Cloudera ODBC

Driver for Apache

Hive
Important Notice

© 2010-2019 Cloudera, Inc. All rights reserved.

Cloudera, the Cloudera logo, and any other product or service names or slogans contained in this
document, except as otherwise disclaimed, are trademarks of Cloudera and its suppliers or
licensors, and may not be copied, imitated or used, in whole or in part, without the prior written
permission of Cloudera or the applicable trademark holder.

Hadoop and the Hadoop elephant logo are trademarks of the Apache Software Foundation. All
other trademarks, registered trademarks, product names and company names or logos
mentioned in this document are the property of their respective owners. Reference to any
products, services, processes or other information, by trade name, trademark, manufacturer,
supplier or otherwise does not constitute or imply endorsement, sponsorship or
recommendation thereof by us.

Complying with all applicable copyright laws is the responsibility of the user. Without limiting the
rights under copyright, no part of this document may be reproduced, stored in or introduced into
a retrieval system, or transmitted in any form or by any means (electronic, mechanical,
photocopying, recording, or otherwise), or for any purpose, without the express written
permission of Cloudera.

Cloudera may have patents, patent applications, trademarks, copyrights, or other intellectual
property rights covering subject matter in this document. Except as expressly provided in any
written license agreement from Cloudera, the furnishing of this document does not give you any
license to these patents, trademarks copyrights, or other intellectual property.

The information in this document is subject to change without notice. Cloudera shall not be liable
for any damages resulting from technical errors or omissions which may be present in this
document, or from use of this document.

Cloudera, Inc.
1001 Page Mill Road, Building 2
Palo Alto, CA 94304-1008
[email protected]
US: 1-888-789-1488
Intl: 1-650-843-0595
www.cloudera.com

Release Information

Version: 2.6.4

Date: March 15, 2019

2 | Cloudera ODBC Driver for Apache Hive

Table of Contents
ABOUT THE CLOUDERA ODBC DRIVER FOR APACHE HIVE 5
WINDOWS DRIVER 6
WINDOWS SYSTEM REQUIREMENTS 6
INSTALLING THE DRIVER ON WINDOWS 6
CREATING A DATA SOURCE N AME ON WINDOWS 7
CONFIGURING A DSN-LESS CONNECTION ON WINDOWS 10
CONFIGURING A UTHENTICATION ON WINDOWS 12
CONFIGURING A DVANCED O PTIONS ON WINDOWS 16
CONFIGURING HTTP O PTIONS ON WINDOWS 18
CONFIGURING SSL V ERIFICATION ON WINDOWS 19
CONFIGURING THE TEMPORARY TABLE FEATURE 20
CONFIGURING SERVER -SIDE PROPERTIES ON WINDOWS 21
CONFIGURING L OGGING O PTIONS ON WINDOWS 22
CONFIGURING KERBEROS A UTHENTICATION FOR WINDOWS 24
V ERIFYING THE DRIVER V ERSION N UMBER ON WINDOWS 28
MAC OS DRIVER 29
MAC OS SYSTEM REQUIREMENTS 29
INSTALLING THE DRIVER ON MAC OS 29
V ERIFYING THE DRIVER V ERSION N UMBER ON MAC OS 29
L INUX DRIVER 31
L INUX SYSTEM REQUIREMENTS 31
INSTALLING THE DRIVER U SING THE RPM FILE 31
INSTALLING THE DRIVER ON DEBIAN 32
V ERIFYING THE DRIVER V ERSION N UMBER ON L INUX 33
AIX DRIVER 34
AIX SYSTEM REQUIREMENTS 34
INSTALLING THE DRIVER ON AIX 34
V ERIFYING THE DRIVER V ERSION N UMBER ON AIX 35
CONFIGURING THE ODBC DRIVER M ANAGER ON N ON-WINDOWS M ACHINES 36
SPECIFYING ODBC DRIVER MANAGERS ON N ON-WINDOWS MACHINES 36
SPECIFYING THE L OCATIONS OF THE DRIVER CONFIGURATION FILES 36
CONFIGURING ODBC CONNECTIONS ON A N ON-WINDOWS M ACHINE 38
CREATING A DATA SOURCE N AME ON A N ON-WINDOWS MACHINE 38
CONFIGURING A DSN-LESS CONNECTION ON A N ON-WINDOWS MACHINE 41
CONFIGURING SERVICE DISCOVERY MODE ON A N ON-WINDOWS MACHINE 43

Cloudera ODBC Driver for Apache Hive | 3

 CONFIGURING A UTHENTICATION ON A N ON-WINDOWS MACHINE 43
CONFIGURING SSL V ERIFICATION ON A N ON-WINDOWS MACHINE 46
CONFIGURING SERVER -SIDE PROPERTIES ON A N ON-WINDOWS MACHINE 47
CONFIGURING L OGGING O PTIONS 48
SETTING DRIVER -WIDE CONFIGURATION O PTIONS ON A N ON-WINDOWS MACHINE 50
TESTING THE CONNECTION 50
AUTHENTICATION M ECHANISMS 52
U SING A CONNECTION STRING 54
DSN CONNECTION STRING E XAMPLE 54
DSN-LESS CONNECTION STRING E XAMPLES 54
FEATURES 57
SQL CONNECTOR FOR HIVEQL 57
DATA TYPES 57
CATALOG AND SCHEMA SUPPORT 59
HIVE_ SYSTEM TABLE 59
SERVER -SIDE PROPERTIES 59
TEMPORARY TABLES 60
GET TABLES WITH QUERY 61
A CTIVE DIRECTORY 61
WRITE-BACK 61
TIMESTAMP FUNCTION SUPPORT 62
DYNAMIC SERVICE DISCOVERY USING ZOOKEEPER 62
SECURITY AND A UTHENTICATION 62
DRIVER CONFIGURATION OPTIONS 64
CONFIGURATION O PTIONS A PPEARING IN THE U SER INTERFACE 64
CONFIGURATION O PTIONS HAVING O NLY KEY N AMES 86
ODBC API CONFORMANCE L EVEL 89
CONTACT U S 91

4 | Cloudera ODBC Driver for Apache Hive

 About the Cloudera ODBC Driver for Apache Hive

About the Cloudera ODBC Driver for Apache Hive

The Cloudera ODBC Driver for Apache Hive is used for direct SQL and HiveQL access to Apache
Hadoop / Hive distributions, enabling Business Intelligence (BI), analytics, and reporting on
Hadoop / Hive-based data. The driver efficiently transforms an application’s SQL query into the
equivalent form in HiveQL, which is a subset of SQL-92. If an application is Hive-aware, then the
driver is configurable to pass the query through to the database for processing. The driver
interrogates Hive to obtain schema information to present to a SQL-based application. Queries,
including joins, are translated from SQL to HiveQL. For more information about the differences
between HiveQL and SQL, see "SQL Connector for HiveQL" on page 57.

The Cloudera ODBC Driver for Apache Hive complies with the ODBC 3.80 data standard and adds
important functionality such as Unicode and 32- and 64-bit support for high-performance
computing environments.

ODBC is one of the most established and widely supported APIs for connecting to and working
with databases. At the heart of the technology is the ODBC driver, which connects an application
to the database. For more information about ODBC, see Data Access Standards on the Simba
Technologies website: https://www.simba.com/resources/data-access-standards-glossary. For
complete information about the ODBC specification, see the ODBC API Reference from the
Microsoft documentation: https://docs.microsoft.com/en-us/sql/odbc/reference/syntax/odbc-
api-reference.

The Installation and Configuration Guide is suitable for users who are looking to access data
residing within Hive from their desktop environment. Application developers might also find the
information helpful. Refer to your application for details on connecting via ODBC.

Cloudera ODBC Driver for Apache Hive | 5

Windows Driver

Windows Driver
Windows System Requirements
The Cloudera ODBC Driver for Apache Hive supports Apache Hive versions 0.11 through 3.1, and
CDH versions 5.0 through 6.2.

Install the driver on client machines where the application is installed. Before installing the driver,
make sure that you have the following:
l Administrator rights on your machine.
l A machine that meets the following system requirements:
l One of the following operating systems:
l Windows 10, 8.1, or 7 SP1
l Windows Server 2016, 2012, or 2008 R2 SP1
l 100 MB of available disk space
l Visual C++ Redistributable for Visual Studio 2013 installed (with the same bitness as
the driver that you are installing).
You can download the installation packages at https://www.microsoft.com/en-
ca/download/details.aspx?id=40784.

Installing the Driver on Windows

On 64-bit Windows operating systems, you can execute both 32- and 64-bit applications.
However, 64-bit applications must use 64-bit drivers, and 32-bit applications must use 32-bit
drivers. Make sure that you use a driver whose bitness matches the bitness of the client
application:
l ClouderaHiveODBC32.msi for 32-bit applications
l ClouderaHiveODBC64.msi for 64-bit applications

You can install both versions of the driver on the same machine.

To install the Cloudera ODBC Driver for Apache Hive on Windows:

1. Depending on the bitness of your client application, double-click to run
ClouderaHiveODBC32.msi or ClouderaHiveODBC64.msi.
2. Click Next.
3. Select the check box to accept the terms of the License Agreement if you agree, and then
click Next.
4. To change the installation location, click Change, then browse to the desired folder, and
then click OK. To accept the installation location, click Next.
5. Click Install.
6. When the installation completes, click Finish.

6 | Cloudera ODBC Driver for Apache Hive

 Windows Driver

Creating a Data Source Name on Windows

Typically, after installing the Cloudera ODBC Driver for Apache Hive, you need to create a Data
Source Name (DSN). A DSN is a data structure that stores connection information so that it can be
used by the driver to connect to Hive.

Alternatively, you can specify connection settings in a connection string or as driver-wide settings.
Settings in the connection string take precedence over settings in the DSN, and settings in the DSN
take precedence over driver-wide settings.

The following instructions describe how to create a DSN. For information about specifying settings
in a connection string, see "Using a Connection String" on page 54. For information about driver-
wide settings, see "Configuring a DSN-less Connection on Windows" on page 10.

To create a Data Source Name on Windows:

1. From the Start menu, go to ODBC Data Sources.

Note:

Make sure to select the ODBC Data Source Administrator that has the same bitness as the
client application that you are using to connect to Hive.

2. In the ODBC Data Source Administrator, click the Drivers tab, and then scroll down as
needed to confirm that the Cloudera ODBC Driver for Apache Hive appears in the
alphabetical list of ODBC drivers that are installed on your system.
3. Choose one:
l To create a DSN that only the user currently logged into Windows can use, click the
User DSN tab.
l Or, to create a DSN that all users who log into Windows can use, click the System
DSN tab.

Note:

It is recommended that you create a System DSN instead of a User DSN. Some
applications load the data using a different user account, and might not be able to detect
User DSNs that are created under another user account.

4. Click Add.
5. In the Create New Data Source dialog box, select Cloudera ODBC Driver for Apache Hive
and then click Finish. The Cloudera ODBC Driver for Apache Hive DSN Setup dialog box
opens.
6. In the Data Source Name field, type a name for your DSN.
7. Optionally, in the Description field, type relevant details about the DSN.

Cloudera ODBC Driver for Apache Hive | 7

Windows Driver

8. In the Hive Server Type drop-down list, select Hive Server 1 or Hive Server 2.

Note:

If you are connecting through Apache ZooKeeper, then Hive Server 1 is not supported.

9. Specify whether the driver uses the ZooKeeper service when connecting to Hive, and
provide the necessary connection information:
l To connect to Hive without using the Apache ZooKeeper service, do the following:
a. From the Service Discovery Mode drop-down list, select No Service
Discovery.
b. In the Host(s) field, type the IP address or host name of the Hive server.
c. In the Port field, type the number of the TCP port that the Hive server uses to
listen for client connections.
l Or, to discover Hive Server 2 services via the ZooKeeper service, do the following:
a. From the Service Discovery Mode drop-down list, select ZooKeeper.
b. In the Host(s) field, type a comma-separated list of ZooKeeper servers. Use the
following format, where [ZK_Host] is the IP address or host name of the
ZooKeeper server and [ZK_Port] is the number of the TCP port that the
ZooKeeper server uses to listen for client connections:
[ZK_Host1]:[ZK_Port1],[ZK_Host2]:[ZK_Port2]

c. In the ZooKeeper Namespace field, type the namespace on ZooKeeper under

which Hive Server 2 znodes are added.
10. In the Database field, type the name of the database schema to use when a schema is not
explicitly specified in a query.

Note:

You can still issue queries on other schemas by explicitly specifying the schema in the
query. To inspect your databases and determine the appropriate schema to use, type the
show databases command at the Hive command prompt.

11. In the Authentication area, configure authentication as needed. For more information, see
"Configuring Authentication on Windows" on page 12.

Note:

Hive Server 1 does not support authentication. Most default configurations of Hive Server
2 require User Name authentication. To verify the authentication mechanism that you
need to use for your connection, check the configuration of your Hadoop / Hive
distribution. For more information, see "Authentication Mechanisms" on page 52.

12. Optionally, if the operations against Hive are to be done on behalf of a user that is different
than the authenticated user for the connection, type the name of the user to be delegated

8 | Cloudera ODBC Driver for Apache Hive

 Windows Driver

in the Delegation UID field. For more information, see "Delegating Authentication to a
Specific User" on page 16.

Note:

This option is applicable only when connecting to a Hive Server 2 instance that supports
this feature.

13. In the Thrift Transport drop-down list, select the transport protocol to use in the Thrift
layer.

Note:

For information about how to determine which Thrift transport protocols your Hive
server supports, see "Authentication Mechanisms" on page 52.

14. If the Thrift Transport option is set to HTTP, then to configure HTTP options such as custom
headers, click HTTP Options. For more information, see "Configuring HTTP Options on
Windows" on page 18.
15. To configure client-server verification over SSL, click SSL Options. For more information, see
"Configuring SSL Verification on Windows" on page 19.

Note:

If you selected User Name as the authentication mechanism, SSL is not available.

16. To configure advanced driver options, click Advanced Options. For more information, see
"Configuring Advanced Options on Windows" on page 16.
17. To configure server-side properties, click Advanced Options and then click Server Side
Properties. For more information, see "Configuring Server-Side Properties on Windows" on
page 21.
18. To configure the Temporary Table feature, click Advanced Options and then click
Temporary Table Configuration. For more information, see "Configuring the Temporary
Table Feature" on page 20 and "Temporary Tables" on page 60.

Important:

When connecting to Hive 0.14 or later, the Temporary Tables feature is always enabled
and you do not need to configure it in the driver.

19. To configure logging behavior for the driver, click Logging Options. For more information,
see "Configuring Logging Options on Windows" on page 22.
20. To test the connection, click Test. Review the results as needed, and then click OK.

Cloudera ODBC Driver for Apache Hive | 9

Windows Driver

Note:

If the connection fails, then confirm that the settings in the Cloudera ODBC Driver for
Apache Hive DSN Setup dialog box are correct. Contact your Hive server administrator as
needed.

21. To save your settings and close the Cloudera ODBC Driver for Apache Hive DSN Setup dialog
box, click OK.
22. To close the ODBC Data Source Administrator, click OK.

Configuring a DSN-less Connection on Windows

Some client applications provide support for connecting to a data source using a driver without a
Data Source Name (DSN). To configure a DSN-less connection, you can use a connection string or
the Cloudera Hive ODBC Driver Configuration tool that is installed with the Cloudera ODBC Driver
for Apache Hive. Settings in a connection string apply only when you connect to Hive using that
particular string, while settings in the driver configuration tool apply to every connection that uses
the Cloudera ODBC Driver for Apache Hive.

The following section explains how to use the driver configuration tool. For information about
using connection strings, see "Using a Connection String" on page 54.

Note:

l Settings in the connection string take precedence over settings in the DSN, and settings in
the DSN take precedence over driver-wide settings.
l The drop-down lists in the driver configuration tool only display one option at a time. Use
the scroll arrows on the right side of the drop-down list to view and select other options.

To configure a DSN-less connection using the driver configuration tool:

1. Choose one:
l If you are using Windows 7 or earlier, click Start > All Programs > Cloudera ODBC
Driver for Apache Hive 2.6 > Driver Configuration.
l Or, if you are using Windows 8 or later, click the arrow button at the bottom of the
Start screen, and then click Cloudera ODBC Driver for Apache Hive 2.6 > Driver
Configuration.

Note:

Make sure to select the Driver Configuration Tool that has the same bitness as the client
application that you are using to connect to Hive.

2. If you are prompted for administrator permission to make modifications to the machine,
click OK.

10 | Cloudera ODBC Driver for Apache Hive

 Windows Driver

Note:

You must have administrator access to the machine to run this application because it
makes changes to the registry.

3. In the Hive Server Type drop-down list, select Hive Server 1 or Hive Server 2.

Note:

If you are connecting through Apache ZooKeeper, then Hive Server 1 is not supported.

4. Specify whether the driver uses the ZooKeeper service when connecting to Hive:
l To connect to Hive without using the Apache ZooKeeper service, from the Service
Discovery Mode drop-down list, select No Service Discovery.
l Or, to discover Hive Server 2 services via the ZooKeeper service, do the following:
a. From the Service Discovery Mode drop-down list, select ZooKeeper.
b. In the Host(s) field, type a comma-separated list of ZooKeeper servers. Use the
following format, where [ZK_Host] is the IP address or host name of the
ZooKeeper server and [ZK_Port] is the number of the TCP port that the
ZooKeeper server uses to listen for client connections:
[ZK_Host1]:[ZK_Port1],[ZK_Host2]:[ZK_Port2]

c. In the ZooKeeper Namespace field, type the namespace on ZooKeeper under

which Hive Server 2 znodes are added.
5. In the Authentication area, configure authentication as needed. For more information, see
"Configuring Authentication on Windows" on page 12.

Note:

Hive Server 1 does not support authentication. Most default configurations of Hive Server
2 require User Name authentication. To verify the authentication mechanism that you
need to use for your connection, check the configuration of your Hadoop / Hive
distribution. For more information, see "Authentication Mechanisms" on page 52.

6. Optionally, if the operations against Hive are to be done on behalf of a user that is different
than the authenticated user for the connection, then in the Delegation UID field, type the
name of the user to be delegated. For more information, see "Delegating Authentication to
a Specific User" on page 16.

Note:

This option is applicable only when connecting to a Hive Server 2 instance that supports
this feature.

7. In the Thrift Transport drop-down list, select the transport protocol to use in the Thrift
layer.

Cloudera ODBC Driver for Apache Hive | 11

Windows Driver

Note:

For information about how to determine which Thrift transport protocols your Hive
server supports, see "Authentication Mechanisms" on page 52.

8. If the Thrift Transport option is set to HTTP, then to configure HTTP options such as custom
headers, click HTTP Options. For more information, see "Configuring HTTP Options on
Windows" on page 18.
9. To configure client-server verification over SSL, click SSL Options. For more information, see
"Configuring SSL Verification on Windows" on page 19.

Note:

If you selected User Name as the authentication mechanism, SSL is not available.

10. To configure advanced options, click Advanced Options. For more information, see
"Configuring Advanced Options on Windows" on page 16.
11. To configure server-side properties, click Advanced Options and then click Server Side
Properties. For more information, see "Configuring Server-Side Properties on Windows" on
page 21.
12. To configure the Temporary Table feature, click Advanced Options and then click
Temporary Table Configuration. For more information, see "Temporary Tables" on page 60
and "Configuring the Temporary Table Feature" on page 20.

Important:

When connecting to Hive 0.14 or later, the Temporary Tables feature is always enabled
and you do not need to configure it in the driver.

13. To save your settings and close the Cloudera Hive ODBC Driver Configuration tool, click OK.

Configuring Authentication on Windows

Some Hive Server 2 instances are configured to require authentication for access. To connect to a
Hive server, you must configure the Cloudera ODBC Driver for Apache Hive to use the
authentication mechanism that matches the access requirements of the server and provides the
necessary credentials.

For information about how to determine the type of authentication your Hive server requires, see
"Authentication Mechanisms" on page 52.

You can specify authentication settings in a DSN, in a connection string, or as driver-wide settings.
Settings in the connection string take precedence over settings in the DSN, and settings in the DSN
take precedence over driver-wide settings.

If cookie-based authentication is enabled in your Hive Server 2 database, you can specify a list of
authentication cookies in the HTTPAuthCookies connection property. In this case, the driver
authenticates the connection once based on the provided authentication credentials. It then uses

12 | Cloudera ODBC Driver for Apache Hive

 Windows Driver

the cookie generated by the server for each subsequent request in the same connection. For
more information, see "HTTPAuthCookies" on page 87.

Note:

On Windows, the HTTPAuthCookies property must be set in a connection string.

Using No Authentication

When connecting to a Hive server of type Hive Server 1, you must use No Authentication. When
you use No Authentication, Binary is the only Thrift transport protocol that is supported.

To configure a connection without authentication:

1. Choose one:
l To access authentication options for a DSN, open the ODBC Data Source
Administrator where you created the DSN, then select the DSN, and then click
Configure.
l Or, to access authentication options for a DSN-less connection, open the Cloudera
Hive ODBC Driver Configuration tool.
2. From the Mechanism drop-down list, select No Authentication.
3. If the Hive server is configured to use SSL, then click SSL Options to configure SSL for the
connection. For more information, see "Configuring SSL Verification on Windows" on page
19.
4. To save your settings and close the dialog box, click OK.

Using Kerberos

This authentication mechanism is available only for Hive Server 2. When you use Kerberos
authentication, the Binary transport protocol is not supported.

If the Use Only SSPI advanced option is disabled, then Kerberos must be installed and configured
before you can use this authentication mechanism. For information about configuring Kerberos
on your machine, see "Configuring Kerberos Authentication for Windows" on page 24. For
information about setting the Use Only SSPI advanced option, see "Configuring Advanced Options
on Windows" on page 16.

To configure Kerberos authentication:

1. Choose one:
l To access authentication options for a DSN, open the ODBC Data Source
Administrator where you created the DSN, then select the DSN, and then click
Configure.
l Or, to access authentication options for a DSN-less connection, open the Cloudera
Hive ODBC Driver Configuration tool.
2. From the Mechanism drop-down list, select Kerberos.
3. Choose one:

Cloudera ODBC Driver for Apache Hive | 13

Windows Driver

l To use the default realm defined in your Kerberos setup, leave the Realm field empty.
l Or, if your Kerberos setup does not define a default realm or if the realm of your Hive
Server 2 host is not the default, then, in the Realm field, type the Kerberos realm of
the Hive Server 2.
4. In the Host FQDN field, type the fully qualified domain name of the Hive Server 2 host.

Note:

To use the Hive server host name as the fully qualified domain name for Kerberos
authentication, in the Host FQDN field, type _HOST.

5. In the Service Name field, type the service name of the Hive server.
6. Optionally, if you are using MIT Kerberos and a Kerberos realm is specified in the Realm
field, then choose one:
l To have the Kerberos layer canonicalize the server's service principal name, leave the
Canonicalize Principal FQDN check box selected.
l Or, to prevent the Kerberos layer from canonicalizing the server's service principal
name, clear the Canonicalize Principal FQDN check box.
7. To allow the driver to pass your credentials directly to the server for use in authentication,
select Delegate Kerberos Credentials.
8. From the Thrift Transport drop-down list, select the transport protocol to use in the Thrift
layer.

Important:

When using this authentication mechanism, the Binary transport protocol is not
supported.

9. If the Hive server is configured to use SSL, then click SSL Options to configure SSL for the
connection. For more information, see "Configuring SSL Verification on Windows" on page
19.
10. To save your settings and close the dialog box, click OK.

Using User Name

This authentication mechanism requires a user name but not a password. The user name labels
the session, facilitating database tracking.

This authentication mechanism is available only for Hive Server 2. Most default configurations of
Hive Server 2 require User Name authentication. When you use User Name authentication, SSL is
not supported and SASL is the only Thrift transport protocol available.

14 | Cloudera ODBC Driver for Apache Hive

 Windows Driver

To configure User Name authentication:

1. Choose one:
l To access authentication options for a DSN, open the ODBC Data Source
Administrator where you created the DSN, then select the DSN, and then click
Configure.
l Or, to access authentication options for a DSN-less connection, open the Cloudera
Hive ODBC Driver Configuration tool.
2. From the Mechanism drop-down list, select User Name.
3. In the User Name field, type an appropriate user name for accessing the Hive server.
4. To save your settings and close the dialog box, click OK.

Using User Name And Password

This authentication mechanism requires a user name and a password.

This authentication mechanism is available only for Hive Server 2.

To configure User Name And Password authentication:

1. Choose one:
l To access authentication options for a DSN, open the ODBC Data Source
Administrator where you created the DSN, then select the DSN, and then click
Configure.
l Or, to access authentication options for a DSN-less connection, open the Cloudera
Hive ODBC Driver Configuration tool.
2. From the Mechanism drop-down list, select User Name And Password.
3. In the User Name field, type an appropriate user name for accessing the Hive server.
4. In the Password field, type the password corresponding to the user name you typed above.
5. To save the password, select the Save Password (Encrypted) check box.

Important:

The password is obscured, that is, not saved in plain text. However, it is still possible for
the encrypted password to be copied and used.

6. From the Thrift Transport drop-down list, select the transport protocol to use in the Thrift
layer.
7. If the Hive server is configured to use SSL, then click SSL Options to configure SSL for the
connection. For more information, see "Configuring SSL Verification on Windows" on page
19.
8. To save your settings and close the dialog box, click OK.

Cloudera ODBC Driver for Apache Hive | 15

Windows Driver

Delegating Authentication to a Specific User

Some Hive Server 2 instances support the ability to delegate all operations against Hive to the
specified user, rather than to the authenticated user for the connection.

To delegate all operations to a specified user:

1. Choose one:
l To access authentication options for a DSN, open the ODBC Data Source
Administrator where you created the DSN, then select the DSN, and then click
Configure.
l Or, to access authentication options for a DSN-less connection, open the Cloudera
Hive ODBC Driver Configuration tool.
2. In the Delegation UID field, type the name of the user to be delegated.
3. To save your settings and close the dialog box, click OK.

If the server returns an error message such as Failed to validate proxy privilege of [RealUser]
for [DelegationUID], you may need to modify the server's core-site.xml configuration file,
as follows:
1. In the server's core-site.xml configuration file, add the following properties:
hadoop.proxyuser.[RealUser].groups=*
hadoop.proxyuser.[RealUser].hosts=*

Where [RealUser] is the authenticated user for the connection.

2. If you are using Kerberos authentication, then in the server's core-site.xml
configuration file, add the following properties:
hadoop.proxyuser.[Principal].groups=*
hadoop.proxyuser.[Principal].hosts=*

Where [Principal] is the primary Kerberos principal user. For example, if the primary
Kerberos principal user is [email protected], replace [Principal] with kerbuser.

For more information on resolving this error, see your Hive Server documentation.

Configuring Advanced Options on Windows

You can configure advanced options to modify the behavior of the driver.

The following instructions describe how to configure advanced options in a DSN and in the driver
configuration tool. You can specify the connection settings described below in a DSN, in a
connection string, or as driver-wide settings. Settings in the connection string take precedence
over settings in the DSN, and settings in the DSN take precedence over driver-wide settings.

16 | Cloudera ODBC Driver for Apache Hive

 Windows Driver

To configure advanced options on Windows:

1. Choose one:
l To access advanced options for a DSN, open the ODBC Data Source Administrator
where you created the DSN, then select the DSN, then click Configure, and then click
Advanced Options.
l Or, to access advanced options for a DSN-less connection, open the Cloudera Hive
ODBC Driver Configuration tool, and then click Advanced Options.
2. To disable the SQL Connector feature, select the Use Native Query check box.

Important:

l When this option is enabled, the driver cannot execute parameterized queries.
l By default, the driver applies transformations to the queries emitted by an
application to convert the queries into an equivalent form in HiveQL. If the
application is Hive-aware and already emits HiveQL, then turning off the translation
avoids the additional overhead of query transformation.

3. To defer query execution to SQLExecute, select the Fast SQLPrepare check box.
4. To allow driver-wide configurations to take precedence over connection and DSN settings,
select the Driver Config Take Precedence check box.
5. To use the asynchronous version of the API call against Hive for executing a query, select
the Use Async Exec check box.

Note:

This option is applicable only when connecting to a Hive cluster running Hive 0.12.0 or
later.

6. To retrieve table names from the database by using the SHOW TABLES query, select the Get
Tables With Query check box.

Note:

This option is applicable only when connecting to Hive Server 2.

7. To enable the driver to return SQL_WVARCHAR instead of SQL_VARCHAR for STRING and
VARCHAR columns, and SQL_WCHAR instead of SQL_CHAR for CHAR columns, select the
Unicode SQL Character Types check box.
8. To enable the driver to return the hive_system table for catalog function calls such as
SQLTables and SQLColumns, select the Show System Table check box.
9. To specify which mechanism the driver uses by default to handle Kerberos authentication,
do one of the following:
l To use the SSPI plugin by default, select the Use Only SSPI check box.
l To use MIT Kerberos by default and only use the SSPI plugin if the GSSAPI library is
not available, clear the Use Only SSPI check box.

Cloudera ODBC Driver for Apache Hive | 17

Windows Driver

10. To enable the driver to automatically open a new session when the existing session is no
longer valid, select the Invalid Session Auto Recover check box.

Note:

This option is applicable only when connecting to Hive Server 2.

11. To have the driver automatically attempt to reconnect to the server if communications are
lost, select Enable Auto Reconnect.
12. In the Rows Fetched Per Block field, type the number of rows to be fetched per block.
13. In the Default String Column Length field, type the maximum data length for STRING
columns.
14. In the Binary Column Length field, type the maximum data length for BINARY columns.
15. In the Decimal Column Scale field, type the maximum number of digits to the right of the
decimal point for numeric data types.
16. In the Socket Timeout field, type the number of seconds that an operation can remain idle
before it is closed.

Note:

This option is applicable only when asynchronous query execution is being used against
Hive Server 2 instances.

17. To save your settings and close the Advanced Options dialog box, click OK.

Configuring HTTP Options on Windows

You can configure options such as custom headers when using the HTTP transport protocol in the
Thrift layer. For information about how to determine if your Hive server supports the
HTTP transport protocol, see "Authentication Mechanisms" on page 52.

The following instructions describe how to configure HTTP options in a DSN and in the driver
configuration tool. You can specify the connection settings described below in a DSN, in a
connection string, or as driver-wide settings. Settings in the connection string take precedence
over settings in the DSN, and settings in the DSN take precedence over driver-wide settings.

To configure HTTP options on Windows:

1. Choose one:
l If you are configuring HTTP for a DSN, open the ODBC Data Source Administrator
where you created the DSN, then select the DSN, then click Configure, and then
make sure that the Thrift Transport option is set to HTTP.
l Or, if you are configuring HTTP for a DSN-less connection, open the Cloudera Hive
ODBC Driver Configuration tool and then make sure that the Thrift Transport option
is set to HTTP.
2. To access HTTP options, click HTTP Options.

18 | Cloudera ODBC Driver for Apache Hive

 Windows Driver

Note:

The HTTP options are available only when the Thrift Transport option is set to HTTP.

3. In the HTTP Path field, type the partial URL corresponding to the Hive server.
4. To create a custom HTTP header, click Add, then type appropriate values in the Key and
Value fields, and then click OK.
5. To edit a custom HTTP header, select the header from the list, then click Edit, then update
the Key and Value fields as needed, and then click OK.
6. To delete a custom HTTP header, select the header from the list, and then click Remove. In
the confirmation dialog box, click Yes.
7. To save your settings and close the HTTP Properties dialog box, click OK.

Configuring SSL Verification on Windows

If you are connecting to a Hive server that has Secure Sockets Layer (SSL) enabled, you can
configure the driver to connect to an SSL-enabled socket. When using SSL to connect to a server,
the driver supports identity verification between the client (the driver itself) and the server.

The following instructions describe how to configure SSL in a DSN and in the driver configuration
tool. You can specify the connection settings described below in a DSN, in a connection string, or
as driver-wide settings. Settings in the connection string take precedence over settings in the DSN,
and settings in the DSN take precedence over driver-wide settings.

Note:

If you selected User Name as the authentication mechanism, SSL is not available.

To configure SSL verification on Windows:

1. Choose one:
l To access SSL options for a DSN, open the ODBC Data Source Administrator where
you created the DSN, then select the DSN, then click Configure, and then click SSL
Options.
l Or, to access advanced options for a DSN-less connection, open the Cloudera Hive
ODBC Driver Configuration tool, and then click SSL Options.
2. Select the Enable SSL check box.
3. To allow authentication using self-signed certificates that have not been added to the list of
trusted certificates, select the Allow Self-signed Server Certificate check box.
4. To allow the common name of a CA-issued SSL certificate to not match the host name of the
Hive server, select the Allow Common Name Host Name Mismatch check box.
5. To specify the CA certificates that you want to use to verify the server, do one of the
following:
l To verify the server using the trusted CA certificates from a specific .pem file, specify
the full path to the file in the Trusted Certificates field and clear the Use System

Cloudera ODBC Driver for Apache Hive | 19

Windows Driver

Trust Store check box.

l Or, to use the trusted CA certificates .pem file that is installed with the driver, leave
the Trusted Certificates field empty, and clear the Use System Trust Store check box.
l Or, to use the Windows trust store, select the Use System Trust Store check box.

Important:

l If you are using the Windows trust store, make sure to import the trusted CA
certificates into the trust store.
l If the trusted CA supports certificate revocation, select the Check Certificate
Revocation check box.

6. From the Minimum TLS Version drop-down list, select the minimum version of TLS to use
when connecting to your data store.
7. To configure two-way SSL verification, select the Two-Way SSL check box and then do the
following:
a. In the Client Certificate File field, specify the full path of the PEM file containing the
client's certificate.
b. In the Client Private Key File field, specify the full path of the file containing the
client's private key.
c. If the private key file is protected with a password, type the password in the Client
Private Key Password field. To save the password, select the Save Password
(Encrypted) check box.

Important:

The password is obscured, that is, not saved in plain text. However, it is still
possible for the encrypted password to be copied and used.

8. To save your settings and close the SSL Options dialog box, click OK.

Configuring the Temporary Table Feature

You can configure the driver to create temporary tables. For more information about this feature,
including details about the statement syntax used for temporary tables, see "Temporary Tables"
on page 60.

Important:

When connecting to Hive 0.14 or later, the Temporary Tables feature is always enabled and you
do not need to configure it in the driver.

To configure the Temporary Table feature:

20 | Cloudera ODBC Driver for Apache Hive

 Windows Driver

1. Choose one:
l To configure the temporary table feature for a DSN, open the ODBC Data Source
Administrator where you created the DSN, then select the DSN and click Configure,
then click Advanced Options, and then click Temporary Table Configuration.
l Or, to configure the temporary table feature for a DSN-less connection, open the
Cloudera Hive ODBC Driver Configuration tool, then click Advanced Options, and
then click Temporary Table Configuration.
2. To enable the Temporary Table feature, select the Enable Temporary Table check box.
3. In the Web HDFS Host field, type the host name or IP address of the machine hosting both
the namenode of your Hadoop cluster and the WebHDFS service. If this field is left blank,
then the host name of the Hive server is used.
4. In the Web HDFS Port field, type the WebHDFS port for the namenode.
5. In the HDFS User field, type the name of the HDFS user that the driver uses to create the
necessary files for supporting the Temporary Table feature.
6. In the Data File HDFS Dir field, type the HDFS directory that the driver uses to store the
necessary files for supporting the Temporary Table feature.

Note:

Due to a known issue in Hive (see https://issues.apache.org/jira/browse/HIVE-4554),

HDFS paths with space characters do not work with versions of Hive prior to 0.12.0.

7. In the Temp Table TTL field, type the number of minutes that a temporary table is
guaranteed to exist in Hive after it is created.
8. To save your settings and close the Temporary Table Configuration dialog box, click OK.

Configuring Server-Side Properties on Windows

You can use the driver to apply configuration properties to the Hive server.

The following instructions describe how to configure server-side properties in a DSN and in the
driver configuration tool. You can specify the connection settings described below in a DSN, in a
connection string, or as driver-wide settings. Settings in the connection string take precedence
over settings in the DSN, and settings in the DSN take precedence over driver-wide settings.

To configure server-side properties on Windows:

1. Choose one:
l To configure server-side properties for a DSN, open the ODBC Data Source
Administrator where you created the DSN, then select the DSN, then click Configure,
then click Advanced Options, and then click Server Side Properties.
l Or, to configure server-side properties for a DSN-less connection, open the Cloudera
Hive ODBC Driver Configuration tool, then click Advanced Options, and then click
Server Side Properties.

Cloudera ODBC Driver for Apache Hive | 21

Windows Driver

2. To create a server-side property, click Add, then type appropriate values in the Key and
Value fields, and then click OK. For example, to set the value of the
mapreduce.job.queuename property to myQueue, type mapreduce.job.queuename
in the Key field and then type myQueue in the Value field.

Note:

For a list of all Hadoop and Hive server-side properties that your implementation
supports, type set -v at the Hive CLI command line or Beeline. You can also execute the
set -v query after connecting using the driver.

3. To edit a server-side property, select the property from the list, then click Edit, then update
the Key and Value fields as needed, and then click OK.
4. To delete a server-side property, select the property from the list, and then click Remove. In
the confirmation dialog box, click Yes.
5. To configure the driver to convert server-side property key names to all lower-case
characters, select the Convert Key Name To Lower Case check box.
6. To change the method that the driver uses to apply server-side properties, do one of the
following:
l To configure the driver to apply each server-side property by executing a query when
opening a session to the Hive server, select the Apply Server Side Properties With
Queries check box.
l Or, to configure the driver to use a more efficient method for applying server-side
properties that does not involve additional network round-tripping, clear the Apply
Server Side Properties With Queries check box.

Note:

The more efficient method is not available for Hive Server 1, and it might not be
compatible with some Hive Server 2 builds. If the server-side properties do not take effect
when the check box is clear, then select the check box.

7. To save your settings and close the Server Side Properties dialog box, click OK.

Configuring Logging Options on Windows

To help troubleshoot issues, you can enable logging. In addition to functionality provided in the
Cloudera ODBC Driver for Apache Hive, the ODBC Data Source Administrator provides tracing
functionality.

Important:

Only enable logging or tracing long enough to capture an issue. Logging or tracing decreases
performance and can consume a large quantity of disk space.

The settings for logging apply to every connection that uses the Cloudera ODBC Driver for
Apache Hive, so make sure to disable the feature after you are done using it.

22 | Cloudera ODBC Driver for Apache Hive

 Windows Driver

To enable driver logging on Windows:

1. To access logging options, open the ODBC Data Source Administrator where you created
the DSN, then select the DSN, then click Configure, and then click Logging Options.
2. From the Log Level drop-down list, select the logging level corresponding to the amount of
information that you want to include in log files:
Logging Level Description

OFF Disables all logging.

FATAL Logs severe error events that lead the driver to abort.

ERROR Logs error events that might allow the driver to continue
running.

WARNING Logs events that might result in an error if action is not taken.

INFO Logs general information that describes the progress of the

driver.

DEBUG Logs detailed information that is useful for debugging the driver.

TRACE Logs all driver activity.

3. In the Log Path field, specify the full path to the folder where you want to save log files.
4. If requested by Technical Support, type the name of the component for which to log
messages in the Log Namespace field. Otherwise, do not type a value in the field.
5. In the Max Number Files field, type the maximum number of log files to keep.

Note:

After the maximum number of log files is reached, each time an additional file is created,
the driver deletes the oldest log file.

6. In the Max File Size field, type the maximum size of each log file in megabytes (MB).

Note:

After the maximum file size is reached, the driver creates a new file and continues logging.

7. Click OK.
8. Restart your ODBC application to make sure that the new settings take effect.

The Cloudera ODBC Driver for Apache Hive produces the following log files at the location you
specify in the Log Path field:

Cloudera ODBC Driver for Apache Hive | 23

Windows Driver

l A clouderahiveodbcdriver.log file that logs driver activity that is not specific to a

connection.
l A clouderahiveodbcdriver_connection_[Number].log file for each
connection made to the database, where [Number] is a number that identifies each log file.
This file logs driver activity that is specific to the connection.

To disable driver logging on Windows:

1. Open the ODBC Data Source Administrator where you created the DSN, then select the
DSN, then click Configure, and then click Logging Options.
2. From the Log Level drop-down list, select LOG_OFF.
3. Click OK.
4. Restart your ODBC application to make sure that the new settings take effect.

Configuring Kerberos Authentication for Windows

Active Directory

The Cloudera ODBC Driver for Apache Hive supports Active Directory Kerberos on Windows. There
are two prerequisites for using Active Directory Kerberos on Windows:
l MIT Kerberos is not installed on the client Windows machine.
l The MIT Kerberos Hadoop realm has been configured to trust the Active Directory realm so
that users in the Active Directory realm can access services in the MIT Kerberos Hadoop
realm.

MIT Kerberos

Downloading and Installing MIT Kerberos for Windows 4.0.1

For information about Kerberos and download links for the installer, see the MIT Kerberos
website: http://web.mit.edu/kerberos/.

To download and install MIT Kerberos for Windows 4.0.1:

1. Download the appropriate Kerberos installer:
l For a 64-bit machine, use the following download link from the MIT Kerberos
website: http://web.mit.edu/kerberos/dist/kfw/4.0/kfw-4.0.1-amd64.msi.
l For a 32-bit machine, use the following download link from the MIT Kerberos
website: http://web.mit.edu/kerberos/dist/kfw/4.0/kfw-4.0.1-i386.msi.

Note:

The 64-bit installer includes both 32-bit and 64-bit libraries. The 32-bit installer includes 32-
bit libraries only.

2. To run the installer, double-click the .msi file that you downloaded above.

24 | Cloudera ODBC Driver for Apache Hive

 Windows Driver

3. Follow the instructions in the installer to complete the installation process.

4. When the installation completes, click Finish.

Setting Up the Kerberos Configuration File

Settings for Kerberos are specified through a configuration file. You can set up the configuration
file as an .ini file in the default location, which is the C:\ProgramData\MIT\Kerberos5
directory, or as a .conf file in a custom location.

Normally, the C:\ProgramData\MIT\Kerberos5 directory is hidden. For information about

viewing and using this hidden directory, refer to Microsoft Windows documentation.

Note:

For more information on configuring Kerberos, refer to the MIT Kerberos documentation.

To set up the Kerberos configuration file in the default location:

1. Obtain a krb5.conf configuration file. You can obtain this file from your Kerberos
administrator, or from the /etc/krb5.conf folder on the machine that is hosting the
Hive Server 2 instance.
2. Rename the configuration file from krb5.conf to krb5.ini.
3. Copy the krb5.ini file to the C:\ProgramData\MIT\Kerberos5 directory and
overwrite the empty sample file.

To set up the Kerberos configuration file in a custom location:

1. Obtain a krb5.conf configuration file. You can obtain this file from your Kerberos
administrator, or from the /etc/krb5.conf folder on the machine that is hosting the
Hive Server 2 instance.
2. Place the krb5.conf file in an accessible directory and make note of the full path name.
3. Open the System window:
l If you are using Windows 7 or earlier, click Start , then right-click Computer, and
then click Properties.
l Or, if you are using Windows 8 or later, right-click This PC on the Start screen, and
then click Properties.
4. Click Advanced System Settings.
5. In the System Properties dialog box, click the Advanced tab and then click Environment
Variables.
6. In the Environment Variables dialog box, under the System Variables list, click New.
7. In the New System Variable dialog box, in the Variable Name field, type KRB5_CONFIG.
8. In the Variable Value field, type the full path to the krb5.conf file.
9. Click OK to save the new variable.
10. Make sure that the variable is listed in the System Variables list.

Cloudera ODBC Driver for Apache Hive | 25

Windows Driver

11. Click OK to close the Environment Variables dialog box, and then click OK to close the
System Properties dialog box.

Setting Up the Kerberos Credential Cache File

Kerberos uses a credential cache to store and manage credentials.

To set up the Kerberos credential cache file:

1. Create a directory where you want to save the Kerberos credential cache file. For example,
create a directory named C:\temp.
2. Open the System window:
l If you are using Windows 7 or earlier, click Start , then right-click Computer, and
then click Properties.
l Or, if you are using Windows 8 or later, right-click This PC on the Start screen, and
then click Properties.
3. Click Advanced System Settings.
4. In the System Properties dialog box, click the Advanced tab and then click Environment
Variables.
5. In the Environment Variables dialog box, under the System Variables list, click New.
6. In the New System Variable dialog box, in the Variable Name field, type KRB5CCNAME.
7. In the Variable Value field, type the path to the folder you created above, and then append
the file name krb5cache. For example, if you created the folder C:\temp, then type
C:\temp\krb5cache.

Note:

krb5cache is a file (not a directory) that is managed by the Kerberos software, and it
should not be created by the user. If you receive a permission error when you first use
Kerberos, make sure that the krb5cache file does not already exist as a file or a
directory.

8. Click OK to save the new variable.

9. Make sure that the variable appears in the System Variables list.
10. Click OK to close the Environment Variables dialog box, and then click OK to close the
System Properties dialog box.
11. To make sure that Kerberos uses the new settings, restart your machine.

Obtaining a Ticket for a Kerberos Principal

A principal refers to a user or service that can authenticate to Kerberos. To authenticate to

Kerberos, a principal must obtain a ticket by using a password or a keytab file. You can specify a
keytab file to use, or use the default keytab file of your Kerberos configuration.

26 | Cloudera ODBC Driver for Apache Hive

 Windows Driver

To obtain a ticket for a Kerberos principal using a password:

1. Open MIT Kerberos Ticket Manager.
2. In MIT Kerberos Ticket Manager, click Get Ticket.
3. In the Get Ticket dialog box, type your principal name and password, and then click OK.

If the authentication succeeds, then your ticket information appears in MIT Kerberos Ticket
Manager.

To obtain a ticket for a Kerberos principal using a keytab file:

1. Open a command prompt:
l If you are using Windows 7 or earlier, click Start , then click All Programs, then click
Accessories, and then click Command Prompt.
l If you are using Windows 8 or later, click the arrow button at the bottom of the Start
screen, then find the Windows System program group, and then click Command
Prompt.
2. In the Command Prompt, type a command using the following syntax:
kinit -k -t [KeytabPath][Principal]

[KeytabPath] is the full path to the keytab file. For example:

C:\mykeytabs\myUser.keytab.

[Principal] is the Kerberos user principal to use for authentication. For example:
[email protected].
3. If the cache location KRB5CCNAME is not set or used, then use the -c option of the kinit
command to specify the location of the credential cache. In the command, the -c
argument must appear last. For example:
kinit -k -t C:\mykeytabs\myUser.keytab [email protected] -c
C:\ProgramData\MIT\krbcache

Krbcache is the Kerberos cache file, not a directory.

To obtain a ticket for a Kerberos principal using the default keytab file:

Note:

For information about configuring a default keytab file for your Kerberos configuration, refer to
the MIT Kerberos documentation.

1. Open a command prompt:

l If you are using Windows 7 or earlier, click Start , then click All Programs, then click
Accessories, and then click Command Prompt.
l If you are using Windows 8 or later, click the arrow button at the bottom of the Start
screen, then find the Windows System program group, and then click Command
Prompt.

Cloudera ODBC Driver for Apache Hive | 27

Windows Driver

2. In the Command Prompt, type a command using the following syntax:

kinit -k [principal]

[principal] is the Kerberos user principal to use for authentication. For example:
[email protected].
3. If the cache location KRB5CCNAME is not set or used, then use the -c option of the kinit
command to specify the location of the credential cache. In the command, the -c
argument must appear last. For example:
kinit -k -t C:\mykeytabs\myUser.keytab [email protected] -c
C:\ProgramData\MIT\krbcache

Krbcache is the Kerberos cache file, not a directory.

Verifying the Driver Version Number on Windows

If you need to verify the version of the Cloudera ODBC Driver for Apache Hive that is installed on
your Windows machine, you can find the version number in the ODBC Data Source Administrator.

To verify the driver version number on Windows:

1. From the Start menu, go to ODBC Data Sources.

Note:

Make sure to select the ODBC Data Source Administrator that has the same bitness as the
client application that you are using to connect to Hive.

2. Click the Drivers tab and then find the Cloudera ODBC Driver for Apache Hive in the list of
ODBC drivers that are installed on your system. The version number is displayed in the
Version column.

28 | Cloudera ODBC Driver for Apache Hive

 macOS Driver

macOS Driver
macOS System Requirements
The Cloudera ODBC Driver for Apache Hive supports Apache Hive versions 0.11 through 3.1, and
CDH versions 5.0 through 6.2.

Install the driver on client machines where the application is installed. Each client machine that
you install the driver on must meet the following minimum system requirements:
l macOS version 10.12, 10.13, or 10.14
l 100 MB of available disk space
l iODBC 3.52.9, 3.52.10, 3.52.11, or 3.52.12

Installing the Driver on macOS

The Cloudera ODBC Driver for Apache Hive is available for macOS as a .dmg file named
ClouderaHiveODBC.dmg. The driver supports both 32- and 64-bit client applications.

To install the Cloudera ODBC Driver for Apache Hive on macOS:

1. Double-click ClouderaHiveODBC.dmg to mount the disk image.
2. Double-click ClouderaHiveODBC.pkg to run the installer.
3. In the installer, click Continue.
4. On the Software License Agreement screen, click Continue, and when the prompt appears,
click Agree if you agree to the terms of the License Agreement.
5. Optionally, to change the installation location, click Change Install Location, then select the
desired location, and then click Continue.

Note:

By default, the driver files are installed in the /opt/cloudera/hiveodbc directory.

6. To accept the installation location and begin the installation, click Install.
7. When the installation completes, click Close.

Next, configure the environment variables on your machine to make sure that the ODBC driver
manager can work with the driver. For more information, see "Configuring the ODBC Driver
Manager on Non-Windows Machines" on page 36.

Verifying the Driver Version Number on macOS

If you need to verify the version of the Cloudera ODBC Driver for Apache Hive that is installed on
your macOS machine, you can query the version number through the Terminal.

Cloudera ODBC Driver for Apache Hive | 29

macOS Driver

To verify the driver version number on macOS:

At the Terminal, run the following command:
pkgutil --info cloudera.hiveodbc

The command returns information about the Cloudera ODBC Driver for Apache Hive that is
installed on your machine, including the version number.

30 | Cloudera ODBC Driver for Apache Hive

 Linux Driver

Linux Driver
For most Linux distributions, you can install the driver using the RPM file. If you are installing the
driver on a Debian machine, you must use the Debian package.

Linux System Requirements

The Cloudera ODBC Driver for Apache Hive supports Apache Hive versions 0.11 through 3.1, and
CDH versions 5.0 through 6.2.

Install the driver on client machines where the application is installed. Each client machine that
you install the driver on must meet the following minimum system requirements:
l One of the following distributions:
o Red Hat® Enterprise Linux® (RHEL) 6 or 7
o CentOS 6 or 7
o SUSE Linux Enterprise Server (SLES) 11 or 12
o Debian 6 or 7
o Ubuntu 14.04
o Oracle Linux 7.5 or 7.6
l 150 MB of available disk space
l One of the following ODBC driver managers installed:
o iODBC 3.52.9, 3.52.10, 3.52.11, or 3.52.12
o unixODBC 2.3.2, 2.3.3, or 2.3.4
l All of the following libsasl libraries installed:
o cyrus-sasl-2.1.22-7 or later
o cyrus-sasl-gssapi-2.1.22-7 or later
o cyrus-sasl-plain-2.1.22-7 or later

Note:

If the package manager in your Linux distribution cannot resolve the dependencies
automatically when installing the driver, then download and manually install the
packages.

To install the driver, you must have root access on the machine.

Installing the Driver Using the RPM File

On 64-bit editions of Linux, you can execute both 32- and 64-bit applications. However, 64-bit
applications must use 64-bit drivers, and 32-bit applications must use 32-bit drivers. Make sure
that you use a driver whose bitness matches the bitness of the client application:

Cloudera ODBC Driver for Apache Hive | 31

Linux Driver

l ClouderaHiveODBC-32bit-[Version]-[Release].i686.rpm for the 32-bit

driver
l ClouderaHiveODBC-[Version]-[Release].x86_64.rpm for the 64-bit driver

The placeholders in the file names are defined as follows:

l [Version] is the version number of the driver.
l [Release] is the release number for this version of the driver.

You can install both the 32-bit and 64-bit versions of the driver on the same machine.

To install the Cloudera ODBC Driver for Apache Hive using the RPM File:
1. Log in as the root user.
2. Navigate to the folder containing the RPM package for the driver.
3. Depending on the Linux distribution that you are using, run one of the following commands
from the command line, where [RPMFileName] is the file name of the RPM package:
l If you are using Red Hat Enterprise Linux or CentOS, run the following command:
yum --nogpgcheck localinstall [RPMFileName]

l Or, if you are using SUSE Linux Enterprise Server, run the following command:
zypper install [RPMFileName]

The Cloudera ODBC Driver for Apache Hive files are installed in the
/opt/cloudera/hiveodbc directory.

Note:

If the package manager in your Linux distribution cannot resolve the libsasl
dependencies automatically when installing the driver, then download and manually
install the packages.

Next, configure the environment variables on your machine to make sure that the ODBC driver
manager can work with the driver. For more information, see "Configuring the ODBC Driver
Manager on Non-Windows Machines" on page 36.

Installing the Driver on Debian

To install the driver on a Debian machine, use the Debian package instead of the RPM file or
tarball package.

On 64-bit editions of Debian, you can execute both 32- and 64-bit applications. However, 64-bit
applications must use 64-bit drivers, and 32-bit applications must use 32-bit drivers. Make sure
that you use the version of the driver that matches the bitness of the client application:
l ClouderaHiveODBC-32bit-[Version]-[Release]_i386.deb for the 32-bit
driver
l ClouderaHiveODBC-[Version]-[Release]_amd64.deb for the 64-bit driver

32 | Cloudera ODBC Driver for Apache Hive

 Linux Driver

[Version] is the version number of the driver, and [Release] is the release number for this
version of the driver.

You can install both versions of the driver on the same machine.

To install the Cloudera ODBC Driver for Apache Hive on Debian:

1. Log in as the root user, and then navigate to the folder containing the Debian package for
the driver.
2. Double-click ClouderaHiveODBC-32bit-[Version]-[Release]_i386.deb or
ClouderaHiveODBC-[Version]-[Release]_amd64.deb.
3. Follow the instructions in the installer to complete the installation process.

The Cloudera ODBC Driver for Apache Hive files are installed in the
/opt/cloudera/hiveodbc directory.

Note:

If the package manager in your Ubuntu distribution cannot resolve the libsasl
dependencies automatically when installing the driver, then download and manually
install the packages required by the version of the driver that you want to install.

Next, configure the environment variables on your machine to make sure that the ODBC driver
manager can work with the driver. For more information, see "Configuring the ODBC Driver
Manager on Non-Windows Machines" on page 36.

Verifying the Driver Version Number on Linux

If you need to verify the version of the Cloudera ODBC Driver for Apache Hive that is installed on
your Linux machine, you can query the version number through the command-line interface if the
driver was installed using an RPM file or Debian package.

To verify the driver version number on Linux:

Depending on your package manager, at the command prompt, run one of the following
commands:
l
yum list | grep ClouderaHiveODBC

l
rpm -qa | grep ClouderaHiveODBC

l
dpkg -l | grep clouderahiveodbc

The command returns information about the Cloudera ODBC Driver for Apache Hive that is
installed on your machine, including the version number.

Cloudera ODBC Driver for Apache Hive | 33

AIX Driver

AIX Driver
AIX System Requirements
The Cloudera ODBC Driver for Apache Hive supports Apache Hive versions 0.11 through 3.1, and
CDH versions 5.0 through 6.2.

Install the driver on client machines where the application is installed. Each machine that you
install the driver on must meet the following minimum system requirements:
l IBM AIX 5.3, 6.1, or 7.1
l 150 MB of available disk space
l One of the following ODBC driver managers installed:
o iODBC 3.52.9, 3.52.10, 3.52.11, or 3.52.12
o unixODBC 2.3.2, 2.3.3, or 2.3.4

To install the driver, you must have root access on the machine.

Installing the Driver on AIX

On 64-bit editions of AIX, you can execute both 32- and 64-bit applications. However, 64-bit
applications must use 64-bit drivers, and 32-bit applications must use 32-bit drivers. Make sure
that you use the version of the driver that matches the bitness of the client application:
l ClouderaHiveODBC-32bit-[Version]-[Release].ppc.rpm for the 32-bit
driver
l ClouderaHiveODBC-[Version]-[Release].ppc.rpm for the 64-bit driver

[Version] is the version number of the driver, and [Release] is the release number for this version
of the driver.

You can install both versions of the driver on the same machine.

To install the Cloudera ODBC Driver for Apache Hive on AIX:

1. Log in as the root user, and then navigate to the folder containing the RPM package for the
driver.
2. Run the following command from the command line, where [RPMFileName] is the file name
of the RPM package:
rpm --install [RPMFileName]

The Cloudera ODBC Driver for Apache Hive files are installed in the
/opt/cloudera/hiveodbc directory.

Next, configure the environment variables on your machine to make sure that the ODBC driver
manager can work with the driver. For more information, see "Configuring the ODBC Driver
Manager on Non-Windows Machines" on page 36.

34 | Cloudera ODBC Driver for Apache Hive

 AIX Driver

Verifying the Driver Version Number on AIX

If you need to verify the version of the Cloudera ODBC Driver for Apache Hive that is installed on
your AIX machine, you can query the version number through the command-line interface.

To verify the driver version number on AIX:

At the command prompt, run the following command:
rpm -qa | grep ClouderaHiveODBC

The command returns information about the Cloudera ODBC Driver for Apache Hive that is
installed on your machine, including the version number.

Cloudera ODBC Driver for Apache Hive | 35

Configuring the ODBC Driver Manager on Non-Windows Machines

Configuring the ODBC Driver Manager on Non-Windows

Machines
To make sure that the ODBC driver manager on your machine is configured to work with the
Cloudera ODBC Driver for Apache Hive, do the following:
l Set the library path environment variable to make sure that your machine uses the correct
ODBC driver manager. For more information, see "Specifying ODBC Driver Managers on
Non-Windows Machines" on page 36.
l If the driver configuration files are not stored in the default locations expected by the
ODBC driver manager, then set environment variables to make sure that the driver
manager locates and uses those files. For more information, see "Specifying the Locations
of the Driver Configuration Files" on page 36.

After configuring the ODBC driver manager, you can configure a connection and access your data
store through the driver.

Specifying ODBC Driver Managers on Non-Windows Machines

You need to make sure that your machine uses the correct ODBC driver manager to load the
driver. To do this, set the library path environment variable.

macOS

If you are using a macOS machine, then set the DYLD_LIBRARY_PATH environment variable to
include the paths to the ODBC driver manager libraries. For example, if the libraries are installed in
/usr/local/lib, then run the following command to set DYLD_LIBRARY_PATH for the current
user session:
export DYLD_LIBRARY_PATH=$DYLD_LIBRARY_PATH:/usr/local/lib

For information about setting an environment variable permanently, refer to the macOS shell
documentation.

Linux or AIX

If you are using a Linux or AIX machine, then set the LD_LIBRARY_PATH environment variable to
include the paths to the ODBC driver manager libraries. For example, if the libraries are installed in
/usr/local/lib, then run the following command to set LD_LIBRARY_PATH for the current
user session:
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/lib

For information about setting an environment variable permanently, refer to the Linux or AIX shell
documentation.

Specifying the Locations of the Driver Configuration Files

By default, ODBC driver managers are configured to use hidden versions of the odbc.ini and
odbcinst.ini configuration files (named .odbc.ini and .odbcinst.ini) located in the
home directory, as well as the cloudera.hiveodbc.ini file in the lib subfolder of the

36 | Cloudera ODBC Driver for Apache Hive

 Configuring the ODBC Driver Manager on Non-Windows Machines

driver installation directory. If you store these configuration files elsewhere, then you must set the
environment variables described below so that the driver manager can locate the files.

If you are using iODBC, do the following:

l Set ODBCINI to the full path and file name of the odbc.ini file.
l Set ODBCINSTINI to the full path and file name of the odbcinst.ini file.
l Set CLOUDERAHIVEINI to the full path and file name of the cloudera.hiveodbc.ini
file.

If you are using unixODBC, do the following:

l Set ODBCINI to the full path and file name of the odbc.ini file.
l Set ODBCSYSINI to the full path of the directory that contains the odbcinst.ini file.
l Set CLOUDERAHIVEINI to the full path and file name of the cloudera.hiveodbc.ini
file.

For example, if your odbc.ini and odbcinst.ini files are located in /usr/local/odbc
and your cloudera.hiveodbc.ini file is located in /etc, then set the environment
variables as follows:

For iODBC:
export ODBCINI=/usr/local/odbc/odbc.ini
export ODBCINSTINI=/usr/local/odbc/odbcinst.ini
export CLOUDERAHIVEINI=/etc/cloudera.hiveodbc.ini

For unixODBC:
export ODBCINI=/usr/local/odbc/odbc.ini
export ODBCSYSINI=/usr/local/odbc
export CLOUDERAHIVEINI=/etc/cloudera.hiveodbc.ini

To locate the cloudera.hiveodbc.ini file, the driver uses the following search order:
1. If the CLOUDERAHIVEINI environment variable is defined, then the driver searches for the
file specified by the environment variable.
2. The driver searches the directory that contains the driver library files for a file named
cloudera.hiveodbc.ini.
3. The driver searches the current working directory of the application for a file named
cloudera.hiveodbc.ini.
4. The driver searches the home directory for a hidden file named
.cloudera.hiveodbc.ini (prefixed with a period).
5. The driver searches the /etc directory for a file named cloudera.hiveodbc.ini.

Cloudera ODBC Driver for Apache Hive | 37

Configuring ODBC Connections on a Non-Windows Machine

Configuring ODBC Connections on a Non-Windows Machine

The following sections describe how to configure ODBC connections when using the Cloudera
ODBC Driver for Apache Hive on non-Windows platforms:
l "Creating a Data Source Name on a Non-Windows Machine" on page 38
l "Configuring a DSN-less Connection on a Non-Windows Machine" on page 41
l "Configuring Service Discovery Mode on a Non-Windows Machine" on page 43
l "Configuring Authentication on a Non-Windows Machine" on page 43
l "Configuring SSL Verification on a Non-Windows Machine" on page 46
l "Configuring Server-Side Properties on a Non-Windows Machine" on page 47
l "Configuring Logging Options" on page 48
l "Setting Driver-Wide Configuration Options on a Non-Windows Machine" on page 50
l "Testing the Connection" on page 50

Creating a Data Source Name on a Non-Windows Machine

Typically, after installing the Cloudera ODBC Driver for Apache Hive, you need to create a Data
Source Name (DSN). A DSN is a data structure that stores connection information so that it can be
used by the driver to connect to Hive.

You can specify connection settings in a DSN (in the odbc.ini file), in a connection string, or as
driver-wide settings (in the cloudera.hiveodbc.ini file). Settings in the connection string
take precedence over settings in the DSN, and settings in the DSN take precedence over driver-
wide settings.

The following instructions describe how to create a DSN by specifying connection settings in the
odbc.ini file. If your machine is already configured to use an existing odbc.ini file, then
update that file by adding the settings described below. Otherwise, copy the odbc.ini file from
the Setup subfolder in the driver installation directory to the home directory, and then update
the file as described below.

For information about specifying settings in a connection string, see "Configuring a DSN-less
Connection on a Non-Windows Machine" on page 41 and "Using a Connection String" on page 54.
For information about driver-wide settings, see "Setting Driver-Wide Configuration Options on a
Non-Windows Machine" on page 50.

To create a Data Source Name on a non-Windows machine:

1. In a text editor, open the odbc.ini configuration file.

Note:

If you are using a hidden copy of the odbc.ini file, you can remove the period (.) from
the start of the file name to make the file visible while you are editing it.

2. In the [ODBC Data Sources] section, add a new entry by typing a name for the DSN,
an equal sign (=), and then the name of the driver.

38 | Cloudera ODBC Driver for Apache Hive

 Configuring ODBC Connections on a Non-Windows Machine

For example, on a macOS machine:

[ODBC Data Sources]
Sample DSN=Cloudera Hive ODBC Driver

As another example, for a 32-bit driver on a Linux/AIX/Debian machine:

[ODBC Data Sources]
Sample DSN=Cloudera Hive ODBC Driver 32-bit

3. Create a section that has the same name as your DSN, and then specify configuration
options as key-value pairs in the section:
a. Set the Driver property to the full path of the driver library file that matches the
bitness of the application.

For example, on a macOS machine:

Driver=/opt
/cloudera/hiveodbc/lib/universal/libclouderahiveodbc.dy
lib

As another example, for a 32-bit driver on a Linux/AIX/Debian machine:

Driver=/opt/cloudera/hiveodbc/lib/32/libclouderahiveodb
c32.so

b. Set the HiveServerType property to one of the following values:

l If you are running Thrift Server 1, set the property to 1.
l If you are running Thrift Server 2, set the property to 2.

For example:
HiveServerType=1

c. Specify whether the driver uses the ZooKeeper service when connecting to Hive, and
provide the necessary connection information:
l To connect to Hive without using the ZooKeeper service do the following:
a. Set the ServiceDiscoveryMode property to 0.
b. Set the Host property to the IP address or host name of the server.
c. Set the Port property to the number of the TCP port that the server
uses to listen for client connections

For example:
ServiceDiscoveryMode=0
Host=192.168.222.160
Port=10000

Cloudera ODBC Driver for Apache Hive | 39

Configuring ODBC Connections on a Non-Windows Machine

l Or, to discover Hive Server 2 services through the ZooKeeper service, set
properties as described in "Configuring Service Discovery Mode on a Non-
Windows Machine" on page 43.
d. If authentication is required to access the server, then specify the authentication
mechanism and your credentials. For more information, see "Configuring
Authentication on a Non-Windows Machine" on page 43.
e. If you want to connect to the server through SSL, then enable SSL and specify the
certificate information. For more information, see "Configuring SSL Verification on a
Non-Windows Machine" on page 46.

Note:

If the AuthMech property is set to 2, SSL is not available.

f. If you want to configure server-side properties, then set them as key-value pairs
using a special syntax. For more information, see "Configuring Server-Side Properties
on a Non-Windows Machine" on page 47.
g. Optionally, set additional key-value pairs as needed to specify other optional
connection settings. For detailed information about all the configuration options
supported by the Cloudera ODBC Driver for Apache Hive, see "Driver Configuration
Options" on page 64.
4. Save the odbc.ini configuration file.

Note:

If you are storing this file in its default location in the home directory, then prefix the file
name with a period (.) so that the file becomes hidden. If you are storing this file in
another location, then save it as a non-hidden file (without the prefix), and make sure
that the ODBCINI environment variable specifies the location. For more information, see
"Specifying the Locations of the Driver Configuration Files" on page 36.

For example, the following is an odbc.ini configuration file for macOS containing a DSN that
connects to a Hive Thrift Server instance directly and authenticates the connection using a user
name and password:
[ODBC Data Sources]
Sample DSN=Cloudera Hive ODBC Driver
[Sample DSN]
Driver=/
opt/cloudera/hiveodbc/lib/universal/libclouderahiveodbc.dylib
HiveServerType=2
ServiceDiscoveryMode=0
Host=192.168.222.160
Port=10000
UID=username
PWD=userpassword

40 | Cloudera ODBC Driver for Apache Hive

 Configuring ODBC Connections on a Non-Windows Machine

As another example, the following is an odbc.ini configuration file for a 32-bit driver on a
Linux/AIX/Debian machine, containing a DSN that connects to a HiveThrift Server instance directly
and authenticates the connection using a user name and password:
[ODBC Data Sources]
Sample DSN=Cloudera Hive ODBC Driver 32-bit
[Sample DSN]
Driver=/opt/cloudera/hiveodbc/lib/32/libclouderahiveodbc32.so
HiveServerType=2
ServiceDiscoveryMode=0
Host=192.168.222.160
Port=10000
UID=username
PWD=userpassword

You can now use the DSN in an application to connect to the data store.

Configuring a DSN-less Connection on a Non-Windows Machine

To connect to your data store through a DSN-less connection, you need to define the driver in the
odbcinst.ini file and then provide a DSN-less connection string in your application.

If your machine is already configured to use an existing odbcinst.ini file, then update that file
by adding the settings described below. Otherwise, copy the odbcinst.ini file from the
Setup subfolder in the driver installation directory to the home directory, and then update the
file as described below.

To define a driver on a non-Windows machine:

1. In a text editor, open the odbcinst.ini configuration file.

Note:

If you are using a hidden copy of the odbcinst.ini file, you can remove the period (.)
from the start of the file name to make the file visible while you are editing it.

2. In the [ODBC Drivers] section, add a new entry by typing a name for the driver, an
equal sign (=), and then Installed.

For example:
[ODBC Drivers]
Cloudera ODBC Driver for Apache Hive=Installed

3. Create a section that has the same name as the driver (as specified in the previous step),
and then specify the following configuration options as key-value pairs in the section:
a. Set the Driver property to the full path of the driver library file that matches the
bitness of the application.

For example, on a macOS machine:

Cloudera ODBC Driver for Apache Hive | 41

Configuring ODBC Connections on a Non-Windows Machine

Driver=/opt
/cloudera/hiveodbc/lib/universal/libclouderahiveodbc.dy
lib

As another example, for a 32-bit driver on a Linux/AIX/Debian machine:

Driver=/opt/cloudera/hiveodbc/lib/32/libclouderahiveodb
c32.so

b. Optionally, set the Description property to a description of the driver.

For example:
Description=Cloudera ODBC Driver for Apache Hive

4. Save the odbcinst.ini configuration file.

Note:

If you are storing this file in its default location in the home directory, then prefix the file
name with a period (.) so that the file becomes hidden. If you are storing this file in
another location, then save it as a non-hidden file (without the prefix), and make sure
that the ODBCINSTINI or ODBCSYSINI environment variable specifies the location. For
more information, see "Specifying the Locations of the Driver Configuration Files" on page
36.

For example, the following is an odbcinst.ini configuration file for macOS:

[ODBC Drivers]
Cloudera Hive ODBC Driver=Installed
[Cloudera Hive ODBC Driver]
Description=Cloudera ODBC Driver for Apache Hive
Driver=/opt/cloudera/hiveodbc/lib/universal/libclouderahiveodbc.
dylib

As another example, the following is an odbcinst.ini configuration file for both the 32- and
64-bit drivers on Linux/AIX/Debian:
[ODBC Drivers]
Cloudera Hive ODBC Driver 32-bit=Installed
Cloudera Hive ODBC Driver 64-bit=Installed
[Cloudera Hive ODBC Driver 32-bit]
Description=Cloudera ODBC Driver for Apache Hive (32-bit)
Driver=/opt/cloudera/hiveodbc/lib/32/libclouderahiveodbc32.so
[Cloudera Hive ODBC Driver 64-bit]
Description=Cloudera ODBC Driver for Apache Hive (64-bit)
Driver=/opt/cloudera/hiveodbc/lib/64/libclouderahiveodbc64.so

You can now connect to your data store by providing your application with a connection string
where the Driver property is set to the driver name specified in the odbcinst.ini file, and all

42 | Cloudera ODBC Driver for Apache Hive

 Configuring ODBC Connections on a Non-Windows Machine

the other necessary connection properties are also set. For more information, see "DSN-less
Connection String Examples" in "Using a Connection String" on page 54.

For instructions about configuring specific connection features, see the following:
l "Configuring Service Discovery Mode on a Non-Windows Machine" on page 43
l "Configuring Authentication on a Non-Windows Machine" on page 43
l "Configuring SSL Verification on a Non-Windows Machine" on page 46
l "Configuring Server-Side Properties on a Non-Windows Machine" on page 47

For detailed information about all the connection properties that the driver supports, see "Driver
Configuration Options" on page 64.

Configuring Service Discovery Mode on a Non-Windows Machine

You can configure the Cloudera ODBC Driver for Apache Hive to discover Hive Server 2 services
through ZooKeeper.

You can set the connection properties described below in a connection string, in a DSN (in the
odbc.ini file), or as a driver-wide setting (in the cloudera.hiveodbc.ini file). Settings in
the connection string take precedence over settings in the DSN, and settings in the DSN take
precedence over driver-wide settings.

To enable Service Discovery Mode on a non-Windows machine:

1. Set the ServiceDiscoveryMode connection attribute to 1.
2. Set the ZKNamespace connection attribute to specify the namespace on ZooKeeper under
which Hive Server 2 znodes are added.
3. Set the Host connection attribute to specify the ZooKeeper ensemble as a comma-
separated list of ZooKeeper servers. For example, type the following, where [ZK_Host] is the
IP address or host name of the ZooKeeper server and [ZK_Port] is the number of the TCP
port that the ZooKeeper server uses to listen for client connections:
[ZK_Host1]:[ZK_Port1],[ZK_Host2]:[ZK_Port2]

Important:

When ServiceDiscoveryMode is set to 1, connections to Hive Server 1 are not supported

and the Port connection attribute is not applicable.

Configuring Authentication on a Non-Windows Machine

Some Hive Server 2 instances are configured to require authentication for access. To connect to a
Hive server, you must configure the Cloudera ODBC Driver for Apache Hive to use the
authentication mechanism that matches the access requirements of the server and provides the
necessary credentials.

For information about how to determine the type of authentication your Hive server requires, see
"Authentication Mechanisms" on page 52.

Cloudera ODBC Driver for Apache Hive | 43

Configuring ODBC Connections on a Non-Windows Machine

You can set the connection properties for authentication in a connection string, in a DSN (in the
odbc.ini file), or as a driver-wide setting (in the cloudera.hiveodbc.ini file). Settings in
the connection string take precedence over settings in the DSN, and settings in the DSN take
precedence over driver-wide settings.

Depending on the authentication mechanism you use, there might be additional connection
attributes that you must define. For more information about the attributes involved in
configuring authentication, see "Driver Configuration Options" on page 64.

If cookie-based authentication is enabled in your Hive Server 2 database, you can specify a list of
authentication cookies in the HTTPAuthCookies connection property. In this case, the driver
authenticates the connection once based on the provided authentication credentials. It then uses
the cookie generated by the server for each subsequent request in the same connection. For
more information, see "HTTPAuthCookies" on page 87.

Using No Authentication

When connecting to a Hive server of type Hive Server 1, you must use No Authentication. When
you use No Authentication, Binary is the only Thrift transport protocol that is supported.

To configure a connection without authentication:

1. Set the AuthMech connection attribute to 0.
2. If the Hive server is configured to use SSL, then configure SSL for the connection. For more
information, see "Configuring SSL Verification on a Non-Windows Machine" on page 46.

Using Kerberos

Kerberos must be installed and configured before you can use this authentication mechanism. For
more information, refer to the MIT Kerberos Documentation: http://web.mit.edu/kerberos/krb5-
latest/doc/.

This authentication mechanism is available only for Hive Server 2. When you use Kerberos
authentication, the Binary transport protocol is not supported.

To configure Kerberos authentication:

1. Set the AuthMech connection attribute to 1.
2. Choose one:
l To use the default realm defined in your Kerberos setup, do not set the KrbRealm
attribute.
l Or, if your Kerberos setup does not define a default realm or if the realm of your Hive
server is not the default, then set the appropriate realm using the KrbRealm
attribute.
3. Optionally, if you are using MIT Kerberos and a Kerberos realm is specified using the
KrbRealm connection attribute, then choose one:
l To have the Kerberos layer canonicalize the server's service principal name, leave the
ServicePrincipalCanonicalization attribute set to 1.

44 | Cloudera ODBC Driver for Apache Hive

 Configuring ODBC Connections on a Non-Windows Machine

l Or, to prevent the Kerberos layer from canonicalizing the server's service principal
name, set the ServicePrincipalCanonicalization attribute to 0.
4. Set the KrbHostFQDN attribute to the fully qualified domain name of the Hive Server 2
host.

Note:

To use the Hive server host name as the fully qualified domain name for Kerberos
authentication, set KrbHostFQDN to _HOST.

5. Set the KrbServiceName attribute to the service name of the Hive .

6. To allow the driver to pass your credentials directly to the server for use in authentication,
set DelegateKrbCreds to 1.
7. Set the ThriftTransport connection attribute to the transport protocol to use in the
Thrift layer.

Important:

When using this authentication mechanism, Binary (ThriftTransport=0) is not

supported.

8. If the Hive server is configured to use SSL, then configure SSL for the connection. For more
information, see "Configuring SSL Verification on a Non-Windows Machine" on page 46.

Using User Name

This authentication mechanism requires a user name but does not require a password. The user
name labels the session, facilitating database tracking.

This authentication mechanism is available only for Hive Server 2. Most default configurations of
Hive Server 2 require User Name authentication. When you use User Name authentication, SSL is
not supported and SASL is the only Thrift transport protocol available.

To configure User Name authentication:

1. Set the AuthMech connection attribute to 2.
2. Set the UID attribute to an appropriate user name for accessing the Hive server.

Using User Name And Password

This authentication mechanism requires a user name and a password.

This authentication mechanism is available only for Hive Server 2.

To configure User Name And Password authentication:

1. Set the AuthMech connection attribute to 3.
2. Set the UID attribute to an appropriate user name for accessing the Hive server.

Cloudera ODBC Driver for Apache Hive | 45

Configuring ODBC Connections on a Non-Windows Machine

3. Set the PWD attribute to the password corresponding to the user name you provided
above.
4. Set the ThriftTransport connection attribute to the transport protocol to use in the
Thrift layer.
5. If the Hive server is configured to use SSL, then configure SSL for the connection. For more
information, see "Configuring SSL Verification on a Non-Windows Machine" on page 46.

Delegating Authentication to a Specific User

Some Hive Server 2 instances support the ability to delegate all operations against Hive to the
specified user, rather than to the authenticated user for the connection.

To delegate all operations to a specified user:

Set the DelegationUID configuration option to the name of the user to be delegated.

If the server returns an error message such as Failed to validate proxy privilege of [RealUser]
for [DelegationUID], you may need to modify the server's core-site.xml configuration file,
as follows:
1. In the server's core-site.xml configuration file, add the following properties:
hadoop.proxyuser.[RealUser].groups=*
hadoop.proxyuser.[RealUser].hosts=*

Where [RealUser] is the authenticated user for the connection.

2. If you are using Kerberos authentication, then in the server's core-site.xml
configuration file, add the following properties:
hadoop.proxyuser.[Principal].groups=*
hadoop.proxyuser.[Principal].hosts=*

Where [Principal] is the primary Kerberos principal user. For example, if the primary
Kerberos principal user is [email protected], replace [Principal] with kerbuser.

For more information on resolving this error, see your Hive Server documentation.

Configuring SSL Verification on a Non-Windows Machine

If you are connecting to a Hive server that has Secure Sockets Layer (SSL) enabled, you can
configure the driver to connect to an SSL-enabled socket. When using SSL to connect to a server,
the driver supports identity verification between the client (the driver itself) and the server.

Note:

If the AuthMech property is set to 2, SSL is not available.

You can set the connection properties described below in a connection string, in a DSN (in the
odbc.ini file), or as a driver-wide setting (in the cloudera.hiveodbc.ini file). Settings in

46 | Cloudera ODBC Driver for Apache Hive

 Configuring ODBC Connections on a Non-Windows Machine

the connection string take precedence over settings in the DSN, and settings in the DSN take
precedence over driver-wide settings.

To configure SSL verification on a non-Windows machine:

1. To enable SSL connections, set the SSL attribute to 1.
2. To allow authentication using self-signed certificates that have not been added to the list of
trusted certificates, set the AllowSelfSignedServerCert attribute to 1.
3. To allow the common name of a CA-issued SSL certificate to not match the host name of the
Hive server, set the CAIssuedCertNamesMismatch attribute to 1.
4. Choose one:
l To configure the driver to load SSL certificates from a specific .pem file when verifying
the server, set the TrustedCerts attribute to the full path of the .pem file.
l Or, to use the trusted CA certificates .pem file that is installed with the driver, do not
specify a value for the TrustedCerts attribute.
5. To configure two-way SSL verification, set the TwoWaySSL attribute to 1 and then do the
following:
a. Set the ClientCert attribute to the full path of the .pem file containing the
client's certificate.
b. Set the ClientPrivateKey attribute to the full path of the file containing the
client's private key.
c. If the private key file is protected with a password, set the
ClientPrivateKeyPassword attribute to the password.
6. To specify the minimum version of TLS to use, set the Min_TLS property to the minimum
version of TLS. Supported options include 1.0 for TLS 1.0, 1.1 for TLS 1.1, and 1.2 for
TLS 1.2.

Configuring Server-Side Properties on a Non-Windows Machine

You can use the driver to apply configuration properties to the Hive server.

You can set the connection properties described below in a connection string, in a DSN (in the
odbc.ini file), or as a driver-wide setting (in the cloudera.hiveodbc.ini file). Settings in
the connection string take precedence over settings in the DSN, and settings in the DSN take
precedence over driver-wide settings.

To configure server-side properties on a non-Windows machine:

1. To set a server-side property, use the syntax SSP_[SSPKey]=[SSPValue], where
[SSPKey] is the name of the server-side property and [SSPValue] is the value to specify for
that property. For example, to set the mapreduce.job.queuename property to
myQueue, type the following:
SSP_mapreduce.job.queuename=myQueue

Cloudera ODBC Driver for Apache Hive | 47

Configuring ODBC Connections on a Non-Windows Machine

Note:

l When setting a server-side property in a connection string, it is recommended that

you enclose the value in braces ({ }) to make sure that special characters can be
properly escaped.
l For a list of all Hadoop and Hive server-side properties that your implementation
supports, type set -v at the Hive CLI command line or Beeline. You can also
execute the set -v query after connecting using the driver.

2. To change the method that the driver uses to apply server-side properties, do one of the
following:
l To configure the driver to apply each server-side property by executing a query when
opening a session to the Hive server, set the ApplySSPWithQueries property to
1.
l Or, to configure the driver to use a more efficient method for applying server-side
properties that does not involve additional network round-tripping, set the
ApplySSPWithQueries property to 0.

Note:

The more efficient method is not available for Hive Server 1, and it might not be
compatible with some Hive Server builds. If the server-side properties do not take effect
when the ApplySSPWithQueries property is set to 0, then set it to 1.

3. To disable the driver's default behavior of converting server-side property key names to all
lower-case characters, set the LCaseSspKeyName property to 0.

Configuring Logging Options

To help troubleshoot issues, you can enable logging in the driver.

Important:

Only enable logging long enough to capture an issue. Logging decreases performance and can
consume a large quantity of disk space.

The settings for logging apply to every connection that uses the Cloudera ODBC Driver for
Apache Hive, so make sure to disable the feature after you are done using it.

Logging is configured through driver-wide settings in the cloudera.hiveodbc.ini file, which

apply to all connections that use the driver.

To enable logging:
1. Open the cloudera.hiveodbc.ini configuration file in a text editor.
2. To specify the level of information to include in log files, set the LogLevel property to one
of the following numbers:

48 | Cloudera ODBC Driver for Apache Hive

 Configuring ODBC Connections on a Non-Windows Machine

LogLevel Value Description

0 Disables all logging.

1 Logs severe error events that lead the driver to abort.

2 Logs error events that might allow the driver to continue

running.

3 Logs events that might result in an error if action is not taken.

4 Logs general information that describes the progress of the

driver.

5 Logs detailed information that is useful for debugging the driver.

6 Logs all driver activity.

3. Set the LogPath key to the full path to the folder where you want to save log files.
4. Set the LogFileCount key to the maximum number of log files to keep.

Note:

After the maximum number of log files is reached, each time an additional file is created,
the driver deletes the oldest log file.

5. Set the LogFileSize key to the maximum size of each log file in bytes.

Note:

After the maximum file size is reached, the driver creates a new file and continues logging.

6. Save the cloudera.hiveodbc.ini configuration file.

7. Restart your ODBC application to make sure that the new settings take effect.

The Cloudera ODBC Driver for Apache Hive produces the following log files at the location you
specify using the LogPath key:
l A clouderahiveodbcdriver.log file that logs driver activity that is not specific to a
connection.
l A clouderahiveodbcdriver_connection_[Number].log file for each
connection made to the database, where [Number] is a number that identifies each log file.
This file logs driver activity that is specific to the connection.

To disable logging:
1. Open the cloudera.hiveodbc.ini configuration file in a text editor.
2. Set the LogLevel key to 0.

Cloudera ODBC Driver for Apache Hive | 49

Configuring ODBC Connections on a Non-Windows Machine

3. Save the cloudera.hiveodbc.ini configuration file.

4. Restart your ODBC application to make sure that the new settings take effect.

Setting Driver-Wide Configuration Options on a Non-Windows Machine

When you specify connection settings in a DSN or connection string, those settings apply only
when you connect to Hive using that particular DSN or string. As an alternative, you can specify
settings that apply to every connection that uses the Cloudera ODBC Driver for Apache Hive by
configuring them in the cloudera.hiveodbc.ini file.

Note:

Settings in the connection string take precedence over settings in the DSN, and settings in the
DSN take precedence over driver-wide settings.

To set driver-wide configuration options on a non-Windows machine:

1. In a text editor, open the cloudera.hiveodbc.ini configuration file.
2. In the [Driver] section, specify configuration options as key-value pairs. Start a new line
for each key-value pair.

For example, to enable User Name authentication using "cloudera" as the user name, type
the following:
AuthMech=2
UID=cloudera

For detailed information about all the configuration options supported by the driver, see
"Driver Configuration Options" on page 64.
3. Save the cloudera.hiveodbc.ini configuration file.

Testing the Connection

To test the connection, you can use an ODBC-enabled client application. For a basic connection
test, you can also use the test utilities that are packaged with your driver manager installation. For
example, the iODBC driver manager includes simple utilities called iodbctest and iodbctestw.
Similarly, the unixODBC driver manager includes simple utilities called isql and iusql.

Using the iODBC Driver Manager

You can use the iodbctest and iodbctestw utilities to establish a test connection with your driver.
Use iodbctest to test how your driver works with an ANSI application, or use iodbctestw to test
how your driver works with a Unicode application.

50 | Cloudera ODBC Driver for Apache Hive

 Configuring ODBC Connections on a Non-Windows Machine

Note:

There are 32-bit and 64-bit installations of the iODBC driver manager available. If you have only
one or the other installed, then the appropriate version of iodbctest (or iodbctestw) is available.
However, if you have both 32- and 64-bit versions installed, then you need to make sure that
you are running the version from the correct installation directory.

For more information about using the iODBC driver manager, see http://www.iodbc.org.

To test your connection using the iODBC driver manager:

1. Run iodbctest or iodbctestw.
2. Optionally, if you do not remember the DSN, then type a question mark (?) to see a list of
available DSNs.
3. Type the connection string for connecting to your data store, and then press ENTER. For
more information, see .

If the connection is successful, then the SQL> prompt appears.

Using the unixODBC Driver Manager

You can use the isql and iusql utilities to establish a test connection with your driver and your
DSN. isql and iusql can only be used to test connections that use a DSN. Use isql to test how your
driver works with an ANSI application, or use iusql to test how your driver works with a Unicode
application.

Note:

There are 32-bit and 64-bit installations of the unixODBC driver manager available. If you have
only one or the other installed, then the appropriate version of isql (or iusql) is available.
However, if you have both 32- and 64-bit versions installed, then you need to make sure that
you are running the version from the correct installation directory.

For more information about using the unixODBC driver manager, see http://www.unixodbc.org.

To test your connection using the unixODBC driver manager:

Run isql or iusql by using the corresponding syntax:
l isql [DataSourceName]
l iusql [DataSourceName]

[DataSourceName] is the DSN that you are using for the connection.

If the connection is successful, then the SQL> prompt appears.

Note:

For information about the available options, run isql or iusql without providing a DSN.

Cloudera ODBC Driver for Apache Hive | 51

Authentication Mechanisms

Authentication Mechanisms
To connect to a Hive server, you must configure the Cloudera ODBC Driver for Apache Hive to use
the authentication mechanism that matches the access requirements of the server and provides
the necessary credentials. To determine the authentication settings that your Hive server
requires, check the server configuration and then refer to the corresponding section below.

Hive Server 1
You must use No Authentication as the authentication mechanism. Hive Server 1 instances do not
support authentication.

Hive Server 2

Note:

Most default configurations of Hive Server 2 require User Name authentication.

Configuring authentication for a connection to a Hive Server 2 instance involves setting the
authentication mechanism, the Thrift transport protocol, and SSL support. To determine the
settings that you need to use, check the following three properties in the hive-site.xml file in
the Hive server that you are connecting to:
l hive.server2.authentication
l hive.server2.transport.mode
l hive.server2.use.SSL

Use the following table to determine the authentication mechanism that you need to configure,
based on the hive.server2.authentication value in the hive-site.xml file:
hive.server2.authentication Authentication Mechanism

NOSASL No Authentication

KERBEROS Kerberos

NONE User Name

LDAP User Name and Password

Use the following table to determine the Thrift transport protocol that you need to configure,
based on the hive.server2.authentication and hive.server2.transport.mode
values in the hive-site.xml file:
hive.server2.authentication hive.server2.transport.mode Thrift Transport Protocol

NOSASL binary Binary

52 | Cloudera ODBC Driver for Apache Hive

 Authentication Mechanisms

hive.server2.authentication hive.server2.transport.mode Thrift Transport Protocol

KERBEROS binary or http SASL or HTTP

NONE binary or http SASL or HTTP

LDAP binary or http SASL or HTTP

To determine whether SSL should be enabled or disabled for your connection, check the
hive.server2.use.SSL value in the hive-site.xml file. If the value is true, then you
must enable and configure SSL in your connection. If the value is false, then you must disable SSL
in your connection.

For detailed instructions on how to configure authentication when using the Windows driver, see
"Configuring Authentication on Windows" on page 12.

For detailed instructions on how to configure authentication when using a non-Windows driver,
see "Configuring Authentication on a Non-Windows Machine" on page 43.

Cloudera ODBC Driver for Apache Hive | 53

Using a Connection String

Using a Connection String

For some applications, you might need to use a connection string to connect to your data source.
For detailed information about how to use a connection string in an ODBC application, refer to the
documentation for the application that you are using.

The connection strings in the following sections are examples showing the minimum set of
connection attributes that you must specify to successfully connect to the data source.
Depending on the configuration of the data source and the type of connection you are working
with, you might need to specify additional connection attributes. For detailed information about
all the attributes that you can use in the connection string, see "Driver Configuration Options" on
page 64.

DSN Connection String Example

The following is an example of a connection string for a connection that uses a DSN:
DSN=[DataSourceName]

[DataSourceName] is the DSN that you are using for the connection.

You can set additional configuration options by appending key-value pairs to the connection
string. Configuration options that are passed in using a connection string take precedence over
configuration options that are set in the DSN.

DSN-less Connection String Examples

Some applications provide support for connecting to a data source using a driver without a DSN.
To connect to a data source without using a DSN, use a connection string instead.

The placeholders in the examples are defined as follows, in alphabetical order:

l [DomainName] is the fully qualified domain name of the Hive server host.
l [Namespace] is the namespace on ZooKeeper under which Hive Server 2 znodes are added.
l [PortNumber] is the number of the TCP port that the Hive server uses to listen for client
connections.
l [Realm] is the Kerberos realm of the Hive server host.
l [Server] is the IP address or host name of the Hive server to which you are connecting.
l [ServiceName] is the Kerberos service principal name of the Hive server.
l [YourPassword] is the password corresponding to your user name.
l [YourUserName] is the user name that you use to access the Hive server.

Connecting to a Hive Server 1 Instance

The following is the format of a DSN-less connection string that connects to a Hive Server 1
instance:
Driver=Cloudera Hive ODBC Driver;HiveServerType=1;
Host=[Server];Port=[PortNumber];

54 | Cloudera ODBC Driver for Apache Hive

 Using a Connection String

For example:
Driver=Cloudera Hive ODBC Driver;HiveServerType=1;
Host=192.168.222.160;Port=10000;

Connecting to a Standard Hive Server 2 Instance

The following is the format of a DSN-less connection string for a standard connection to a Hive
Server 2 instance. By default, the driver is configured to connect to a Hive Server 2 instance that
requires User Name authentication, and the driver uses anonymous as the user name.
Driver=Cloudera Hive ODBC Driver;Host=[Server];
Port=[PortNumber];

For example:
Driver=Cloudera Hive ODBC Driver;Host=192.168.222.160;
Port=10000;

Connecting using Dynamic Service Discovery

The following is the format of a DSN-less connection string that discovers Hive Server 2 services via
the ZooKeeper service.
Driver=Cloudera Hive ODBC Driver;ServiceDiscoveryMode=1;
Host=[Server1]:[PortNumber1], [Server2]:[PortNumber2],
[Server3]:[PortNumber3];ZKNamespace=[Namespace];

For example:
Driver=Cloudera Hive ODBC Driver;ServiceDiscoveryMode=1;
Host=192.168.222.160:10000, 192.168.222.165:10000,
192.168.222.231:10000;ZKNamespace=hiveserver;

Connecting to a Hive Server 2 Instance Without Authentication

The following is the format of a DSN-less connection string for a Hive Server 2 instance that does
not require authentication.
Driver=Cloudera Hive ODBC Driver;Host=[Server];
Port=[PortNumber];AuthMech=0;

For example:
Driver=Cloudera Hive ODBC Driver;Host=192.168.222.160;
Port=10000;AuthMech=0;

Connecting to a Hive Server that Requires Kerberos Authentication

The following is the format of a DSN-less connection string that connects to a Hive Server 2
instance requiring Kerberos authentication:
Driver=Cloudera Hive ODBC Driver;Host=[Server];
Port=[PortNumber];AuthMech=1;KrbRealm=[Realm];

Cloudera ODBC Driver for Apache Hive | 55

Using a Connection String

KrbHostFQDN=[DomainName];KrbServiceName=[ServiceName];

For example:
Driver=Cloudera Hive ODBC Driver;Host=192.168.222.160;
Port=10000;AuthMech=1;KrbRealm=CLOUDERA;
KrbHostFQDN=localhost.localdomain;KrbServiceName=hive;

Connecting to a Hive Server that Requires User Name And Password Authentication (LDAP)

The following is the format of a DSN-less connection string that connects to a Hive Server 2
instance requiring User Name And Password authentication:
Driver=Cloudera Hive ODBC Driver;Host=[Server];
Port=[PortNumber];AuthMech=3;UID=[YourUserName];
PWD=[YourPassword];

For example:
Driver=Cloudera Hive ODBC Driver;Host=192.168.222.160;
Port=10000;AuthMech=3;UID=cloudera;PWD=cloudera;

56 | Cloudera ODBC Driver for Apache Hive

 Features

Features
For more information on the features of the Cloudera ODBC Driver for Apache Hive, see the
following:
l "SQL Connector for HiveQL" on page 57
l "Data Types" on page 57
l "Catalog and Schema Support" on page 59
l "hive_system Table" on page 59
l "Server-Side Properties" on page 59
l "Get Tables With Query" on page 61
l "Active Directory" on page 61
l "Write-back" on page 61
l "Timestamp Function Support" on page 62
l "Dynamic Service Discovery using ZooKeeper" on page 62
l "Security and Authentication" on page 62

SQL Connector for HiveQL

The native query language supported by Hive is HiveQL. For simple queries, HiveQL is a subset of
SQL-92. However, the syntax is different enough that most applications do not work with native
HiveQL.

To bridge the difference between SQL and HiveQL, the SQL Connector feature translates standard
SQL-92 queries into equivalent HiveQL queries. The SQL Connector performs syntactical
translations and structural transformations. For example:
l Quoted Identifiers: The double quotes (") that SQL uses to quote identifiers are translated
into back quotes (`) to match HiveQL syntax. The SQL Connector needs to handle this
translation because even when a driver reports the back quote as the quote character,
some applications still generate double-quoted identifiers.
l Table Aliases: Support is provided for the AS keyword between a table reference and its
alias, which HiveQL normally does not support.
l JOIN, INNER JOIN, and CROSS JOIN: SQL JOIN, INNER JOIN, and CROSS JOIN syntax is
translated to HiveQL JOIN syntax.
l TOP N/LIMIT: SQL TOP N queries are transformed to HiveQL LIMIT queries.

Data Types
The Cloudera ODBC Driver for Apache Hive supports many common data formats, converting
between Hive data types and SQL data types.

The following table lists the supported data type mappings.

Cloudera ODBC Driver for Apache Hive | 57

Features

Hive Type SQL Type

BIGINT SQL_BIGINT

BINARY SQL_VARBINARY

BOOLEAN SQL_BIT

CHAR(n) SQL_CHAR

Note:

SQL_WCHAR is returned instead if the Unicode SQL Character Types

configuration option (the UseUnicodeSqlCharacterTypes
key) is enabled.

DATE SQL_TYPE_DATE

DECIMAL SQL_DECIMAL

DECIMAL(p,s) SQL_DECIMAL

DOUBLE SQL_DOUBLE

FLOAT SQL_REAL

INT SQL_INTEGER

SMALLINT SQL_SMALLINT

STRING SQL_VARCHAR

Note:

SQL_WVARCHAR is returned instead if the Unicode SQL Character

Types configuration option (the
UseUnicodeSqlCharacterTypes key) is enabled.

TIMESTAMP SQL_TYPE_TIMESTAMP

TINYINT SQL_TINYINT

VARCHAR(n) SQL_VARCHAR

58 | Cloudera ODBC Driver for Apache Hive

 Features

Note:

The aggregate types (ARRAY, MAP, and STRUCT) are not supported. Columns of aggregate types
are treated as STRING columns.

The interval types (YEAR TO MONTH and DAY TIME) are supported only in query expressions and
predicates. Interval types are not supported as column data types in tables.

Catalog and Schema Support

The Cloudera ODBC Driver for Apache Hive supports both catalogs and schemas to make it easy
for the driver to work with various ODBC applications. Since Hive only organizes tables into
schemas/databases, the driver provides a synthetic catalog named HIVE under which all of the
schemas/databases are organized. The driver also maps the ODBC schema to the Hive
schema/database.

hive_system Table
A pseudo-table called hive_system can be used to query for Hive cluster system environment
information. The pseudo-table is under the pseudo-schema called hive_system. The table has two
STRING type columns, envkey and envvalue. Standard SQL can be executed against the hive_
system table. For example:
SELECT * FROM HIVE.hive_system.hive_system WHERE envkey LIKE
'%hive%'

The above query returns all of the Hive system environment entries whose key contains the word
"hive". A special query, set -v, is executed to fetch system environment information. Some
versions of Hive do not support this query. For versions of Hive that do not support querying
system environment information, the driver returns an empty result set.

Server-Side Properties
The Cloudera ODBC Driver for Apache Hive allows you to set server-side properties via a DSN.
Server-side properties specified in a DSN affect only the connection that is established using the
DSN.

You can also specify server-side properties for connections that do not use a DSN. To do this, use
the Cloudera Hive ODBC Driver Configuration tool that is installed with the Windows version of the
driver, or set the appropriate configuration options in your connection string or the
cloudera.hiveodbc.ini file. Properties specified in the driver configuration tool or the
cloudera.hiveodbc.ini file apply to all connections that use the Cloudera ODBC Driver for
Apache Hive.

For more information about setting server-side properties when using the Windows driver, see
"Configuring Server-Side Properties on Windows" on page 21. For information about setting
server-side properties when using the driver on a non-Windows platform, see "Configuring Server-
Side Properties on a Non-Windows Machine" on page 47.

Cloudera ODBC Driver for Apache Hive | 59

Features

Temporary Tables
The driver supports the creation of temporary tables and lets you insert literal values into
temporary tables. Temporary tables are only accessible by the ODBC connection that created
them, and are dropped when the connection is closed.

CREATE TABLE Statement for Temporary Tables

The driver supports the following DDL syntax for creating temporary tables:
<create table statement> := CREATE TABLE <temporary table name>
<left paren><column definition list><right paren>
<column definition list> := <column definition>[, <column
definition>]*
<column definition> := <column name> <data type>
<temporary table name> := <double quote><number sign><table
name><double quote>
<left paren> := (
<right paren> := )
<double quote> := "
<number sign> := #

The following is an example of a SQL statement for creating a temporary table:

CREATE TABLE "#TEMPTABLE1" (C1 DATATYPE_1, C2 DATATYPE_2, ...,
Cn DATATYPE_n)

The temporary table name in a SQL query must be surrounded by double quotes ("), and the
name must begin with a number sign (#).

Note:

You can only use data types that are supported by Hive.

INSERT Statement for Temporary Tables

The driver supports the following DDL syntax for inserting data into temporary tables:
<insert statement> := INSERT INTO <temporary table name> <left
paren><column name list><right paren> VALUES <left
paren><literal value list><right paren>
<column name list> := <column name>[, <column name>]*
<literal value list> := <literal value>[, <literal value>]*
<temporary table name> := <double quote><number sign><table
name><double quote>
<left paren> := (
<right paren> := )
<double quote> := "
<number sign> := #

60 | Cloudera ODBC Driver for Apache Hive

 Features

The following is an example of a SQL statement for inserting data into temporary tables:
INSERT INTO "#TEMPTABLE1" values (VAL(C1), VAL(C2) ... VAL(Cn) )

VAL(C1) is the literal value for the first column in the table, and VAL(Cn) is the literal value for the
nth column in the table.

Note:

The INSERT statement is only supported for temporary tables.

Get Tables With Query

The Get Tables With Query configuration option allows you to choose whether to use the SHOW
TABLES query or the GetTables API call to retrieve table names from a database.

Hive Server 2 has a limit on the number of tables that can be in a database when handling the
GetTables API call. When the number of tables in a database is above the limit, the API call will
return a stack overflow error or a timeout error. The exact limit and the error that appears
depends on the JVM settings.

As a workaround for this issue, enable the Get Tables with Query configuration option or the
GetTablesWithQuery key to use the query instead of the API call.

Active Directory
The Cloudera ODBC Driver for Apache Hive supports Active Directory Kerberos on Windows. There
are two prerequisites for using Active Directory Kerberos on Windows:
l MIT Kerberos is not installed on the client Windows machine.
l The MIT Kerberos Hadoop realm has been configured to trust the Active Directory realm so
that users in the Active Directory realm can access services in the MIT Kerberos Hadoop
realm.

Write-back
The Cloudera ODBC Driver for Apache Hive supports translation for the following syntax when
connecting to a Hive Server 2 instance that is running Hive 0.14 or later:
l INSERT
l UPDATE
l DELETE
l CREATE
l DROP

If the statement contains non-standard SQL-92 syntax, then the driver is unable to translate the
statement to SQL and instead falls back to using HiveQL.

Cloudera ODBC Driver for Apache Hive | 61

Features

Timestamp Function Support

The Cloudera ODBC Driver for Apache Hive supports the following ODBC functions for working
with data of type TIMESTAMP:
l TIMESTAMPADD: You can call this function to increment a TIMESTAMP value by a specified
interval of time.
l TIMESTAMPDIFF: You can call this function to calculate the interval of time between two
specified TIMESTAMP values.

The types of time intervals that are supported for these functions might vary depending on the
Hive server version that you are connecting to. To return a list of the intervals supported for
TIMESTAMPADD, call the SQLGetInfo catalog function using SQL_TIMEDATE_ADD_INTERVALS as
the argument. Similarly, to return a list of the intervals supported for TIMESTAMPDIFF, call
SQLGetInfo using SQL_TIMEDATE_DIFF_INTERVALS as the argument.

Note:

The SQL_TSI_FRAC_SECOND interval is not supported by Hive.

Dynamic Service Discovery using ZooKeeper

The Cloudera ODBC Driver for Apache Hive can be configured to discover Hive Server 2 services via
the ZooKeeper service.

For information about configuring this feature in the Windows driver, see "Creating a Data Source
Name on Windows" on page 7 or "Configuring a DSN-less Connection on Windows" on page 10.
For information about configuring this feature when using the driver on a non-Windows platform,
see "Configuring Service Discovery Mode on a Non-Windows Machine" on page 43.

Security and Authentication

To protect data from unauthorized access, some Hive data stores require connections to be
authenticated with user credentials or the SSL protocol. The Cloudera ODBC Driver for Apache
Hive provides full support for these authentication protocols.

Note:

In this documentation, "SSL" refers to both TLS (Transport Layer Security) and SSL (Secure
Sockets Layer). The driver supports TLS 1.0, 1.1, and 1.2. The SSL version used for the
connection is the highest version that is supported by both the driver and the server.

The driver provides mechanisms that enable you to authenticate your connection using the
Kerberos protocol, your Hive user name only, or your Hive user name and password. You must
use the authentication mechanism that matches the security requirements of the Hive server. For
information about determining the appropriate authentication mechanism to use based on the
Hive server configuration, see "Authentication Mechanisms" on page 52. For detailed driver
configuration instructions, see "Configuring Authentication on Windows" on page 12 or
"Configuring Authentication on a Non-Windows Machine" on page 43.

62 | Cloudera ODBC Driver for Apache Hive

 Features

Additionally, the driver supports the following types of SSL connections:

l No identity verification
l One-way authentication
l Two-way authentication

It is recommended that you enable SSL whenever you connect to a server that is configured to
support it. SSL encryption protects data and credentials when they are transferred over the
network, and provides stronger security than authentication alone. For detailed configuration
instructions, see "Configuring SSL Verification on Windows" on page 19 or "Configuring SSL
Verification on a Non-Windows Machine" on page 46.

Cloudera ODBC Driver for Apache Hive | 63

Driver Configuration Options

Driver Configuration Options

Driver Configuration Options lists the configuration options available in the Cloudera ODBC Driver
for Apache Hive alphabetically by field or button label. Options having only key names, that is, not
appearing in the user interface of the driver, are listed alphabetically by key name.

When creating or configuring a connection from a Windows machine, the fields and buttons are
available in the Cloudera Hive ODBC Driver Configuration tool and the following dialog boxes:
l Cloudera ODBC Driver for Apache Hive DSN Setup
l Advanced Options
l Server Side Properties
l SSL Options
l HTTP Properties

When using a connection string or configuring a connection from a Linux/macOS/AIX machine,

use the key names provided.

Note:

If you are using the driver on a non-Windows machine, you can set driver configuration
properties in a connection string, in a DSN (in the odbc.ini file), or as a driver-wide setting (in
the cloudera.hiveodbc.ini file). Settings in the connection string take precedence over
settings in the DSN, and settings in the DSN take precedence over driver-wide settings.

Configuration Options Appearing in the User Interface

The following configuration options are accessible via the Windows user interface for the Cloudera
ODBC Driver for Apache Hive, or via the key name when using a connection string or configuring a
connection from a Linux/macOS/AIX/Debian machine:

l "Allow Common Name Host Name l "Invalid Session Auto Recover" on page
Mismatch" on page 65 75
l "Allow Self-Signed Server Certificate" l "Log Level" on page 75
on page 66 l "Log Path" on page 76
l "Apply Properties with Queries" on l "Max File Size" on page 77
page 66
l "Max Number Files" on page 77
l "Binary Column Length" on page 67
l "Mechanism" on page 78
l "Canonicalize Principal FQDN" on page
67 l "Minimum TLS" on page 78
l "Check Certificate Revocation" on page l "Password" on page 78
67 l "Port" on page 79
l "Client Certificate File" on page 68 l "Realm" on page 79
l "Client Private Key File" on page 68 l "Rows Fetched Per Block" on page 79

64 | Cloudera ODBC Driver for Apache Hive

 Driver Configuration Options

l "Client Private Key Password" on page l "Save Password (Encrypted)" on page

68 79
l "Convert Key Name to Lower Case" on l "Service Discovery Mode" on page 80
page 69 l "Service Name" on page 80
l "Data File HDFS Dir" on page 69 l "Show System Table" on page 80
l "Database" on page 70 l "Socket Timeout" on page 81
l "Decimal Column Scale" on page 70 l "Temp Table TTL" on page 81
l "Default String Column Length" on l "Thrift Transport" on page 81
page 70
l "Trusted Certificates" on page 82
l "Delegate Kerberos Credentials" on
page 70 l "Two-Way SSL" on page 82
l "Delegation UID" on page 71 l "Unicode SQL Character Types" on
page 83
l "Driver Config Take Precedence" on
page 71 l "Use Async Exec" on page 83
l "Enable Auto Reconnect" on page 71 l "Use Native Query" on page 84
l "Enable SSL" on page 72 l "Use Only SSPI" on page 84
l "Enable Temporary Table" on page 72 l "Use System Trust Store" on page 84
l "Fast SQLPrepare" on page 73 l "User Name" on page 85
l "Get Tables With Query" on page 73 l "Web HDFS Host" on page 85
l "HDFS User" on page 73 l "Web HDFS Port" on page 85
l "Hive Server Type" on page 74 l "ZooKeeper Namespace" on page 86
l "Host(s)" on page 74
l "Host FQDN" on page 74
l "HTTP Path" on page 75

Allow Common Name Host Name Mismatch

Key Name Default Value Required

CAIssuedCertNamesMismatch Clear (0) No

Description

This option specifies whether a CA-issued SSL certificate name must match the host name of the
Hive server.
l Enabled (1): The driver allows a CA-issued SSL certificate name to not match the host name
of the Hive server.
l Disabled (0): The CA-issued SSL certificate name must match the host name of the Hive
server.

Cloudera ODBC Driver for Apache Hive | 65

Driver Configuration Options

Note:

This setting is applicable only when SSL is enabled.

Allow Self-Signed Server Certificate

Key Name Default Value Required

AllowSelfSigned Clear (0) No

ServerCert

Description

This option specifies whether the driver allows a connection to a Hive server that uses a self-signed
certificate, even if this certificate is not in the list of trusted certificates. This list is contained in the
Trusted Certificates file, or in the system Trust Store if the system Trust Store is used instead of a
file.
l Enabled (1): The driver authenticates the Hive server even if the server is using a self-signed
certificate that has not been added to the list of trusted certificates.
l Disabled (0): The driver does not allow self-signed certificates from the server unless they
have already been added to the list of trusted certificates.

Note:

This setting is applicable only when SSL is enabled.

Apply Properties with Queries

Key Name Default Value Required

ApplySSPWithQueries Selected (1) No

Description

This option specifies how the driver applies server-side properties.

l Enabled (1): The driver applies each server-side property by executing a set
SSPKey=SSPValue query when opening a session to the Hive server.
l Disabled (0): The driver uses a more efficient method for applying server-side properties
that does not involve additional network round-tripping. However, some Hive Server 2
builds are not compatible with the more efficient method.

Note:

When connecting to a Hive Server 1 instance, this option is always enabled.

66 | Cloudera ODBC Driver for Apache Hive

 Driver Configuration Options

Binary Column Length

Key Name Default Value Required

BinaryColumnLength 32767 No

Description

The maximum data length for BINARY columns.

By default, the columns metadata for Hive does not specify a maximum data length for BINARY
columns.

Canonicalize Principal FQDN

Key Name Default Value Required

ServicePrincipal Selected (1) No

Canonicalization

Description

This option specifies whether the Kerberos layer canonicalizes the host FQDN in the server’s
service principal name.
l Enabled (1): The Kerberos layer canonicalizes the host FQDN in the server’s service principal
name.
l Disabled (0): The Kerberos layer does not canonicalize the host FQDN in the server’s service
principal name.

Note:

l This option only affects MIT Kerberos, and is ignored when using Active Directory
Kerberos.
l This option can only be disabled if the Kerberos Realm or KrbRealm key is specified.

Check Certificate Revocation

Key Name Default Value Required

CheckCertRevocation No

Description

This option specifies whether the driver checks to see if a certificate has been revoked while
retrieving a certificate chain from the Windows Trust Store.

This option is only applicable if you are using a CA certificate from the Windows Trust Store (see
"Use System Trust Store" on page 84).

Cloudera ODBC Driver for Apache Hive | 67

Driver Configuration Options

l Enabled (1): The driver checks for certificate revocation while retrieving a certificate chain
from the Windows Trust Store.
l Disabled (0): The driver does not check for certificate revocation while retrieving a certificate
chain from the Windows Trust Store.

Note:

This option is only available on Windows.

Client Certificate File

Key Name Default Value Required

ClientCert None No

Description

The full path to the .pem file containing the client's SSL certificate.

Note:

This setting is applicable only when two-way SSL is enabled.

Client Private Key File

Key Name Default Value Required

ClientPrivateKey None Yes, if two-way SSL

verification is enabled.

Description

The full path to the .pem file containing the client's SSL private key.

If the private key file is protected with a password, then provide the password using the driver
configuration option "Client Private Key Password" on page 68.

Note:

This setting is applicable only when two-way SSL is enabled.

Client Private Key Password

Key Name Default Value Required

ClientPrivateKeyPassword None Yes, if two-way SSL

68 | Cloudera ODBC Driver for Apache Hive

 Driver Configuration Options

Key Name Default Value Required

verification is enabled
and the client's private
key file is protected with
a password.

Description

The password of the private key file that is specified in the Client Private Key File field
(ClientPrivateKey).

Convert Key Name to Lower Case

Key Name Default Value Required

LCaseSspKeyName Selected (1) No

Description

This option specifies whether the driver converts server-side property key names to all lower-case
characters.
l Enabled (1): The driver converts server-side property key names to all lower-case
characters.
l Disabled (0): The driver does not modify the server-side property key names.

Data File HDFS Dir

Key Name Default Value Required

HDFSTempTableDir /tmp/simba No

Description

The HDFS directory that the driver uses to store the necessary files for supporting the Temporary
Table feature.

Note:

Due to a problem in Hive (see https://issues.apache.org/jira/browse/HIVE-4554), HDFS paths

with space characters do not work with versions of Hive prior to 0.12.0.

This option is not applicable when connecting to Hive 0.14 or later.

Cloudera ODBC Driver for Apache Hive | 69

Driver Configuration Options

Database
Key Name Default Value Required

Schema default No

Description

The name of the database schema to use when a schema is not explicitly specified in a query. You
can still issue queries on other schemas by explicitly specifying the schema in the query.

Note:

To inspect your databases and determine the appropriate schema to use, at the Hive command
prompt, type show databases.

Decimal Column Scale

Key Name Default Value Required

DecimalColumnScale 10 No

Description

The maximum number of digits to the right of the decimal point for numeric data types.

Default String Column Length

Key Name Default Value Required

Default 255 No
StringColumnLength

Description

The maximum number of characters that can be contained in STRING columns.

By default, the columns metadata for Hive does not specify a maximum length for STRING
columns.

Delegate Kerberos Credentials

Key Name Default Value Required

DelegateKrbCreds 0 No

70 | Cloudera ODBC Driver for Apache Hive

 Driver Configuration Options

Description

This option specifies whether your Kerberos credentials are forwarded to the server and used for
authentication.

Note:

This option is only applicable when Authentication Mechanism is set to Kerberos

(AuthMech=1).

Delegation UID
Key Name Default Value Required

DelegationUID None No

Description

If a value is specified for this setting, the driver delegates all operations against Hive to the
specified user, rather than to the authenticated user for the connection.

Note:

This option is applicable only when connecting to a Hive Server 2 instance that supports this
feature.

Driver Config Take Precedence

Key Name Default Value Required

DriverConfigTakePrecedence Clear (0) No

Description

This option specifies whether driver-wide configuration settings take precedence over connection
and DSN settings.
l Enabled (1): Driver-wide configurations take precedence over connection and DSN settings.
l Disabled (0): Connection and DSN settings take precedence instead.

Enable Auto Reconnect

Key Name Default Value Required

AutoReconnect Selected (1) Yes

Cloudera ODBC Driver for Apache Hive | 71

Driver Configuration Options

Description

This option specifies whether the driver attempts to automatically reconnect to the server when a
communication link error occurs.
l Enabled (1): The driver attempts to reconnect.
l Disabled (0): The driver does not attempt to reconnect.

Enable SSL
Key Name Default Value Required

SSL Clear (0) No

Description

This option specifies whether the client uses an SSL encrypted connection to communicate with
the Hive server.
l Enabled (1): The client communicates with the Hive server using SSL.
l Disabled (0): SSL is disabled.

SSL is configured independently of authentication. When authentication and SSL are both
enabled, the driver performs the specified authentication method over an SSL connection.

Note:

l This option is applicable only when connecting to a Hive server that supports SSL.
l If you selected User Name as the authentication mechanism, SSL is not available.

Enable Temporary Table

Key Name Default Value Required

EnableTempTable Clear (0) No

Description

This option specifies whether the driver supports the creation and use of temporary tables.
l Enabled (1): The driver supports the creation and use of temporary tables.
l Disabled (0): The driver does not support temporary tables.

Important:

When connecting to Hive 0.14 or later, the Temporary Tables feature is always enabled and you
do not need to configure it in the driver.

72 | Cloudera ODBC Driver for Apache Hive

 Driver Configuration Options

Fast SQLPrepare
Key Name Default Value Required

FastSQLPrepare Clear (0) No

Description

This option specifies whether the driver defers query execution to SQLExecute.
l Enabled (1): The driver defers query execution to SQLExecute.
l Disabled (0): The driver does not defer query execution to SQLExecute.

Note:

When using Native Query mode, the driver executes the HiveQL query to retrieve the result set
metadata for SQLPrepare. As a result, SQLPrepare might be slow. If the result set metadata is
not required after calling SQLPrepare, then enable Fast SQLPrepare.

Get Tables With Query

Key Name Default Value Required

GetTablesWithQuery Clear (0) No

Description

This option specifies whether the driver uses the SHOW TABLES query or the GetTables Thrift API
call to retrieve table names from the database.
l Enabled (1): The driver uses the SHOW TABLES query to retrieve table names.
l Disabled (0): The driver uses the GetTables Thrift API call to retrieve table names.

Note:

This option is applicable only when connecting to a Hive Server 2 instance.

HDFS User
Key Name Default Value Required

HDFSUser hdfs No

Description

The name of the HDFS user that the driver uses to create the necessary files for supporting the
Temporary Tables feature.

This option is not applicable when connecting to Hive 0.14 or later.

Cloudera ODBC Driver for Apache Hive | 73

Driver Configuration Options

Hive Server Type

Key Name Default Value Required

HiveServerType Hive Server 2 (2) No

Description

This option specifies the type of Hive server.

l Hive Server 1 (1): The driver connects to a Hive Server 1 instance.
l Hive Server 2 (2): The driver connects to a Hive Server 2 instance.

Note:

If Service Discovery Mode is enabled, then connections to Hive Server 1 are not supported.

Host(s)
Key Name Default Value Required

Host None Yes

Description

If Service Discovery Mode is disabled, the IP address or host name of the Hive server.

If Service Discovery Mode is enabled, specify a comma-separated list of ZooKeeper servers in the
following format, where [ZK_Host] is the IP address or host name of the ZooKeeper server and
[ZK_Port] is the number of the TCP port that the ZooKeeper server uses to listen for client
connections:
[ZK_Host1]:[ZK_Port1],[ZK_Host2]:[ZK_Port2]

Host FQDN
Key Name Default Value Required

KrbHostFQDN _HOST No

Description

The fully qualified domain name of the Hive Server 2 host.

When the value of Host FQDN is _HOST, the driver uses the Hive server host name as the fully
qualified domain name for Kerberos authentication. If Service Discovery Mode is disabled, then
the driver uses the value specified in the Host connection attribute. If Service Discovery Mode is
enabled, then the driver uses the Hive Server 2 host name returned by ZooKeeper.

74 | Cloudera ODBC Driver for Apache Hive

 Driver Configuration Options

HTTP Path
Key Name Default Value Required

HTTPPath /hive2 if using Windows No

Azure HDInsight Service (6).
/ if using non-Windows
Azure HDInsight Service with
Thrift Transport set to HTTP
(2).

Description

The partial URL corresponding to the Hive server.

The driver forms the HTTP address to connect to by appending the HTTP Path value to the host
and port specified in the DSN or connection string. For example, to connect to the HTTP address
http://localhost:10002/gateway/sandbox/hive/version, you would set HTTP
Path to /gateway/sandbox/hive/version.

Note:

By default, Hive servers use cliservice as the partial URL.

Invalid Session Auto Recover

Key Name Default Value Required

InvalidSessionAutoRecover Selected (1) No

Description

This option specifies whether the driver automatically opens a new session when the existing
session is no longer valid.
l Enabled (1): The driver automatically opens a new session when the existing session is no
longer valid.
l Disabled (0): The driver does not automatically open new sessions.

Note:

This option is applicable only when connecting to Hive Server 2.

Log Level
Key Name Default Value Required

LogLevel OFF (0) No

Cloudera ODBC Driver for Apache Hive | 75

Driver Configuration Options

Description

Use this property to enable or disable logging in the driver and to specify the amount of detail
included in log files.

Important:

l Only enable logging long enough to capture an issue. Logging decreases performance and
can consume a large quantity of disk space.
l The settings for logging apply to every connection that uses the Cloudera ODBC Driver for
Apache Hive, so make sure to disable the feature after you are done using it.
l This option is not supported in connection strings. To configure logging for the Windows
driver, you must use the Logging Options dialog box. To configure logging for a non-
Windows driver, you must use the cloudera.hiveodbc.ini file.

Set the property to one of the following values:

l OFF (0): Disable all logging.
l FATAL (1): Logs severe error events that lead the driver to abort.
l ERROR (2): Logs error events that might allow the driver to continue running.
l WARNING (3): Logs events that might result in an error if action is not taken.
l INFO (4): Logs general information that describes the progress of the driver.
l DEBUG (5): Logs detailed information that is useful for debugging the driver.
l TRACE (6): Logs all driver activity.

When logging is enabled, the driver produces the following log files at the location you specify in
the Log Path (LogPath) property:
l A clouderahiveodbcdriver.log file that logs driver activity that is not specific to a
connection.
l A clouderahiveodbcdriver_connection_[Number].log file for each
connection made to the database, where [Number] is a number that identifies each log file.
This file logs driver activity that is specific to the connection.

Log Path
Key Name Default Value Required

LogPath None Yes, if logging is enabled.

Description

The full path to the folder where the driver saves log files when logging is enabled.

76 | Cloudera ODBC Driver for Apache Hive

 Driver Configuration Options

Important:

This option is not supported in connection strings. To configure logging for the Windows driver,
you must use the Logging Options dialog box. To configure logging for a non-Windows driver,
you must use the cloudera.hiveodbc.ini file.

Max File Size

Key Name Default Value Required

LogFileSize 20971520 No

Description

The maximum size of each log file in bytes. After the maximum file size is reached, the driver
creates a new file and continues logging.

If this property is set using the Windows UI, the entered value is converted from megabytes (MB)
to bytes before being set.

Important:

This option is not supported in connection strings. To configure logging for the Windows driver,
you must use the Logging Options dialog box. To configure logging for a non-Windows driver,
you must use the cloudera.hiveodbc.ini file.

Max Number Files

Key Name Default Value Required

LogFileCount 50 No

Description

The maximum number of log files to keep. After the maximum number of log files is reached, each
time an additional file is created, the driver deletes the oldest log file.

Important:

This option is not supported in connection strings. To configure logging for the Windows driver,
you must use the Logging Options dialog box. To configure logging for a non-Windows driver,
you must use the cloudera.hiveodbc.ini file.

Cloudera ODBC Driver for Apache Hive | 77

Driver Configuration Options

Mechanism
Key Name Default Value Required

AuthMech No Authentication (0) if you No

are connecting to Hive
Server 1. User Name (2) if
you are connecting to Hive
Server 2.

Description

The authentication mechanism to use.

Select one of the following settings, or set the key to the corresponding number:
l No Authentication (0)
l Kerberos (1)
l User Name (2)
l User Name And Password (3)

Minimum TLS
Key Name Default Value Required

Min_TLS TLS 1.2 (1.2) No

Description

The minimum version of TLS/SSL that the driver allows the data store to use for encrypting
connections. For example, if TLS 1.1 is specified, TLS 1.0 cannot be used to encrypt connections.
l TLS 1.0 (1.0): The connection must use at least TLS 1.0.
l TLS 1.1 (1.1): The connection must use at least TLS 1.1.
l TLS 1.2 (1.2): The connection must use at least TLS 1.2.

Password
Key Name Default Value Required

PWD None Yes, if the authentication

mechanism is User Name
And Password (3).

Description

The password corresponding to the user name that you provided in the User Name field (the UID
key).

78 | Cloudera ODBC Driver for Apache Hive

 Driver Configuration Options

Port
Key Name Default Value Required

Port 10000 Yes, if Service Discovery

Mode is disabled.

Description

The number of the TCP port that the Hive server uses to listen for client connections.

Realm
Key Name Default Value Required

KrbRealm Depends on your Kerberos No

configuration.

Description

The realm of the Hive Server 2 host.

If your Kerberos configuration already defines the realm of the Hive Server 2 host as the default
realm, then you do not need to configure this option.

Rows Fetched Per Block

Key Name Default Value Required

RowsFetchedPerBlock 10000 No

Description

The maximum number of rows that a query returns at a time.

Valid values for this setting include any positive 32-bit integer. However, testing has shown that
performance gains are marginal beyond the default value of 10000 rows.

Save Password (Encrypted)

Key Name Default Value Required

N/A Selected No

Description

This option specifies whether the password is saved in the registry.

Cloudera ODBC Driver for Apache Hive | 79

Driver Configuration Options

l Enabled: The password is saved in the registry.

l Disabled: The password is not saved in the registry.

This option is available only in the Windows driver. It appears in the Cloudera ODBC Driver for
Apache Hive DSN Setup dialog box and the SSL Options dialog box.

Important:

The password is obscured (not saved in plain text). However, it is still possible for the encrypted
password to be copied and used.

Service Discovery Mode

Key Name Default Value Required

ServiceDiscoveryMode No Service Discovery (0) No

Description

This option specifies whether the driver uses the ZooKeeper service.
l Enabled (1): The driver discovers Hive Server 2 services via the ZooKeeper service.
l Disabled (0): The driver connects to Hive without using a discovery service.

Service Name
Key Name Default Value Required

KrbServiceName None No

Description

The Kerberos service principal name of the Hive server.

Show System Table

Key Name Default Value Required

ShowSystemTable Clear (0) No

Description

This option specifies whether the driver returns the hive_system table for catalog function calls
such as SQLTables and SQLColumns.
l Enabled (1): The driver returns the hive_system table for catalog function calls such as
SQLTables and SQLColumns.
l Disabled (0): The driver does not return the hive_system table for catalog function calls.

80 | Cloudera ODBC Driver for Apache Hive

 Driver Configuration Options

Socket Timeout
Key Name Default Value Required

SocketTimeout 60 No

Description

The number of seconds that an operation can remain idle before it is closed.

Note:

This option is applicable only when asynchronous query execution is being used against Hive
Server 2 instances.

Temp Table TTL

Key Name Default Value Required

TempTableTTL 10 No

Description

The number of minutes a temporary table is guaranteed to exist in Hive after it is created.

This option is not applicable when connecting to Hive 0.14 or later.

Thrift Transport
Key Name Default Value Required

ThriftTransport Binary (0) if you are No

connecting to Hive Server 1.
SASL (1)
if you are connecting to Hive
Server 2.

Description

The transport protocol to use in the Thrift layer.

Select one of the following settings, or set the key to the number corresponding to the desired
setting:
l Binary (0)
l SASL (1)
l HTTP (2)

Cloudera ODBC Driver for Apache Hive | 81

Driver Configuration Options

Note:

For information about how to determine which Thrift transport protocols your Hive server
supports, see "Authentication Mechanisms" on page 52.

Trusted Certificates
Key Name Default Value Required

TrustedCerts The cacerts.pem file in No

the \lib subfolder within
the driver's installation
directory.
The exact file path varies
depending on the version of
the driver that is installed.
For example, the path for
the Windows driver is
different from the path for
the macOS driver.

Description

The full path of the .pem file containing trusted CA certificates, for verifying the server when using
SSL.

If this option is not set, then the driver defaults to using the trusted CA certificates .pem file
installed by the driver.

Important:

If you are connecting from a Windows machine and the Use System Trust Store option is
enabled, the driver uses the certificates from the Windows trust store instead of your specified
.pem file. For more information, see "Use System Trust Store" on page 84.

Two-Way SSL
Key Name Default Value Required

TwoWaySSL Clear (0) No

Description

This option specifies whether two-way SSL is enabled.

l Enabled (1): The client and the Hive server verify each other using SSL. See also the driver
configuration options "Client Certificate File" on page 68, "Client Private Key File" on page
68, and "Client Private Key Password" on page 68.

82 | Cloudera ODBC Driver for Apache Hive

 Driver Configuration Options

l Disabled (0): The server does not verify the client. Depending on whether one-way SSL is
enabled, the client might verify the server. For more information, see "Enable SSL" on page
72.

Note:

This option is applicable only when connecting to a Hive server that supports SSL. You must
enable SSL before Two Way SSL can be configured. For more information, see "Enable SSL" on
page 72.

Unicode SQL Character Types

Key Name Default Value Required

UseUnicodeSqlCharacterTypes Clear (0) No

Description

This option specifies the SQL types to be returned for string data types.
l Enabled (1): The driver returns SQL_WVARCHAR for STRING and VARCHAR columns, and
returns SQL_WCHAR for CHAR columns.
l Disabled (0): The driver returns SQL_VARCHAR for STRING and VARCHAR columns, and
returns SQL_CHAR for CHAR columns.

Use Async Exec

Key Name Default Value Required

EnableAsyncExec Clear (0) No

Description

This option specifies whether to execute queries synchronously or asynchronously.

l Enabled (1): The driver uses an asynchronous version of the API call against Hive for
executing a query.
l Disabled (0): The driver executes queries synchronously.

Due to a problem in Hive 0.12.0 (see https://issues.apache.org/jira/browse/HIVE-5230), Hive

returns generic error messages for errors that occur during query execution. To see the actual
error message relevant to the problem, turn off asynchronous query execution and execute the
query again.

Note:

This option only takes effect when connecting to a Hive cluster running Hive 0.12.0 or higher.

Cloudera ODBC Driver for Apache Hive | 83

Driver Configuration Options

Use Native Query

Key Name Default Value Required

UseNativeQuery Clear (0) No

Description

This option specifies whether the driver uses native HiveQL queries, or converts the queries
emitted by an application into an equivalent form in HiveQL. If the application is Hive-aware and
already emits HiveQL, then enable this option to avoid the extra overhead of query
transformation.
l Enabled (1): The driver does not transform the queries emitted by an application, and
executes HiveQL queries directly.
l Disabled (0): The driver transforms the queries emitted by an application and converts
them into an equivalent form in HiveQL.

Important:

When this option is enabled, the driver cannot execute parameterized queries.

Use Only SSPI

Key Name Default Value Required

UseOnlySSPI Clear (0) No

Description

This option specifies how the driver handles Kerberos authentication: either with the SSPI plugin
or with MIT Kerberos.
l Enabled (1): The driver handles Kerberos authentication by using the SSPI plugin instead of
MIT Kerberos by default.
l Disabled (0): The driver uses MIT Kerberos to handle Kerberos authentication, and only
uses the SSPI plugin if the GSSAPI library is not available.

Important:

This option is available only in the Windows driver.

Use System Trust Store

Key Name Default Value Required

UseSystemTrustStore Clear (0) No

84 | Cloudera ODBC Driver for Apache Hive

 Driver Configuration Options

Description

This option specifies whether to use a CA certificate from the system trust store, or from a
specified .pem file.
l Enabled (1): The driver verifies the connection using a certificate in the system trust store.
l Disabled (0): The driver verifies the connection using a specified .pem file. For information
about specifying a .pem file, see "Trusted Certificates" on page 82.

Note:

This option is only available on Windows.

User Name
Key Name Default Value Required

UID For User Name (2) Yes, if the authentication

authentication only, the mechanism is User Name
default value is anonymous And Password (3).
No, if the authentication
mechanism is User Name (2).

Description

The user name that you use to access Hive Server 2.

Web HDFS Host

Key Name Default Value Required

WebHDFSHost The Hive server host. No

Description

The host name or IP address of the machine hosting both the namenode of your Hadoop cluster
and the WebHDFS service.

This option is not applicable when connecting to Hive 0.14 or later.

Web HDFS Port

Key Name Default Value Required

WebHDFSPort 50070 No

Description

The WebHDFS port for the namenode.

Cloudera ODBC Driver for Apache Hive | 85

Driver Configuration Options

This option is not applicable when connecting to Hive 0.14 or later.

ZooKeeper Namespace
Key Name Default Value Required

ZKNamespace None Yes, if Service Discovery

Mode is enabled.

Description

The namespace on ZooKeeper under which Hive Server 2 znodes are added.

Configuration Options Having Only Key Names

The following configuration options do not appear in the Windows user interface for the Cloudera
ODBC Driver for Apache Hive. They are accessible only when you use a connection string or
configure a connection from a Linux/macOS/AIX machine:
l "DelegationUserIDCase" on page 86
l "Driver" on page 86
l "HTTPAuthCookies" on page 87
l "http.header." on page 87
l "SSP_" on page 88

DelegationUserIDCase
Key Name Default Value Required

DelegationUserIDCase Unchanged No

Description

This option specifies whether the driver changes the Delegation UID (or DelegationUID) value
to all upper-case or all lower-case. The following values are supported:
l Upper: Change the delegated user name to all upper-case.
l Lower: Change the delegated user name to all lower-case.
l Unchanged: Do not modify the delegated user name.

For more information about delegating a user name, see "Delegation UID" on page 71.

Driver
Key Name Default Value Required

Driver Cloudera ODBC Yes

86 | Cloudera ODBC Driver for Apache Hive

 Driver Configuration Options

Key Name Default Value Required

Driver for Apache

Hive when installed on
Windows, or the absolute
path of the driver shared
object file when installed on
a non-Windows machine.

Description

On Windows, the name of the installed driver (Cloudera ODBC Driver for Apache
Hive).

On other platforms, the name of the installed driver as specified in odbcinst.ini, or the
absolute path of the driver shared object file.

HTTPAuthCookies
Key Name Default Value Required

HTTPAuthCookies hive.server2.auth, No
JSessionID

Description

A comma-separated list of authentication cookies that are supported by the driver.

If cookie-based authentication is enabled in your server, the driver authenticates the connection
once based on the provided authentication credentials. It then uses the cookie generated by the
server for each subsequent request in the same connection.

http.header.
Key Name Default Value Required

http.header None No

Description

Set a custom HTTP header by using the following syntax, where [HeaderKey] is the name of the
header to set and [HeaderValue] is the value to assign to the header:
http.header.[HeaderKey]=[HeaderValue]

For example:
http.header.AUTHENTICATED_USER=john

Cloudera ODBC Driver for Apache Hive | 87

Driver Configuration Options

After the driver applies the header, the http.header. prefix is removed from the DSN entry, leaving
an entry of [HeaderKey]=[HeaderValue]

The example above would create the following custom HTTP header:
AUTHENTICATED_USER: john

Note:

The http.header. prefix is case-sensitive. This option is applicable only when you are using
HTTP as the Thrift transport protocol. For more information, see "Thrift Transport" on page 81.

SSP_
Key Name Default Value Required

SSP_ None No

Description

Set a server-side property by using the following syntax, where [SSPKey] is the name of the server-
side property and [SSPValue] is the value for that property:
SSP_[SSPKey]=[SSPValue]

For example:
SSP_mapred.queue.names=myQueue

After the driver applies the server-side property, the SSP_ prefix is removed from the DSN entry,
leaving an entry of [SSPKey]=[SSPValue].

Note:

l The SSP_ prefix must be upper case.

l When setting a server-side property in a connection string, it is recommended that you
enclose the value in braces ({ }) to make sure that special characters can be properly
escaped.

88 | Cloudera ODBC Driver for Apache Hive

 ODBC API Conformance Level

ODBC API Conformance Level

The following table lists the ODBC interfaces that the Cloudera ODBC Driver for Apache Hive
implements and the ODBC compliance level of each interface.

ODBC compliance levels are Core, Level 1, and Level 2. These compliance levels are defined in the
ODBC Specification published with the Interface SDK from Microsoft.

Interfaces include both the Unicode and non-Unicode versions. For more information, see
"Unicode Function Arguments" in the ODBC Programmer's Reference:
http://msdn.microsoft.com/en-us/library/ms716246%28VS.85%29.aspx.

Conformance Level INTERFACES Conformance Level INTERFACES

Core SQLAllocHandle Core SQLGetStmtAttr

Core SQLBindCol Core SQLGetTypeInfo

Core SQLBindParameter Core SQLNativeSql

Core SQLCancel Core SQLNumParams

Core SQLCloseCursor Core SQLNumResultCols

Core SQLColAttribute Core SQLParamData

Core SQLColumns Core SQLPrepare

Core SQLConnect Core SQLPutData

Core SQLCopyDesc Core SQLRowCount

Core SQLDescribeCol Core SQLSetConnectAttr

Core SQLDisconnect Core SQLSetCursorName

Core SQLDriverconnect Core SQLSetDescField

Core SQLEndTran Core SQLSetDescRec

Core SQLExecDirect Core SQLSetEnvAttr

Core SQLExecute Core SQLSetStmtAttr

Core SQLFetch Core SQLSpecialColumns

Core SQLFetchScroll Core SQLStatistics

Cloudera ODBC Driver for Apache Hive | 89

ODBC API Conformance Level

Conformance Level INTERFACES Conformance Level INTERFACES

Core SQLFreeHandle Core SQLTables

Core SQLFreeStmt Core SQLBrowseConnect

SQLGetConnectAttr
Core Core SQLPrimaryKeys

SQLGetCursorName
Core Core SQLGetInfo

SQLProcedureColumns
Core SQLGetData Level 1

Core SQLGetDescField Level 1 SQLProcedures

Core SQLGetDescRec Level 2 SQLColumnPrivileges

Core SQLGetDiagField Level 2 SQLDescribeParam

Core SQLGetDiagRec Level 2 SQLForeignKeys

Core SQLGetEnvAttr Level 2 SQLTablePrivileges

Core SQLGetFunctions

90 | Cloudera ODBC Driver for Apache Hive

 Contact Us

Contact Us
If you are having difficulties using the driver, our Community Forum may have your solution. In
addition to providing user to user support, our forums are a great place to share your questions,
comments, and feature requests with us.

If you are a Subscription customer you may also use the Cloudera Support Portal to search the
Knowledge Base or file a Case.

Important:

To help us assist you, prior to contacting Cloudera Support please prepare a detailed summary
of the client and server environment including operating system version, patch level, and
configuration.

Cloudera ODBC Driver for Apache Hive | 91