Troubleshooting Taskomatic

Repository metadata regeneration is a relatively intensive process, so Taskomatic can take several minutes to complete. Additionally, if Taskomatic crashes, repository metadata regeneration can be interrupted.

To access a shell inside the Server container run mgrctl term on the container host.

If Taskomatic is still running, or if the process has crashed, package updates can seem available in the Web UI, but do not appear on the client, and attempts to update the client fail. In this case, the zypper ref command shows an error like this:

Valid metadata not found at specified URL

To correct this, determine if Taskomatic is still in the process of generating repository metadata, or if a crash could have occurred. Wait for metadata regeneration to complete or restart Taskomatic after a crash in order for client updates to be carried out correctly.

Procedure: Resolving Taskomatic Problems

  1. On the Uyuni Server, check the /var/log/rhn/rhn_taskomatic_daemon.log file to determine if any metadata regeneration processes are still running, or if a crash occurred.

  2. Restart taskomatic:

    service taskomatic restart

  3. In the Taskomatic log files, you can identify the section related to metadata regeneration by looking for opening and closing lines that look like this:

    <YYYY-DD-MM> <HH:MM:SS>,174 [Thread-584] INFO  com.redhat.rhn.taskomatic.task.repomd.RepositoryWriter - Generating new repository metadata for channel 'cloned-2018-q1-sles12-sp3-updates-x86_64'(sha256) 550 packages, 140 errata

...

<YYYY-DD-MM> <HH:MM:SS>,704 [Thread-584] INFO  com.redhat.rhn.taskomatic.task.repomd.RepositoryWriter - Repository metadata generation for 'cloned-2018-q1-sles12-sp3-updates-x86_64' finished in 4 seconds