From matrix:
@music Is there anything odd going on with koji / kojira right now? This side-tag build has been in waitrepo state for almost an hour: https://koji.fedoraproject.org/koji/taskinfo?taskID=124248491 @decathorpe looks like the last successful newRepo tasks finished ~8 hours ago: https://koji.fedoraproject.org/koji/tasks?state=all&view=tree&method=newRepo&order=-id others seem stuck. @churchyard I mostly get Service Unavailable
Now I get Gateway Timeout.
Could you please investigate? Thanks
today
Metadata Update from @zlopez: - Issue priority set to: Waiting on Assignee (was: Needs Review) - Issue tagged with: Needs investigation, koji
Metadata Update from @zlopez: - Issue assigned to zlopez
Not sure why, but the db-koji01 was under heavy load, after restarting it all the nagios alerts are gone.
@churchyard Could you confirm that the issue is solved?
Hopefully due to restarting but my build completed with Could not execute build: database outage.
Could not execute build: database outage
Thanks. I can now access the Koji web interface. The waitrepos are still in progress, will check in a bit.
@hobbes1069 This probably happened even before the restart as the database wasn't responding under heavy load.
But you should be able to restart the build now.
https://koji.fedoraproject.org/koji/tasks?state=all&view=tree&method=newRepo&order=-id still seem stuck. Should we cancel the old jobs?
I'm not sure if the koji will catch up with them. It will be better to start them again.
freetasking them works. I'll do that
https://koji.fedoraproject.org/koji/tasks?state=all&view=tree&method=newRepo&order=-id is back in business.
Thanks again.
Metadata Update from @churchyard: - Issue close_status updated to: Fixed - Issue status updated to: Closed (was: Open)
Log in to comment on this ticket.