Adam Williamson
Adam Williamson
If you look closely at past blivet-gui updates you can see some of the weirdness here, e.g. https://bodhi.fedoraproject.org/updates/FEDORA-2023-2f3ec99d17 . The gating status went to `waiting` first, then `failed` a couple...
Well, per this ticket description I don't *think* any updates moving to testing are gated. Not even critpath ones. The example update I linked to above - https://bodhi.fedoraproject.org/updates/FEDORA-2023-2f3ec99d17 , which...
While working on https://pagure.io/fedora-ci/general/issue/436 , I found a test that implies there is at least one path on which some kind of push-to-testing gate is implemented: `bodhi-server/tests/consumers/test_signed.py` test `test_consume_from_tag(self)` creates...
No, that doesn't really change this. I was planning to look at this next, if I get time.
note: I found out today that waiving 'missing' results *does* work.
https://github.com/fedora-infra/bodhi/issues/5397#issuecomment-1706981034 would partly address this by not showing running tests as waived even if a waiver is filed.
yeah, I'm not really sure what would be best, was gonna consider all the options.
yeah, I'm aware of that, I think it actually originates in openQA (I contributed some of the definitions there :>). the implementation can get a bit finicky, though. I'm kinda...
Yes, that's how I read it too - but my point is, none of that makes a lot of sense. > yes, the use of `proventester` is odd, I assume...
Man, the scheduling code has gotten a bit complex over time, but reviewing it, it looks like this message - the topic is `bodhi.update.request.testing` , right? - sends us down...