#10776 kojipkgs.fedoraproject.org returning 503 errors
Closed: Fixed with Explanation 2 years ago by kevin. Opened 2 years ago by rjones.

Koji is not quite working at the moment.

If you look at RPMs or build logs you get a 503, eg:
https://kojipkgs.fedoraproject.org//work/tasks/1597/88451597/build.log

Also this seems to affect all new builds which fail with creating the SRPM, eg:

$ fedpkg build --target=f37-build-side-54560
Using coq.spec
Building coq-8.15.2-1.fc37 for f37-build-side-54560
Created task: 88455061
Task info: https://koji.fedoraproject.org/koji/taskinfo?taskID=88455061
Watching tasks (this may be safely interrupted)...
88455061 build (f37-build-side-54560, /rpms/coq.git:04c47e99afd22c747737056a418588c3ed7b2fa7): free
88455061 build (f37-build-side-54560, /rpms/coq.git:04c47e99afd22c747737056a418588c3ed7b2fa7): free -> FAILED: BuildrootError: could not init mock buildroot, mock exited with status 30; see root.log for more information
  0 free  0 open  0 done  1 failed
  88455064 buildSRPMFromSCM (/rpms/coq.git:04c47e99afd22c747737056a418588c3ed7b2fa7): FAILED: BuildrootError: could not init mock buildroot, mock exited with status 30; see root.log for more information

88455061 build (f37-build-side-54560, /rpms/coq.git:04c47e99afd22c747737056a418588c3ed7b2fa7) failed

Yeah, kojipkgs01/02 both oom killed varnish. We need to figure out why that happened, but I have restarted them now... so everything should be back to working.

I am going to leave this open to track the underlying issue.

Can confirm that whatever you did has fixed the issue. I'll leave the bug open as you wish.

Koji seems rather slow and builds don't seem to be starting for me currently

Thats likely me trying to tag the f37-python builds into f37... it's hammering things pretty hard and also firing off eln builds. ;)

Metadata Update from @phsmoura:
- Issue priority set to: Waiting on Assignee (was: Needs Review)
- Issue tagged with: high-gain, high-trouble, ops

2 years ago

I set varnish to restart on failure here, so that should take care of it at least somewhat.

Metadata Update from @kevin:
- Issue close_status updated to: Fixed with Explanation
- Issue status updated to: Closed (was: Open)

2 years ago

Login to comment on this ticket.

Metadata
Boards 1
ops Status: Backlog