#10146 Inconsistent state in koji (build complete, task failed)
Closed: Insufficient data 2 years ago by kevin. Opened 2 years ago by ellert.

  • Describe the issue

The build https://koji.fedoraproject.org/koji/buildinfo?buildID=1758546 (root-6.22.08-5.fc34) has ended up in an inconsistent state. It says "State complete", but the associated build https://koji.fedoraproject.org/koji/taskinfo?taskID=69245201 says "State failed".

The build is also not tagged as an f34-candidate.

The error for the failed build says

GenericError: Build already exists (id=1758546, state=COMPLETE): {'name': 'root', 'version': '6.22.08', 'release': '5.fc34', 'epoch': None, 'task_id': 69245201, 'source': 'git+https://src.fedoraproject.org/rpms/root.git#c4fdaa58588eda8dea0b6216d23cdd0bfb1ddfb6', 'extra': '{"source": {"original_url": "git+https://src.fedoraproject.org/rpms/root.git#c4fdaa58588eda8dea0b6216d23cdd0bfb1ddfb6"}}', 'owner': 741, 'state': 0, 'completion_time': None, 'pkg_id': 10727, 'start_time': 'NOW', 'volume_id': 0}

I am not sure how to read that, but it looks like a self-reference: The task of build 1758546 can not be completed, because build 1758546 is already completed?

  • When do you need this? (YYYY/MM/DD)

  • When is this no longer needed or useful? (YYYY/MM/DD)

  • If we cannot complete your request, what is the impact?


I suspect that the failure is a result of the outage last week. @kevin can you please confirm, or bring some light to this?

This was actually before any outage... but it's definitely a very weird state. ;(

@tkopecek any ideas here?

I mean short term, just bump and rebuild it... but this seems like a bug that it could get in this state.

Metadata Update from @mohanboddu:
- Issue tagged with: low-gain, low-trouble, ops

2 years ago

Not sure how to replicate this. I would incline to some builder/hub restart. You can double-check if there was no other buildArch task working on same build (and find parent for that).

SELECT id, arch FROM task WHERE method='buildArch' AND request LIKE '%build%root-6.22.08-5.fc34%';

id | arch
----+------
(0 rows)

I guess we just close this and see if it happens again? Or anything else we can check?

Metadata Update from @kevin:
- Issue close_status updated to: Insufficient data
- Issue status updated to: Closed (was: Open)

2 years ago

Login to comment on this ticket.

Metadata
Boards 1
Ops Status: Backlog