Troubleshooting DevOps Deploy when running as a container

When you install HCL DevOps Deploy (Deploy) server by using the provided Helm chart, two containers are created. The first is named init that performs some initialization steps, including initialization of your database. The second is hcl-launch-prod, which is the main server container. You can debug issues by checking the logs of each of these containers.

Checking the init container

  • To get logging information for the init container, you must run the following command:
    kubectl logs [pod-name] -c init
  • A part of the initialization process involves connecting to the database and pulling the Deploy server version information from the database. If you are installing a new release of Deploy with a new database, the database is not initialized and does not contain a Deploy version. In this case, you might get the following error message: 
    [Error] Table DS_DATABASE_VERSION does not exist

    After this error message, you see another message indicating the database has been initialized. It might take some time for the database to initialize.

    When you connect to a previously initialized database, the version of the Deploy server used to initialize the database is displayed.

Checking the server container (hcl-launch-prod)

After the init container startup finishes, the main server container starts.

  • You can view the installation and setup logs by running the following command:
    kubectl logs [pod-name] -c hcl-launch-prod
  • After the container has started, you can run the following command inside the container: 
    kubectl exec -it [pod-name] -- /bin/bash

    The Deploy server is installed in the /opt/hcl-launch/server directory. You can navigate to the directory and explore settings and log files as you do with an on-premises Deploy.