So this was never submitted for/in testing for some reason?

Perhaps bodhi needs to handle this case? Either force it to go testing first, or add the needed tags when it is submitted for stable in this case?

CC: @adamwill @mattia

Well, it got submitted to stable before it ever made it to testing. But...that happens fairly often, and should work.

It's odd. The other two packages in the update - python-pymdown-extensions-10.12-1.fc41 and mkdocs-material-9.5.48-2.fc41 - have three tags: f41-build-side-101966 , f41-updates-candidate and f41-updates-testing-pending . The fact that they have f41-updates-candidate means they would pass this check (let's disregard for now that the list of tags it checks against looks...weird). It's only python-paginate-0.5.6-1.fc41 which has the wrong tags. From the message it looks like at the time the push was attempted it had tags 'f41-updates-pending' and 'f41-build-side-101966'. Now, it only has f41-build-side-101966 .

Hum. So this is a side tag update that includes a newpackage. Maybe the particular circumstance of a side tag update containing a new package being submitted for stable before reaching testing triggers a bug in the tagging flow somewhere?

I've tagged the package just to keep this moving, but we should try and fix it so it doesn't happen again.

Comparing the history of two builds, seems that python-paginate has never been through the signing process:

$ koji list-history --build=python-paginate-0.5.6-1.fc41
Thu Dec 12 12:37:11 2024 python-paginate-0.5.6-1.fc41 tagged into f41-build-side-101966 by dcavalca [still active]
Thu Dec 12 15:48:00 2024 python-paginate-0.5.6-1.fc41 tagged into f41-updates-pending by bodhi
Fri Dec 13 01:14:06 2024 python-paginate-0.5.6-1.fc41 untagged from f41-updates-pending by bodhi
Sat Dec 14 01:31:19 2024 python-paginate-0.5.6-1.fc41 tagged into f41-updates-candidate by kevin [still active]
$ koji list-history --build=mkdocs-material-9.5.48-2.fc41
Thu Dec 12 15:35:55 2024 mkdocs-material-9.5.48-2.fc41 tagged into f41-build-side-101966 by dcavalca [still active]
Thu Dec 12 15:39:47 2024 mkdocs-material-9.5.48-2.fc41 tagged into f41-signing-pending by bodhi
Thu Dec 12 15:39:51 2024 mkdocs-material-9.5.48-2.fc41 tagged into f41-updates-candidate by bodhi [still active]
Thu Dec 12 15:40:47 2024 mkdocs-material-9.5.48-2.fc41 untagged from f41-signing-pending by autosign/autosign02.iad2.fedoraproject.org
Thu Dec 12 15:40:47 2024 mkdocs-material-9.5.48-2.fc41 tagged into f41-updates-testing-pending by autosign/autosign02.iad2.fedoraproject.org [still active]
Thu Dec 12 15:47:56 2024 mkdocs-material-9.5.48-2.fc41 tagged into f41-updates-pending by bodhi
Fri Dec 13 01:14:06 2024 mkdocs-material-9.5.48-2.fc41 untagged from f41-updates-pending by bodhi

That's weird, Bodhi uses a Koji multiCall to submit all builds for an update to signing, so I think something happened Koji side.
However, I have been unable to look at the related Koji task. All I can say it's that the task didn't fail, searching for the history of completed task returns a timeout...

Should I manually "push to testing" this update to get it moving? I don't want to get in the way of your investigation.

Should I manually "push to testing" this update to get it moving? I don't want to get in the way of your investigation.

IMO, please do so. I don't want to block the update waiting for that as I don't know how can I look deeper into it. Unless Kevin or Adam say different.

Yeah, sorry, I thought I had already when I retagged it. ;(

Done now.