethods, go to the
Documentation Feedback Form on the webMethods Bookshelf.
Additional Documentation. All webMethods documentation is available on the
webMethods Bookshelf.
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Deployment Steps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Overview
webMethods Deployer is a tool you use to deploy items that reside on source
webMethods servers to target webMethods servers. For example, you might want to
deploy items you have developed on servers in a development environment to servers in
a test or production environment.
The sections below list the items you can deploy and the source and target servers for
those items.
Note: A deployed process model
does not contain the design‐time
source code for the project, just
the run‐time elements that are
necessary for process execution.
Important! Do not deploy packages
that are webMethods products.
For example, do not deploy the
WmMonitor package, which is
webMethods Monitor. Instead,
install such packages using the
webMethods Installer.
Deployment Steps
The deployment process involves these steps:
1 Create a deployment project by assigning the project a name, setting the project’s
properties, and authorizing users to work on the project.
2 Define the project by identifying its contents.
3 Build the project.
4 Map the contents of the project to target servers.
5 Deploy the project to the target servers.
Note: If you want to deploy process models from
multiple ProcessModel servers, you must define a
deployment set for each ProcessModel server.
The items you include in a deployment set depend on the target servers to which you
want to deploy the items. For example, if you want to deploy a set of Integration Server
items to the same target servers, you can define a single deployment set that identifies
those items.
project
deployment set target servers
If you want to deploy some Integration Server items to one set of target servers and other
Integration Server items to a different set of target servers, you would have to define two
different deployment sets.
project
deployment sets target servers
As you define a deployment set, the Deployer determines when items that are in a
deployment set require other items that are not in the deployment set. The Deployer
identifies these missing items as unresolved dependencies. For example, if you add a trigger
to an IS & TN deployment set, but do not add the service that is invoked by the trigger,
the Deployer identifies the missing service as an unresolved dependency. The Deployer
gives you several options for resolving unresolved dependencies.
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
Overview
Some servers require a special component to communicate with the Deployer; you must
install the appropriate communications component on each server. For most servers, you
must configure communication between the Deployer and the server.
The Deployer and Integration Server Administrator use the same log on user name and
password. If you just installed the Deployer with a new Integration Server, the defaults
are user name Administrator and password manage.
To log out of the Deployer, click Log Off on the top right corner of any Deployer page.
Important! The Deployer cannot use SSL to connect to Brokers.
In the Deployer, go to the ServersBroker page. For every source and target Broker, click
Configure Broker Server. In the Configure Broker Server area, complete these fields:
Box Entry
Name Name to assign to the Broker.
Host Host name or IP address of the Broker.
Port Port for the Broker.
Client Group Client group to use to access the Broker. For target Brokers, type admin.
Broker Name Name of the Broker.
Click Configure. If you want to test the connection, click .
1 In the Deployer, go to the ServersProcessModel page. For every source and target
ProcessModel server, click Configure ProcessModel Server. In the Configure ProcessModel
Server area, complete these fields:
Box Entry
Name Name to assign to the server.
Host Host name or IP address of the server.
Port Port for the server.
User User name for a user account with Administrator authority that
Deployer can use to access the server.
Box Entry
Password Password that is associated with the user name.
Use SSL Whether to encrypt the communication between the server and the
Deployer. If you want to encrypt the communication, choose the Yes
option.
You can only use SSL if the server is configured to use SSL. When
defining the HTTPS port that the server uses for SSL, configure the
port so that it does not request client certificates. For instructions on
defining the HTTPS port, see the webMethods Integration Server
Administrator’s Guide.
Click Configure. If you want to test the connection, click .
2 In Designer, a logical‐to‐physical server mapping is defined for each process model.
For deployment purposes, you must duplicate the mapping for each process model
you want to deploy on the model’s source and target ProcessModel servers. In the
Integration Server Administrator for each of the servers, do the following:
a Define the physical servers in the mapping as remote servers. For instructions, see
the webMethods Integration Server Administrator’s Guide.
b Go to the PackagesManagement page and click for the WmDesigner package.
c Click Add Logical Server and complete these fields:
Box Entry
Name Name of a logical server in the mapping for the process model.
Type Integration Server.
Physical Physical server to which the logical server is mapped.
Server
1 Open the Integration Server Administrator for the Integration Server that hosts the
Deployer and define the following as remote servers:
All source Integration Servers
All target Integration Servers
The Integration Server that hosts the Deployer, if you will be using it as a source
or target server (that is, define the Integration Server as a remote server to itself)
For instructions on defining remote servers, see the webMethods Integration Server
Administrator’s Guide.
2 Install the WmDeployerResource package on each Integration Server. In the
Deployer, go to the ServersIS & TN page; the page lists all Integration Servers you
defined as remote servers. In the Install column, select the check box next to each
Integration Server, then click Install.
In the Deployer, go to the ServersMWS page. For every source and target My
webMethods Server, click Configure MWS Server. In the Configure MWS Server area, complete
these fields:
Box Entry
Name Name to assign to the My webMethods Server.
Host Host name or IP address of the My webMethods Server.
Port Port for the My webMethods Server.
User User name for a user account with Administrator authority that
Deployer can use to access the My webMethods Server.
Password Password that is associated with the user name.
Include security When resolving dependencies for an item, whether to include
dependencies security realms that contain that item, as well as user/group/role
references in the itemʹs security ACLs.
Root folder aliases My webMethods Server aliases to use as root folders when
selecting pages to deploy. Separate the folders using commas.
Box Entry
Maximum folder Maximum number of items to display within My webMethods
object count Server folders when you are defining and choosing items to
include in an MWS deployment set.
Enable additional Whether to log debug information about items you select to source
MWS logging My webMethods Server logs, and about items the Deployer
deploys to target My webMethods Server logs.
Maximum folder Maximum number of My webMethods Server folder levels to
depth display when you are defining and choosing items to include in an
MWS deployment set.
Use SSL Whether to encrypt the communication between the My
webMethods Server and the Deployer.
You can only use SSL if the My webMethods Server is configured
to use SSL. When using the My webMethods Server Configurator
to define the HTTPS port that the server uses for SSL, configure the
port so that it does not request client certificates. For instructions,
see the My webMethods Server Administrator’s Guide.
Click Configure. If you want to test the connection, click .
In the Deployer, go to the ServersOptimize page. For every source and target Optimize
server, click Configure Optimize Server. In the Configure Optimize Server area, complete these
fields:
Box Entry
Name Name to assign to the server.
Host Host name or IP address of the server.
Port Port for the server.
Click Configure. If you want to test the connection, click .
Create a Project . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Option Description
Automatic Tells the Deployer to automatically check dependencies regularly as you
modify the project and progress through the different phases of
deployment.
Reduced Tells the Deployer to automatically check dependencies when you create
a project build and when you deploy. You can trigger additional
dependency checking at different steps yourself.
Manual You trigger dependency checking at different steps yourself.
Option Entry
Checkpoint To have the Deployer automatically create a checkpoint for a
Creation deployment candidate before it is deployed, click Automatic. If you only
want to generate checkpoints when you choose to (see “Generate a
Checkpoint” on page 84), click Manual.
Rollback on To have the Deployer automatically roll back a deployment candidate
Error after a failed deployment, click Automatic. If you only want to roll back
the project when you choose to (see “Roll Back the Target Servers” on
page 84), click Manual.
Cluster If you want to deploy to each Integration Server 6.5 in a cluster yourself,
Support click Manual.
If you want to deploy to a single Integration Server 6.5 in a cluster and
have that Integration Server automatically propagate the deployment to
the other Integration Servers in the cluster, click Automatic. For
instructions on setting up this feature, see Chapter 12, “Deploying to
Clustered 6.5 Integration Servers”.
Important! If you choose All, the Deployer suspends execution and
document retrieval for ALL triggers on the target Integration
Servers, not just for the triggers that you include in the project.
Not suspend triggers. None
Choose individual triggers to suspend from the Selected
packages in the project. You can use this option under
these conditions:
The source and target Integration Servers are
release 6.0.1 and are equipped with SP1 and Fix 144,
and the triggers were developed on a 6.0.1
Developer equipped with SP1_Fix9
The source and target Integration Servers are
release 6.1 and are equipped with Fix 22
The source and target Integration Servers are
release 6.5
You choose the triggers when you set package
properties. For instructions, see “Set the Package
Properties” on page 49.
Ports Whether to disable ports on the target Integration Servers that
match ports you are trying to deploy.
Note: Tasks that are already running at deployment time are not
affected by deployment.
2 Under the Overwrite Existing area, indicate how the Deployer should proceed when it
finds that items you are trying to deploy already exist on target Integration Servers.
For this
option... Indicate whether the Deployer should...
TN Rules Overwrite existing Trading Networks processing rules or append
the processing rules you are trying to deploy to the ones that
already exist.
ACL Maps Deploy the mapping of ACLs to services for any services you
choose to deploy. You would choose to deploy ACL maps if you
want to assign the same ACLs to the deployed services on the
target Integration Server that you assigned to the source services
on the source Integration Servers.
Other Non- Overwrite existing items. This option applies to all items except
Package Items the following:
Trading Networks processing rules (see TN Rules, above).
Integration Server ACL maps (see ACL Maps, above).
Integration Server packages. You specify the overwrite option
for Integration Server packages on a package‐by‐package
basis, as described in “Set the Package Properties” on page 49.
Note: Before you deploy a project, you can find out which items the Deployer will
overwrite by generating the simulation report.
Note: If you choose to activate ports, and one of the ports you deploy uses the same
port number as an existing port on a target Integration Server, the Deployer will
display a message to that effect and will not activate the port.
4 Click Save.
Create a Project
You can create a project by creating a new, blank project or by copying an existing project
and modifying it.
To create a project:
1 On the Projects page, create a project using one of these methods:
Create... Steps
New 1 Click Create Project.
2 In the Name box, type the name to use for the new project. The
name can be up to 32 characters long and can include any
characters that are valid for a file name in your operating system.
3 In the Description box, type a description for the project. The
description length has no limit and can include any characters.
4 Click Create.
From 1 Click Copy Project.
existing
2 In the Project to Copy box, click the project to copy.
3 In the New Project Name box, type the name to use for the new
project. The name can be up to 32 characters long and can include
any characters that are valid for a file name in your operating
system.
4 Click Copy Project.
5 Review the default properties for projects in the right pane and override any if
necessary, then click Save.
If the project is for My webMethods Server, you can set these properties for the project
under the MWS Deployment Options area:
6 Click Save.
1 All users you want to authorize to perform certain tasks (by authorizing their groups
to perform the tasks) must belong to the Developers security group. You can also
create intuitively named groups for use with the Deployer (for example, you could
create groups named viewProjects, buildProjects, mapProjects, deployProjects, and
defineProjects). For instructions on assigning users to security groups and creating
security groups, see the webMethods Integration Server Administrator’s Guide.
2 In the Deployer, go to the Projects page.
3 Locate the project to which you want to authorize users. In the Authorize column for
the project, click .
4 In the Select Authorization list, click a task to which you want to authorize users.
5 The Not Specified box lists all groups defined on the local Integration Server. Using the
arrow buttons, move each group that you want to assign to the selected task into the
Allowed box. Move each group that you do not want to assign to the selected task into
the Denied box.
6 Click Update.
The Resulting users with this Authorization lists all users that belong to the groups you
assigned to the task (that is, the groups you moved into the Allowed box).
To view the home page for the project, go to the Deployer Projects page and click in the
Home column for the project.
Resolve Dependencies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
1 In the Deployer, go to the Projects page. In the Name column, click the project in which
to create a Broker deployment set.
2 In the right pane, click Define.
3 Click Create Deployment Set.
4 In the Type list, click Broker, then complete these fields:
Box Entry
Name Name to use for the deployment set. The name can be up to 32
characters long and can include any characters that are valid
for a file name in your operating system.
Description Description for the deployment set. The description length has
no limit and can include any characters.
All other items After you choose the source Brokers, the Deployer will display
all items on those Brokers. You can use this field to narrow the
display. Specify a regular expression that specifies the text the
item names must contain in order to be listed.
5 Click Create.
1 On the Deployer ProjectsprojectDefine page, in the left pane in the Name column,
click the name of the deployment set for which to identify source Brokers. In the right
pane in the Select Source Broker Servers area, the Deployer lists all Brokers you set up to
work with the Deployer.
Note: If a Broker you want to use as a source does not appear in the list, you have not
yet set it up to work with the Deployer. For instructions, see “Configure for Brokers”
on page 19.
2 In the Select column, select the check box next to each Broker that contains items to
add to the deployment set.
3 Click Save.
1 In the Deployment Sets area, under the deployment set to which you want to add
webMethods Broker items, click the Broker folder. In the right pane, the Deployer lists
the source Brokers you identified.
2 In the right pane, open the tree to show the items on the source Brokers, then select
the check box next to each item to add to the deployment set.
Important! Do not add any client groups with a queue type of persistent and a lifecycle
set to explicit destroy. The Deployer cannot deploy such client groups.
3 Click Save. The Deployer shows your choices in the left pane under the Broker folder
for the deployment set.
Resolve Dependencies
For any project, the Deployer can determine when items that are in a deployment set
require other items that are not in the deployment set. The items that require other items
are called dependent items, while the items that are required are called referenced items. The
Deployer identifies missing referenced items as unresolved dependencies. For example, if
you add a client group to a Broker deployment set, but do not add the documents that the
client group can publish or subscribe to, then the Deployer identifies the documents as
unresolved dependencies.
In the project properties (“Set the Default for Dependency Checking” on page 24), you
indicated how you want to check dependencies in the deployment sets. When the
Deployer automatically checks dependencies and finds unresolved dependencies in a
deployment set, it shows in the Unresolved Dependencies or References column for the
deployment set; when there are no unresolved dependencies, the Deployer shows in
the column. When you can check dependencies manually, the Deployer shows in the
Unresolved Dependencies or References column for each deployment set; click Check next to
the . If necessary, you can later un‐resolve a dependency you have resolved and re‐
resolve it a different way.
To resolve dependencies:
1 In the Unresolved Dependencies column for the deployment set, click Check. The
Deployer shows all unresolved dependencies on the Unresolved Dependencies page.
The Referenced Item column lists the missing referenced items. The next column offers
the possible ways you can resolve the unresolved dependency. The Project Item column
shows the dependent items.
2 Tell the Deployer how to resolve each unresolved dependency as described below. If
you want to resolve all items in a folder the same way, you can set the resolution at the
folder level rather than at the level of the individual items.
Option Description
Add If the referenced item does not exist on the target servers and you want
to deploy the referenced item to them, use this option. The Deployer
adds the referenced item to the deployment set.
Exists If you believe the referenced item already exists on the target servers
and you want to continue working, but you want the Deployer to make
sure the item does in fact exist later, use this option. The Deployer will
check for the referenced item when you map the project to target
servers. If the Deployer does not find the item, an icon alerts you during
the mapping task.
If you do not address the problem during the mapping task, the
Deployer will write a message about the problem to the simulation
report. If you deploy without addressing the problem, the Deployer will
not deploy the deployment set.
Ignore If you want to bypass dependency checking for the referenced item at
this time so you can continue working, use this option. You might use
this option if the referenced item is missing on the source server. Missing
referenced items are marked with a question mark (?) on the Unresolved
Dependencies page.
Before deploying, make sure either that the referenced item exists on the
target server or that the referenced item is unnecessary. If the referenced
item does not exist on the target server, the Deployer might not be able
to deploy correctly; if it can deploy, the deployed items will not run
correctly.
The Deployer will list ignored items in the simulation report and in the
deployment report.
3 Click Save. The Deployer moves dependencies you resolved using the Exists or Ignore
option to the Resolved Dependencies page.
4 To see the resolved dependencies, click Resolved Dependencies.
You can un‐resolve a resolved dependency and re‐resolve it differently, as follows:
To un‐resolve a dependency you resolved using the Exists option or the Ignore
option, go to the Resolved Dependencies page, select the check box in the Delete
column for the resolved dependency, and click Delete.
To un‐resolve a dependency you resolved using the Add option, return to the
ProjectsprojectDefine page, open the folder that contains the item, navigate to
the item in the tree in the right pane, cancel the selection of the item by clearing
the appropriate check box, and save the deployment set.
When you un‐resolve a resolved dependency, the Deployer returns the dependency to
the Unresolved Dependencies page. Go to that page and re‐resolve the dependency.
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
Resolve Dependencies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
Overview
A ProcessModel deployment set can deploy process models from only one ProcessModel
server. If you want to deploy process models from multiple ProcessModel servers, you
must create a deployment set for each of those ProcessModel servers.
1 In the Deployer, go to the Projects page. In the Name column, click the project in which
to create a ProcessModel deployment set.
2 In the right pane, click Define.
3 Click Create Deployment Set.
4 From the Type list, select ProcessModel, then complete these fields:
Box Entry
Name Name to use for the deployment set. The name can be up to 32
characters long and can include any characters that are valid
for a file name in your operating system.
Description Description for the deployment set. The description length has
no limit and can include any characters.
Packages After you choose the source ProcessModel servers, the
Deployer will display all packages on those servers. You can
use this field to narrow the display. Specify a regular
expression that specifies the text the package names must
contain in order to be listed.
All other items After you choose the source ProcessModel servers, the
Deployer will display all items on those servers. You can use
this field to narrow the display. Specify a regular expression
that specifies the text the item names must contain in order to
be listed.
5 Click Create.
1 On the Deployer ProjectsprojectDefine page, in the left pane in the Name column,
click the name of the deployment set for which to identify the source ProcessModel
server. In the right pane in the Select Source ProcessModel Server area, the Deployer lists
all ProcessModel servers you set up to work with the Deployer.
Note: If the ProcessModel server you want to use as the source does not appear in the
list, you have not yet set it up to work with the Deployer. For instructions, see
“Configure for ProcessModel Servers” on page 19.
2 In the Select column, click the ProcessModel server that contains process models to
add to the deployment set.
3 Click Save.
1 In the Deployment Sets area, under the deployment set to which you want to add
process models, click the ProcessModel folder. In the right pane, the Deployer lists the
source ProcesModel server you identified.
2 In the right pane, open the tree to show the process models on the ProcessModel
server, then select the check box next to each process model to add to the deployment
set.
3 Click Save. The Deployer shows your choices in the left pane under the ProcessModel
folder for the deployment set.
Resolve Dependencies
For any project, the Deployer can determine when items that are in a deployment set
require other items that are not in the deployment set. The items that require other items
are called dependent items, while the items that are required are called referenced items. The
Deployer identifies missing referenced items as unresolved dependencies. For example, if
you add a process model to a ProcessModel deployment set, but do not add the flow
services that are called by that process model, the Deployer identifies the flow services as
unresolved dependencies.
In the project properties (“Set the Default for Dependency Checking” on page 24), you
indicated how you want to check dependencies in the deployment sets. When the
Deployer automatically checks dependencies and finds unresolved dependencies in a
deployment set, it shows in the Unresolved Dependencies or References column for the
deployment set; when there are no unresolved dependencies, the Deployer shows in
the column. When you can check dependencies manually, the Deployer shows in the
Unresolved Dependencies or References column for each deployment set; click Check next to
the . If necessary, you can later un‐resolve a dependency you have resolved and re‐
resolve it a different way.
To resolve dependencies:
1 In the Unresolved Dependencies column for the deployment set, click Check. The
Deployer shows all unresolved dependencies on the Unresolved Dependencies page.
The Referenced Item column lists the missing referenced items. The next column offers
the possible ways you can resolve the unresolved dependency. The Project Item column
shows the dependent items.
2 Tell the Deployer how to resolve each unresolved dependency as described below. If
you want to resolve all items in a folder the same way, you can set the resolution at the
folder level rather than at the level of the individual items.
Option Description
Add If the referenced item does not exist on the target servers and you want
to deploy the referenced item to them, use this option. The Deployer
adds the referenced item to the deployment set.
Exists If you believe the referenced item already exists on the target servers
and you want to continue working, but you want the Deployer to make
sure the item does in fact exist later, use this option. The Deployer will
check for the referenced item when you map the project to target
servers. If the Deployer does not find the item, an icon alerts you during
the mapping task.
If you do not address the problem during the mapping task, the
Deployer will write a message about the problem to the simulation
report. If you deploy without addressing the problem, the Deployer will
not deploy the deployment set.
Option Description
Ignore If you want to bypass dependency checking for the referenced item at
this time so you can continue working, use this option. You might use
this option if the referenced item is missing on the source server. Missing
referenced items are marked with a question mark (?) on the Unresolved
Dependencies page.
Before deploying, make sure either that the referenced item exists on the
target server or that the referenced item is unnecessary. If the referenced
item does not exist on the target server, the Deployer might not be able
to deploy correctly; if it can deploy, the deployed items will not run
correctly.
The Deployer will list ignored items in the simulation report and in the
deployment report.
3 Click Save. The Deployer moves dependencies you resolved using the Exists or Ignore
option to the Resolved Dependencies page.
4 To see the resolved dependencies, click Resolved Dependencies.
You can un‐resolve a resolved dependency and re‐resolve it differently, as follows:
To un‐resolve a dependency you resolved using the Exists option or the Ignore
option, go to the Resolved Dependencies page, select the check box in the Delete
column for the resolved dependency, and click Delete.
To un‐resolve a dependency you resolved using the Add option, return to the
ProjectsprojectDefine page, open the folder that contains the item, navigate to
the item in the tree in the right pane, cancel the selection of the item by clearing
the appropriate check box, and save the deployment set.
When you un‐resolve a resolved dependency, the Deployer returns the dependency to
the Unresolved Dependencies page. Go to that page and re‐resolve the dependency.
Resolve Dependencies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
1 In the Deployer, go to the Projects page. In the Name column, click the project in which
to create an IS & TN deployment set.
2 In the right pane, click Define.
3 Click Create Deployment Set.
4 From the Type list, select IS & TN, then complete these fields:
Box Entry
Name Name to use for the deployment set. The name can be up to 32
characters long and can include any characters that are valid
for a file name in your operating system.
Description Description for the deployment set. The description length has
no limit and can include any characters.
Packages After you choose the source Integration Server, the Deployer
will display all packages on those Integration Servers. You can
use this field to narrow the display. Specify a regular
expression that specifies the text the package names must
contain in order to be listed.
All other items After you choose the source Integration Servers, the Deployer
will display all items on those Integration Servers. You can use
this field to narrow the display. Specify a regular expression
that specifies the text the item names must contain in order to
be listed.
5 Click Create.
1 On the Deployer ProjectsprojectDefine page, in the left pane in the Name column,
click the deployment set for which you want to identify source Integration Servers. In
the right pane, in the Select Source Servers area, the Deployer lists all remote servers
defined to the Integration Server that hosts the Deployer.
Note: If an Integration Server you want to use as a source does not appear in the list,
you must define it as a remote server to the host Integration Server. This includes the
host Integration Server; to use it as a source, you must define it as a remote server to
itself. For instructions, see the webMethods Integration Server Administrator’s Guide.
After you define a new remote server with the Integration Server Administrator, click
Refresh this Page on the Deployer to update the list of servers.
2 In the Select column, select the check box next to each Integration Server that contains
items to add to the deployment set.
3 Click Save.
4 Make sure every Integration Server you identified as a source is equipped with the
Deployer resource package. To do so, go to the ServersIS & TN page and check
whether the Resource Package column displays the word “Installed” for those
Integration Servers. If the column does not show “Installed,” you must install the
resource package on the Integration Servers. In the Install column, select the check box
next to each Integration Server you identified as a source, then click Install.
1 In the Deployment Sets area, under the deployment set to which you want to add
Integration Server administration items, click the Administration folder. In the right
pane, the Deployer lists the source Integration Servers you identified.
2 In the right pane, open the tree to show the administration items on the source
Integration Servers, then select the check box next to each item to add to the
deployment set.
3 Click Save. The Deployer shows your choices in the left pane under the Administration
folder for the deployment set.
4 If you are not going to add any more items to the deployment set, go to “Resolve
Dependencies” on page 53.
Add selected package components only
Add selected package files only
A package from which you have added only selected components or files is called a partial
package. You can add both entire packages and partial packages to a deployment set.
If you add a partial package of only selected files to a deployment set and the package
already exists on target Integration Servers, you can have the Deployer delete specified
files from the existing package on the target Integration Servers after deployment. You
might use this feature if the existing package contains a service that has been superseded;
you would deploy the files that make up the new service and delete the files that make up
the old service.
1 In the Deployment Sets area, under the deployment set to which you want to add
packages, click the Packages folder. In the right pane, the Deployer lists the source
Integration Servers you identified.
2 In the right pane, open the tree to show the packages on the source Integration Servers,
then select the check boxes next to the packages you want to add in their entirety.
3 Click Save. The Deployer shows the entire package icon ( ) in the left pane under the
Packages folder for the deployment set. The Deployer also shows the entire package
icon for selected packages in the package list in the right pane, and the check boxes
next to the packages are marked with a black check mark.
4 If you are done adding packages to the deployment set, go to “Set the Package
Properties” on page 49.
1 In the Deployment Sets area, under the deployment set to which you want to add
package components, click the Packages folder. In the right pane, the Deployer lists
the source Integration Servers you identified.
2 In the right pane, open the tree to show the packages on the source Integration Servers,
then click the name of a package that contains components you want to add to the
deployment set.
3 In the Select Components area, open the tree to show the components in the package,
then select the check box next to each component to add to the deployment set.
4 Click Save.
Note: If you later decide you want to include the entire package (as opposed to
selected components only), cancel the selection of the components by clicking the
name of the partial package, clearing all checked boxes, and clicking Save. Then save
the deployment set and select the entire package as explained in “Add an Entire
Package” on page 46.
6 If you are done adding packages to the deployment set, go to “Set the Package
Properties” on page 49.
1 In the Deployment Sets area, under the deployment set to which to add package files,
click the Packages folder. In the right pane, the Deployer lists the source Integration
Servers you identified.
2 In the right pane, open the tree to show the packages on the source Integration
Servers, then click the name of a package that contains files you want to add to the
deployment set.
3 Click Select Files. The Deployer lists all files in the package.
Do one of the following:
4 If a package of the same name as this partial package already exists on one of the
deployment set’s target Integration Servers, and the existing package contains files
you want to delete after deployment, type the fully qualified names of the files to
delete in the Files to Delete from Target box. Type each file name on its own line, and end
each line with a semicolon (;). For example:
code/classes/wm/administratorResource/admin.class;
code/classes/wm/administratorResource/user.class;
ns/wm/administratorResource/
5 Click Save.
Note: If you later decide you want to include the entire package (as opposed to
selected files only), cancel the selection of the files by navigating to the file selection
page and clicking None, then save the deployment set, and then select the entire
package as explained in “Add an Entire Package” on page 46.
7 If you are done adding packages to the deployment set, go to “Set the Package
Properties” on page 49.
1 In the Deployment Sets area, under the deployment set to which you added entire or
partial packages, open the tree under the Packages folder and click a package.
2 In the package_name Properties area, specify the following:
Box Entry
Package Type Use this property when the source package already exists on the
target Integration Servers. You can use the options below for entire
packages and for partial packages.
If you want the Deployer to... Click...
Deploy the source package, replacing the existing package Full
entirely. When you choose to deploy an entire package,
this is the default.
Box Entry
Deploy the components and files in the source package Patch
over the corresponding components and files in the
existing package. When you choose to deploy package
components, package files, or both, this is the default.
Note: Before you deploy a project, you can find out which items the
Deployer will overwrite by generating the simulation report.
Version Supply the version number to use for the source package in
comparisons with existing packages on target Integration Servers.
Whether the Deployer actually deploys depends on the version
numbers of the source package and the existing package. If the
source package’s version number is the same or higher than the
existing package’s version number, the Deployer deploys. If the
source package’s version number is lower than the existing
package’s version number, the Deployer does not deploy.
Note: The version number for the source package on the source
Integration Server is not affected by your entry here.
Build Supply the build number to assign to the deployed package on the
target Integration Servers. Specify a build number only if you
selected Full for Package Type.
Patches Supply the list of patches that have been applied to the deployed
Included package on the target Integration Servers. Specify the patch
numbers, separated by commas (for example, 44, 45, 55). Specify
patches only if you selected Full for Package Type.
Brief Supply a description to use for the deployed package on the target
Description Integration Servers (for example, “December 2003 release with
patches to correct Order Process problem.”) Specify a description
only if you selected Full for Package Type.
time the user makes a build instead of the package version and build numbers
specified in the package_name Properties area.
5 In the Package Deployment Options area, specify the following:
Option Entry
Activate After How the Deployer should deploy the package.
Deployment
If you want to... Click
The Deployer deploys the package to the Integration Activate
Server_directory\replicate\inbound directory on the
target Integration Servers, then enables the package.
The Deployer deploys the package to the Integration Install
Server_directory\replicate\inbound directory on the Only
target Integration Servers and installs it but does not
enable it.
The Deployer deploys the package to the Integration Inbound
Server_directory\replicate\inbound directory on the Only
target Integration Servers but does not install or
enable it.
Sync Whether the Deployer should synchronize the publishable IS
Document document types in the source package with documents types on
Types the Brokers that are connected to the target Integration Servers.
The connected Brokers must be available at deployment time for
synchronization to occur. If a connected Broker is not available, IS
document types are not synchronized for the Integration Server to
which the Broker is connected. The Deployer writes a message to
that effect to the deployment report.
Note: The Deployer can detect Broker unavailability when you
generate the simulation report and will write a message advising
you of the problem to the report.
Synchronize all publishable IS document types in the New
package that are new to the target Integration Servers.
Do not synchronize IS document types in the package
that already exist on the target Integration Servers,
even if they have been modified.
Option Entry
Synchronize all publishable IS document types in the All
package.
Not synchronize any IS document types. None
6 If you indicated in the project properties that you want to suspend individual triggers
during deployment, click Suspend Triggers, select the check box next to each trigger to
suspend, click Suspend, and then return to the previous page.
7 If you indicated in the project properties that you want to suspend individual adapter
notifications during deployment, click Suspend Notifications, select the check box next
to each notification to suspend, click Suspend, and then return to the previous page.
Important! If you suspend a particular adapter notification but the notification does not
exist on a target Integration Server, you will not be able to deploy. Only suspend
notifications that already exist on all target Integration Servers.
8 Click Save.
9 Repeat these steps for each package in the deployment set.
10 If you are not going to add any more items to the deployment set, go to “Resolve
Dependencies” on page 53.
1 In the Deployment Sets area, under the deployment set to which you want to add
webMethods files, click the webMethods Files folder. In the right pane, the Deployer
lists the source Integration Servers you identified.
2 In the right pane, open the tree to show the webMethods installation directory and its
contents on the source Integration Servers. Select the check box next to each file to add
to the deployment set.
3 Click Save. The Deployer shows your choices in the left pane under the webMethods
Files folder for the deployment set.
4 If you are not going to add any more items to the deployment set, go to “Resolve
Dependencies” on page 53.
1 In the Deployment Sets area, under the deployment set to which you want to add
Trading Networks items, click the Trading Networks folder. In the right pane, the
Deployer lists the source Integration Servers you identified.
2 In the right pane, open the tree to show the Trading Networks items on the source
Integration Servers, then select the check box next to each item to add to the
deployment set.
Important! If you add a processing rule from a 6.0.1 Trading Networks to a deployment
set, the processing rule might require custom contact types. If the custom contact
types do not already exist on the target Integration Servers, you must add them to the
deployment set. If you do not, the Deployer will log an error to the deployment report
and the processing rule will not be deployed. (This is not an issue when adding
processing rules from a 6.1 Trading Networks.)
If you add a TN document type with the Duplicate Checking option set to Use
Custom Service, the Deployer does not detect the dependency on the service. If the
service does not already exist on the target Integration Servers, you must add the
service to the deployment set. If you do not, the Deployer will log an error to the
deployment report and will not deploy the TN document type.
3 Click Save. The Deployer shows your choices in the left pane under the Trading
Networks folder for the deployment set.
4 If you are not going to add any more items to the deployment set, go to “Resolve
Dependencies”, below.
Resolve Dependencies
For any project, the Deployer can determine when items that are in a deployment set
require other items that are not in the deployment set. The items that require other items
are called dependent items, while the items that are required are called referenced items. The
Deployer identifies missing referenced items as unresolved dependencies. For example, if
you add a trigger to an IS & TN deployment set, but do not add the service that is invoked
by the trigger, the Deployer identifies the service as an unresolved dependency.
In the project properties (“Set the Default for Dependency Checking” on page 24), you
indicated how you want to check dependencies in the deployment sets. When the
Deployer automatically checks dependencies and finds unresolved dependencies in a
deployment set, it shows in the Unresolved Dependencies or References column for the
deployment set; when there are no unresolved dependencies, the Deployer shows in
the column. When you can check dependencies manually, the Deployer shows in the
Unresolved Dependencies or References column for each deployment set; click Check next to
the . If necessary, you can later un‐resolve a dependency you have resolved and re‐
resolve it a different way.
To resolve dependencies:
1 In the Unresolved Dependencies column for the deployment set, click Check. The
Deployer shows all unresolved dependencies on the Unresolved Dependencies page.
The Referenced Item column lists the missing referenced items. The next column offers
the possible ways you can resolve the unresolved dependency. The Project Item column
shows the dependent items.
2 Tell the Deployer how to resolve each unresolved dependency as described below. If
you want to resolve all items in a folder the same way, you can set the resolution at the
folder level rather than at the level of the individual items.
Option Description
Add If the referenced item does not exist on the target servers and you want
to deploy the referenced item to them, use this option. The Deployer
adds the referenced item to the deployment set.
Exists If you believe the referenced item already exists on the target servers
and you want to continue working, but you want the Deployer to make
sure the item does in fact exist later, use this option. The Deployer will
check for the referenced item when you map the project to target
servers. If the Deployer does not find the item, an icon alerts you during
the mapping task.
If you do not address the problem during the mapping task, the
Deployer will write a message about the problem to the simulation
report. If you deploy without addressing the problem, the Deployer will
not deploy the deployment set.
Option Description
Ignore If you want to bypass dependency checking for the referenced item at
this time so you can continue working, use this option. You might use
this option if the referenced item is missing on the source server. Missing
referenced items are marked with a question mark (?) on the Unresolved
Dependencies page.
Before deploying, make sure either that the referenced item exists on the
target server or that the referenced item is unnecessary. If the referenced
item does not exist on the target server, the Deployer might not be able
to deploy correctly; if it can deploy, the deployed items will not run
correctly.
The Deployer will list ignored items in the simulation report and in the
deployment report.
3 Click Save. The Deployer moves dependencies you resolved using the Exists or Ignore
option to the Resolved Dependencies page.
4 To see the resolved dependencies, click Resolved Dependencies.
You can un‐resolve a resolved dependency and re‐resolve it differently, as follows:
To un‐resolve a dependency you resolved using the Exists option or the Ignore
option, go to the Resolved Dependencies page, select the check box in the Delete
column for the resolved dependency, and click Delete.
To un‐resolve a dependency you resolved using the Add option, return to the
ProjectsprojectDefine page, open the folder that contains the item, navigate to
the item in the tree in the right pane, cancel the selection of the item by clearing
the appropriate check box, and save the deployment set.
When you un‐resolve a resolved dependency, the Deployer returns the dependency to
the Unresolved Dependencies page. Go to that page and re‐resolve the dependency.
about it to the simulation report and, if you do not resolve it at that time, to the
deployment report.
1 Go to the Resolved Dependencies page as explained in the previous section.
2 Under the Manually Add Dependency area, in the Referenced Package box, type the name
of the package that contains the referenced component.
3 In the Referenced Component box, type the name of the referenced component.
4 Click Add.
You can remove a dependency you added manually. To do so, return to the
ProjectsprojectDefine page, open the folder that contains the item, navigate to the
item in the tree in the right pane, cancel the selection of the item by clearing the
appropriate check box, and save the deployment set.
Resolve Dependencies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
1 In the Deployer, go to the Projects page. In the Name column, click the project in which
to create a MWS deployment set.
2 In the right pane, click Define.
3 Click Create Deployment Set.
4 In the Type list, click MWS, then complete these fields:
Box Entry
Name Name to use for the deployment set. The name can be up to 32
characters long and can include any characters that are valid
for a file name in your operating system.
Description Description for the deployment set. The description length has
no limit and can include any characters.
All other items After you choose the source My webMethods Servers, the
Deployer will display all items on those My webMethods
Servers. You can use this field to narrow the display. Specify a
regular expression that specifies the text the item names must
contain in order to be listed.
5 Click Create.
1 On the Deployer ProjectsprojectDefine page, in the left pane in the Name column,
click the name of the deployment set for which to identify source My webMethods
Servers. In the right pane in the Select Source MWS Servers area, the Deployer lists all
My webMethods Servers you set up to work with the Deployer.
Note: If a My webMethods Server you want to use as a source does not appear in the
list, you have not yet set it up to work with the Deployer. For instructions, see
“Configure for My webMethods Servers” on page 21.
2 In the Select column, select the check box next to each My webMethods Server that
contains items to add to the deployment set.
3 Click Save.
1 In the Deployment Sets area, under the deployment set to which you want to add My
webMethods Server items, click the MWS folder. In the right pane, the Deployer lists
the source My webMethods Servers you identified.
2 In the right pane, open the tree to show the items on the My webMethods Servers,
then select the check box next to each item to add to the deployment set.
Note: In the Deployer, each My webMethods Server folder is listed twice within its
directory, as a container preceded by and as an item preceded by . If you want to
add a folder with all the items it contains to the deployment set, select the folder
where it appears next to the square icon. If you want to add individual items in the
folder without adding the folder itself, open the folder where it appears as a container
and click the items to add.
3 Click Save. The Deployer shows your choices in the left pane under the My webMethods
Server folder for the deployment set.
Resolve Dependencies
For any project, the Deployer can determine when items that are in a deployment set
require other items that are not in the deployment set. The items that require other items
are called dependent items, while the items that are required are called referenced items. The
Deployer identifies missing referenced items as unresolved dependencies. For example, if
you add a page to an MWS deployment set, but do not add the portlets that are referenced
by that page, the Deployer identifies the portlets as unresolved dependencies.
In the project properties, you indicated whether you want the Deployer to automatically
check the deployment sets in the project for unresolved dependencies each time you add
an item to a deployment set, or whether you want to initiate dependency checking for
deployment sets manually.
If you chose automatic checking, the Deployer automatically checks dependencies each
time you add an item to a deployment set or go to the ProjectsprojectDefine page. If it
finds unresolved dependencies, the Deployer shows in the Unresolved Dependencies
column for the deployment set. If there are no unresolved dependencies, the Deployer
shows in that column.
You must resolve all unresolved dependencies before you can continue to the next task
(that is, the project build task). If necessary, you can later un‐resolve a dependency you
have resolved and re‐resolve it a different way.
To resolve dependencies:
1 In the Unresolved Dependencies column for the deployment set, click Check. The
Deployer shows all unresolved dependencies on the Unresolved Dependencies page.
The Referenced Item column lists the missing referenced items. The next column offers
the possible ways you can resolve the unresolved dependency. The Project Item
column shows the dependent items.
2 Tell the Deployer how to resolve each unresolved dependency as described below. If
you want to resolve all items in a folder the same way, you can set the resolution at the
folder level rather than at the level of the individual items.
Option Description
Add If the referenced item does not exist on the target servers and you want
to deploy the referenced item to them, use this option. The Deployer
adds the referenced item to the deployment set.
Exists If you believe the referenced item already exists on the target servers
and you want to continue working, but you want the Deployer to make
sure the item does in fact exist later, use this option. The Deployer will
check for the referenced item when you map the project to target servers.
If the Deployer does not find the item, an icon alerts you during the
mapping task.
If you do not address the problem during the mapping task, the
Deployer will write a message about the problem to the simulation
report. If you deploy without addressing the problem, the Deployer will
not deploy the deployment set.
Ignore If you want to bypass dependency checking for the referenced item at
this time so you can continue working, use this option. You might use
this option if the referenced item is missing on the source server. Missing
referenced items are marked with a question mark (?) on the Unresolved
Dependencies page.
Before deploying, make sure either that the referenced item exists on the
target server or that the referenced item is unnecessary. If the referenced
item does not exist on the target My webMethods Server, the Deployer
might not be able to deploy correctly; if it can deploy, the deployed items
will not run correctly.
The Deployer will list ignored items in the simulation report and in the
deployment report.
3 Click Save. The Deployer moves dependencies you resolved using the Exists or Ignore
option to the Resolved Dependencies page.
4 To see the resolved dependencies, click Resolved Dependencies.
You can un‐resolve a resolved dependency and re‐resolve it differently, as follows:
To un‐resolve a dependency you resolved using the Exists option or the Ignore
option, go to the Resolved Dependencies page, select the check box in the Delete
column for the resolved dependency, and click Delete.
To un‐resolve a dependency you resolved using the Add option, return to the
ProjectsprojectDefine page, open the folder that contains the item, navigate to
the item in the tree in the right pane, cancel the selection of the item by clearing
the appropriate check box, and save the deployment set.
When you un‐resolve a resolved dependency, the Deployer returns the dependency to
the Unresolved Dependencies page. Go to that page and re‐resolve the dependency.
Add Items . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
Resolve Dependencies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
1 In the Deployer, go to the Projects page. In the Name column, click the project in which
to create an Optimize deployment set.
2 In the right pane, click Define.
3 Click Create Deployment Set.
4 In the Type list, click Optimize, then complete these fields:
Box Entry
Name Name to use for the deployment set. The name can be up to 32
characters long and can include any characters that are valid
for a file name in your operating system.
Description Description for the deployment set. The description length has
no limit and can include any characters.
All other items After you choose the source Optimize servers, the Deployer
will display all items on those servers. You can use this field to
narrow the display. Specify a regular expression that specifies
the text the item names must contain in order to be listed.
5 Click Create.
1 On the Deployer ProjectsprojectDefine page, in the left pane in the Name column,
click the name of the deployment set for which to identify source Optimize servers. In
the right pane in the Select Source Optimize Servers area, the Deployer lists all Optimize
servers you set up to work with the Deployer.
Note: If an Optimize server you want to use as a source does not appear in the list, you
have not yet set it up to work with the Deployer. For instructions, see “Configure for
Optimize Servers” on page 22.
2 In the Select column, select the check box next to each Optimize server that contains
items to add to the deployment set.
3 Click Save.
Add Items
To add items:
1 In the Deployment Sets area, under the deployment set to which you want to add
Optimize items, click the Optimize folder. In the right pane, the Deployer lists the
source Optimize servers you identified.
2 In the right pane, open the tree to show the items on the Optimize servers, then select
the check box next to each item to add to the deployment set.
3 Click Save. The Deployer shows your choices in the left pane under the Optimize folder
for the deployment set.
Resolve Dependencies
For any project, the Deployer can determine when items that are in a deployment set
require other items that are not in the deployment set. The items that require other items
are called dependent items, while the items that are required are called referenced items. The
Deployer identifies missing referenced items as unresolved dependencies. For example, if
you add a rule to a deployment set, but do not add the dimensions used by that rule, the
Deployer identifies the dimensions as unresolved dependencies.
In the project properties (“Set the Default for Dependency Checking” on page 24), you
indicated how you want to check dependencies in the deployment sets. When the
Deployer automatically checks dependencies and finds unresolved dependencies in a
deployment set, it shows in the Unresolved Dependencies or References column for the
deployment set; when there are no unresolved dependencies, the Deployer shows in
the column. When you can check dependencies manually, the Deployer shows in the
Unresolved Dependencies or References column for each deployment set; click Check next to
the . If necessary, you can later un‐resolve a dependency you have resolved and re‐
resolve it a different way.
To resolve dependencies:
1 In the Unresolved Dependencies column for the deployment set, click Check. The
Deployer shows all unresolved dependencies on the Unresolved Dependencies page.
The Referenced Item column lists the missing referenced items. The next column offers
the possible ways you can resolve the unresolved dependency. The Project Item column
shows the dependent items.
2 Tell the Deployer how to resolve each unresolved dependency as described below. If
you want to resolve all items in a folder the same way, you can set the resolution at the
folder level rather than at the level of the individual items.
Option Description
Add If the referenced item does not exist on the target servers and you want
to deploy the referenced item to them, use this option. The Deployer
adds the referenced item to the deployment set.
Exists If you believe the referenced item already exists on the target servers
and you want to continue working, but you want the Deployer to make
sure the item does in fact exist later, use this option. The Deployer will
check for the referenced item when you map the project to target
servers. If the Deployer does not find the item, an icon alerts you during
the mapping task.
If you do not address the problem during the mapping task, the
Deployer will write a message about the problem to the simulation
report. If you deploy without addressing the problem, the Deployer will
not deploy the deployment set.
Ignore If you want to bypass dependency checking for the referenced item at
this time so you can continue working, use this option. You might use
this option if the referenced item is missing on the source server. Missing
referenced items are marked with a question mark (?) on the Unresolved
Dependencies page.
Before deploying, make sure either that the referenced item exists on the
target server or that the referenced item is unnecessary. If the referenced
item does not exist on the target server, the Deployer might not be able
to deploy correctly; if it can deploy, the deployed items will not run
correctly.
The Deployer will list ignored items in the simulation report and in the
deployment report.
3 Click Save. The Deployer moves dependencies you resolved using the Exists or Ignore
option to the Resolved Dependencies page.
4 To see the resolved dependencies, click Resolved Dependencies.
You can un‐resolve a resolved dependency and re‐resolve it differently, as follows:
To un‐resolve a dependency you resolved using the Exists option or the Ignore
option, go to the Resolved Dependencies page, select the check box in the Delete
column for the resolved dependency, and click Delete.
To un‐resolve a dependency you resolved using the Add option, return to the
ProjectsprojectDefine page, open the folder that contains the item, navigate to
the item in the tree in the right pane, cancel the selection of the item by clearing
the appropriate check box, and save the deployment set.
When you un‐resolve a resolved dependency, the Deployer returns the dependency to
the Unresolved Dependencies page. Go to that page and re‐resolve the dependency.
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
Rebuild a Build . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
Overview
You can create a project build using the Deployer interface or a script.
If a project build you create becomes out of sync with the current project definition, or if
the build contains items that you know have changed on the source servers, you can
rebuild the project build to bring it up to date. You can rebuild a project build using the
Deployer interface or a script.
If the Deployer in the source environment cannot access the target environment, you must
use the source environment Deployer to export the project build, then use the target
environment Deployer to import (and later, map and deploy) the build. You can export
and import a project build using the Deployer interface or a script.
Important! If you encounter unresolved dependencies during this step, resolve them using
the instructions in the “Resolve Dependencies” section in any chapter on defining
projects.
1 In the Deployer, go to the Projects page.
2 In the Name column, click the name of the project for which you want to create a build.
3 In the Description box, you can type a description for the deployment map. The
description can be of any length and can include any characters.
4 Click Build. The Deployer displays the ProjectsprojectBuild page and lists all
builds that exist for the selected project.
Note: The Status column on the ProjectsprojectBuild page indicates whether each
project build is in sync with the current project definition. If the build and the current
project definition are in sync, the column shows . If the project definition has
changed since the build was created, the column shows . You can rebuild such a
project if you want. For instructions, see “Rebuild a Build” on page 72.
5 In the left pane, click Create Build.
6 In the Name box accept the default build name or replace it with a name that you
choose. The name can be up to 32 characters long and can include any characters that
are valid for a file name in your operating system.
7 Click Create. The Deployer creates the build and stores it as a file in the Integration
Server_directory/packages/WmDeployer/persist/projects/project_name/builds directory.
Important! If the project for which you are trying to create the build contains
unresolved dependencies, you will receive a message to that effect and the build
process will fail. For instructions on viewing and resolving unresolved dependencies,
see the appropriate section in the chapters on defining projects.
Parameter Description
-h host -p port Host name or IP address and port for the Deployer host
machine.
-u user -x password User name and password to use to log onto the Deployer.
-r build_name Name to use for the build. The name can be up to 32
characters long and can include any characters that are
valid for a file name in your operating system.
-f project Name of the project from which to create the build.
[-d] Optional. Tells the Deployer to write debug information to
standard output. By default, standard output echoes to
your console.
Rebuild a Build
The Status column on the ProjectsprojectBuild page indicates whether each project
build is in sync with the current project definition. If the build and the current project
definition are in sync, the column shows . If the project definition has changed since the
build was created, the column shows .
If a project build is out of sync with the current project definition or contains items that
you know have changed on the source servers, and you want to recreate the build to bring
it up to date, click in the Rebuild column for the build.
If you want to rebuild a build using a script, use the CreateBuild script as described in
“Create a Build Using a Script”, above, and specify the name of the existing build on the ‐
r build_name parameter.
1 Export the build from the source environment as follows:
a In the Deployer in the source environment, go to the ProjectsprojectBuild page.
b Locate the build to export and click in the build’s Export column. The Deployer
creates a file that contains the build. The file is named project_build and is stored in
the Integration Server_directory\packages\WmDeployer\replicate\outbound
folder.
2 Move the build into the target environment as follows:
a Copy the project_build file to the Integration Server_directory\packages\
WmDeployer\replicate\inbound directory on the machine that hosts the
Deployer in the target environment.
b In the Deployer in the target environment, go to the ToolsImport Build page.
c In the Project Build list, click the project_build file you just copied to the inbound
directory.
d Click Import.
3 Use the Deployer in the target environment to map and deploy the build as described
in the appropriate chapters in this guide.
Parameter Description
-h host -p port Host name or IP address and port for the Deployer host
machine.
-u user -x password User name and password to use to log onto the Deployer.
-f project Name of the project from which the build you are
exporting was created.
-r build Name of the build to export.
-o {true|false} Indicates whether to overwrite an existing project_build
file, if there is one.
[-d] Optional. Tells the Deployer to write debug information to
standard output. By default, standard output echoes to
your console.
The syntax for the ImportBuild script is as follows:
ImportBuild -h host -p port -u user -x password -r project_build [-d]
Parameter Description
-h host -p port Host name or IP address and port for the Deployer host
machine.
-u user -x password User name and password to use to log onto the Deployer.
-r project_build Name of the file to import (project_build).
[-d] Optional. Tells the Deployer to write debug information to
standard output. By default, standard output echoes to
your console.
Important! If you encounter unresolved dependencies during this step, resolve them using
the instructions in the “Resolve Dependencies” section in any chapter on defining
projects.
1 In the Deployer, go to the Projects page.
2 In the Name column, click the name of the project to map.
3 In the right pane, click Map. The Deployer displays the Projects project Map
page and lists all maps that exist for the selected project.
4 In the left pane, click Create Deployment Map.
5 In the Name box, accept the default deployment map name or replace it with a name
that you choose. The name can be up to 32 characters long and can include any
characters that are valid for a file name in your operating system.
6 In the Description box, type a description for the deployment map. The description
length has no limit and can include any characters.
7 Click Create.
8 Under the Deployment Map Topology area, in the Deployment Set Mapping column for a
deployment set, click Add Target Server, then do the following:
Note: If the Integration Servers you want to use as targets do
not appear in the list, you must define them as remote servers
to the host Integration Server. For instructions, see the
webMethods Integration Server Administrator’s Guide. After you
define a new remote server with the Integration Server
Administrator, click Refresh this Page on the Deployer to
update the list of servers.
9 When the Deployer returns to the map Properties page, the Reference Items column in
the Deployment Map Topology area might show . This icon indicates that you resolved
an unresolved dependency using the Exists option, but the Deployer has found that
the referenced item does not exist on target servers. Click to see the missing
referenced item. You can then place the referenced item on the target servers, or you
can return to the project definition stage and re‐resolve the dependency in a different
way (see the appropriate section in the chapters on defining projects).
Important! If you do not address the problem at this time, the Deployer will write a
message about it to the simulation report. If you deploy without addressing the
problem, the Deployer will not deploy the deployment set.
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
Generate a Checkpoint . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
Overview
You can deploy a project using the Deployer interface or using a script.
Before you deploy, you can generate a simulation report that contains information such as
the following:
Items that will be suspended during deployment
Items that will be enabled after deployment
Changes that will occur on the target servers, such as the items that will be added or
overwritten, and configuration parameters that will be changed
Messages about problems, such as unresolved dependencies
You set up the checkpoint and rollback features in the project properties (see “Set the
Default Properties for Deployment” on page 24). If you chose automatic checkpoint or
rollback, the Deployer automatically generates a checkpoint as the first step when you
actually deploy the project. If you chose automatic rollback and the deployment fails, the
Deployer automatically rolls back the target servers to the checkpoint.
If you chose manual checkpoint, you can generate a checkpoint at any time before you
deploy. If you take multiple checkpoints for a deployment candidate, only the latest is
retained. If you chose manual rollback, you can roll back the target servers manually at
any time after a deployment.
Important! If you encounter unresolved dependencies during this step, resolve them using
the instructions in the “Resolve Dependencies” section in any chapter on defining
projects.
1 If you chose to suspend triggers, ports, and scheduled tasks, but a service is triggered
by one of these items before the Deployer suspends them, and the service is a long‐
running service, the Deployer might overwrite the service during deployment. Make
sure long‐running services have completed.
2 In the Deployer, go to the Projects page. In the Name column, click the name of the
project you want to deploy.
3 In the right pane, click Deploy. The Deployer displays the ProjectsprojectDeploy
page and lists all deployment candidates that exist for the selected project.
4 In the left pane, click Create Deployment Candidate.
5 In the Name box, accept the default deployment candidate name or replace it with a
name that you choose. The name can be up to 32 characters long and can include any
characters that are valid for a file name in your operating system.
In the Description box, type a description for the deployment set. The description
length has no limit and can include any characters.
6 In the Project Build list, click the project build to deploy.
7 In the Deployment Map list, click the deployment map that identifies the target servers
to which to deploy the items in the selected project build.
If you do not address the problem during the mapping task, the Deployer will write a
message about the problem to the simulation report. If you deploy without
addressing the problem, the Deployer will not deploy the deployment set.
8 Click Create. In the candidate list in the left pane, if the selected build and the current
project definition are in sync, the Status column shows . If the project definition has
changed since the build was created, the column shows . You can rebuild the
project build before proceeding. For instructions, see “Rebuild a Build” on page 72.
Important! If you do not address all problems at this time, you will probably experience
errors during the deployment.
10 To deploy the project, click in the Deploy column for the deployment candidate.
The Deployer does the following:
If you chose automatic checkpointing or automatic rollback in the project
properties, automatically generates a checkpoint at this time.
Deploys the items in the selected project build to the target servers you identified
in the selected deployment map.
Creates a deployment report and lists the report in the Deployment History area.
Click next to Deployment Report in the Report Type column to display the report.
The report contains similar information to the simulation report, except that the
events have actually occurred at this point.
Parameter Description
-h host -p port Host name or IP address and port for the
Deployer host machine.
-u user -x password User name and password to use to log onto the
Deployer.
-r project Name of the project from which the build you are
deploying was created.
-n new_candidate_name Name to use for a new deployment candidate.
The name can be up to 32 characters long and can
include any characters that are valid for a file
name in your operating system.
Parameter Description
-s new_candidate_description Brief description of the new deployment
candidate. The description length has no limit
and can include any characters that are valid for a
file name in your operating system.
-b build Name of the project build to deploy.
-m map Name of the deployment map that identifies the
target servers to which to deploy the items in the
selected project build.
-l {true|false} Tells the Deployer to either generate the
simulation report (true) or to deploy the
deployment candidate (false).
[-d] Optional. Tells the Deployer to write debug
information to standard output. By default,
standard output echoes to your console.
If you are deploying an existing deployment candidate, the syntax for the Deploy script is
as follows:
Deploy -h host -p port -u user -x password -e candidate -r project
-l {true|false} [-d]
Parameter Description
-h host -p port Host name or IP address and port for the
Deployer host machine.
-u user -x password User name and password to use to log onto the
Deployer.
-e candidate Name of an existing deployment candidate to
deploy.
-r project Name of the project from which the build you are
deploying was created.
-l {true|false} Tells the Deployer to either generate the
simulation report (true) or to deploy the
deployment candidate (false).
[-d] Optional. Tells the Deployer to write debug
information to standard output. By default,
standard output echoes to your console.
Generate a Checkpoint
The target servers must be available for the checkpoint generation to be successful.
To generate a checkpoint:
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
Overview
The Deployer can deploy to clustered 6.5 Integration Servers. The Deployer can deploy all
Integration Server items in a clustered environment except existing scheduled tasks (see
“Deploy Scheduled Tasks in a Cluster”). For complete information on Integration Server
clustering, see the webMethods Integration Server Clustering Guide.
Note: When you checkpoint a cluster, the assets are checkpointed from the primary node.
Important! Do not deploy to multiple Integration Servers in the cluster. The Trading
Networks database component would be updated multiple times, and you might
experience unexpected results.
1 In the Integration Server Administrator for the Integration Server you want to use as
the primary node, do the following:
a Specify a remote server alias for every Integration Server, or node, in the cluster,
including the primary node.
b On the watt.server.cluster.aliasList property, list the remote server alias
name of every node in the cluster, including the primary node. Separate the alias
names using a comma.
For detailed instructions, see the webMethods Integration Server Administrator’s Guide.
Important! Integration Server uses this list when executing the remote invokes that
update the other cluster nodes. If you do not include a cluster node on this list, the
Deployer will not propagate the deployment to that node.
2 Install the WmDeployerResource package to the primary node only. For instructions,
see “Configure for Integration Servers” on page 21.
3 Map projects for the cluster to the primary cluster node only. For instructions, see
Chapter 10, “Mapping a Project”.
Important! Do not map to any other nodes in the cluster, or the project will be deployed
multiple times.
1 Create a separate project that contains the scheduled tasks to deploy.
2 Define every Integration Server in the cluster as a remote server. For instructions, see
the webMethods Integration Server Administrator’s Guide.
3 Map the project to every Integration Server in the cluster.
4 Checkpoint, deploy, and, if necessary, roll back the project as you would normally do
in an unclustered environment.