Failed the application diagnostics during the Engineering Lifecycle Management instance upgrade with a reverse proxy in the deployment topology

When you upgrade to a major or interim fix version of the Engineering Lifecycle Management instance by using the reverse proxy in the deployment topology, the upgrade might fail with the following error. The Applications Events table shows the message Application diagnostics has failed with the reason code AppDiagnosticsErr. Review the following troubleshooting information to resolve the error.

Problem

  1. In the Red Hat® OpenShift® Container Platform web console, go to Operators > Installed Operators and select the Engineering Lifecycle Management operator.
  2. Go to the ELM tab and select the Engineering Lifecycle Management instance.
  3. On the Details tab, verify the Engineering Lifecycle Management application status in the Applications Events table. It shows the message as Application diagnostics has failed with the reason code AppDiagnosticsErr.
  4. Review the Engineering Lifecycle Management application log for any of the following error messages. For more information about accessing the application pod logs, see Accessing application pod logs.
    The server returned the HTTP error 504 with this error text: Gateway Time-out.
    The server returned the HTTP error 502 with this error text: Bad Gateway

    The timeout parameter configurations in the external reverse proxy server that is required in the Red Hat OpenShift topology might cause these error messages.

Resolution

To resolve the timeout errors with the external reverse proxy server, the timeout parameters in the proxy configurations are to be updated.

If you use the HaProxy as an external reverse proxy, adjust the following parameters.
timeout client
timeout server

The timeout values of these parameters depend on the factors such as network latency, database load, and database performance. You must consider these factors when you adjust the value for the parameters.

An example with the sample values:

timeout client 5m
timeout server 5m

For more information about the parameters in HaProxy, see timeout client and timeout server.

When the timeout parameters are updated, resume the upgrade process by running the following command in the terminal.
/elm/server/resolveUpgradeErr.sh REPOTOOLS_DIAGNOSTICS
If the previous step did not resolve the error, complete the following steps to resolve the error.
  1. Review the Engineering Lifecycle Management application log /elm/server/conf/<elm-application-name>/logs/repotools-<elm-application-name>_executeDiagnostics.log to identify the cause of the failure. For more information about accessing the application pod logs, see Accessing application pod logs.
    Note: For more information about executeDiagnostics command, see Repository tools command to run diagnostics on the server.
  2. When the failure cause is resolved, run the following command to resume the upgrade process.
    /elm/server/resolveUpgradeErr.sh REPOTOOLS_DIAGNOSTICS

    The upgrade process resumes with the application diagnostics stage. The application diagnostics event is updated as Completed in the Upgrade Events table.

  3. After the upgrade is completed, go to the 'Server Administration' page in Jazz® Team Server application and run the application diagnostics manually. If you see any error message that is related to service unreachable or gateway timeouts, refresh the web page to get the updated results.