... | ... | @@ -961,6 +961,34 @@ Delivered mail 641c797273ba1_86be948d03829@gitlab-02.mail (7.2ms) |
|
|
|
|
|
[issue 139]: https://gitlab.torproject.org/tpo/tpa/gitlab/-/issues/139
|
|
|
|
|
|
### Sidekiq jobs stuck
|
|
|
|
|
|
If merge requests don't display properly, email notifications don't go
|
|
|
out, and, in general, GitLab is being weird, it could be Sidekiq
|
|
|
having trouble. You are likely going to see a `SidekiqQueueSize` alert
|
|
|
that looks like this:
|
|
|
|
|
|
Sidekiq queue default on gitlab-02.torproject.org is too large
|
|
|
|
|
|
The solution to this is unclear. During one incident
|
|
|
([tpo/tpa/team#42218](https://gitlab.torproject.org/tpo/tpa/team/-/issues/42218)), the server was running out of disk space
|
|
|
(but did *not* actually run out completely, it still had about 1.5GB
|
|
|
of disk available), so the disk was resized, GitLab was upgraded, and
|
|
|
the server rebooted a couple of times. Then Sidekiq was able to go
|
|
|
through its backlog in a couple of minutes and service was restored.
|
|
|
|
|
|
Look for lack of disk space, and in all of GitLab's logs:
|
|
|
|
|
|
tail -f /var/log/gitlab/*.log
|
|
|
|
|
|
Try to restart GitLab:
|
|
|
|
|
|
gitlab-ctl restart
|
|
|
|
|
|
... or rebooting the server.
|
|
|
|
|
|
Update this section with future incidents as you find them.
|
|
|
|
|
|
### Gitlab registry troubleshooting
|
|
|
|
|
|
If something goes with the GitLab Registry feature, you should first
|
... | ... | |