Common issues that occur after an upgrade

Authors: UpgradeTroubleshootingTeam
Build basis: CLM 4.x

This page will discuss some troubleshooting steps to take when issues are incurred after the upgrade.

Known issues (application non-specific)

This section will discuss some known issues which occur after the upgrade.

Unable to create an LPA project

Symptom: When creating a new project, you may see errors such as the following:

Error instantiating the template: CRJCA0003E 

 CRJCA0003E The Lifecycle Project Administration (LPA) application made a request that was answered by an unexpected response code. The request was: https://.../ccm/clmSampleProject. The response was: HTTP/1.1 400 Bad Request. 

Cause: Process templates need to be redeployed
Resolution: See Redeploying Predefined Templates for instructions on how to redeploy the templates.

Symptom: Selecting 'save' button on LPA Creation page results in nothing
Environment: CLM in a distributed environment
Cause: One of the applications needs to be authorized
Resolution: Navigate to one of the other links under the admin context (members, projects). At the top of the page, one of your applications should prompt for authorization.

Failing integrations

Symptom: After upgrading your CLM install, the previously configured integrations (OSLC links, synchronization, connectors) are not functioning as they did prior to the upgrade.
Cause:

• Versions are no longer supported
• Configuration is incorrect
• New connector version must be downloaded

Resolution: Before upgrading, ensure that the new versions of the product are supported. Once the upgrade has occured, ensure that the configuration is still accurate. For more information on troubleshooting integrations, see the Integrations Troubleshooting section of the wiki.

Failing Queries or Dashboard query issues

If you are unable to run a query, or see dashboards failing to load, you may need to Reindex the applications.

Failing Reports or Dashboard reports issues

Symptom: After upgrading your CLM install, reports or dashboard reports fail, or show incorrect information. Environment: Apache Derby Database for Data Warehouse
Cause: The Apache Derby Database for Data Warehouse was not moved from the pre-upgrade directory into the new directory. Therefore, the current Data Warehouse database is empty.
*Resolution:* Copy the Apache Derby Data Warehouse location in the pre-upgrade directory (../server/conf/jts/derby/DataWarehouse) to the new directory. Afterwards, complete the Data Warehouse upgrade again by running repotools-jts -upgradeWarehouse.

Users unable to login

Symptom: After upgrading your CLM install, users are unable to login to CLM. The following error may occur

 This user was not found in the directory service. This user will not be able to login unless they have an account in the directory service.ID CRJAZ1532I 

Environment: CLM deployed on Apache Tomcat, using local file based authentication (tomcat-users.xml)
Cause: The file write behaviour for the tomcat-users.xml file was changed between Tomcat 5.5 (RTC 3.x) and Tomcat 7 (RTC 4.x)
Resolution: See Technote 1614661: Tomcat-users.xml is file not updated after Upgrade

Symptom: When accessing RRC (DOORS NG) when JTS has been upgraded to 5.x, users cannot access RRC/DOORS NG
Environment: JTS at 5.0, RRC at 4.x or 5.x
Cause: RRC has changed to DOORS NG.
Resolution: You need to activate the DOORS NG licenses, or apply new licenses. These licenses should already be assigned to your users, but the new CAL needs to be applied.

Unable to load or create work items

Symptom: After upgrade, work items cannot be viewed or created.
Environment: CLM versions 4.0.4 and below
Cause: This is a known defect in 4.0.4
Resolution: See Creating or viewing a work item results in a blank page and not an object error with RTC 4.0.4

Known Issues (application specific)

Navigate to Known Issues and click on the application and version to see known issue and workaround details (specific to application and version).

RTC: Workspaces and Streams are disconnected

Symptom: After upgrade, workspaces and streams are disconnected. You may see a NullPointerException.
Environment: CLM versions 5.0, Oracle backend database
Cause: This is a known defect.
Resolution: See NullPointerExceptions occur with Oracle DB after migration to Rational Team Concert 5.0

Where do I go from here?

If you are unable to resolve your issue using the available online resources, please open a service request with IBM Rational Support. Refer to Additional Troubleshooting Resources for further details.

Recommended data gathering and subsequent analysis steps

When investigating problems that occur after the upgrade, collect the following information:

• ISALite Log collection
• All upgrade repotools log files located in the following directory:

 <Install Directory>/server 

• screenshots or relevant errors

It will also be helpful to gather the following information:

• What was the behaviour like before the upgrade?
• Besides the upgrade, did any other configuration change occur?

Related topics:

• For more troubleshooting topics, refer to Upgrade Troubleshooting.

External links:

• None

Additional contributors: StephanieBagot