Ran into this trying to test the proposed critpath openQA gating change. It was causing spurious "missing" test results in the decision.

Ouch, nice find.

Use list comprehension:

            return [
                result for result in self.cache[cache_key]
                if result['testcase']['name'] == testcase
            ]

edit: never mind, it is better.

+1 Thanks.

Thanks. @kevin @lholecek can we get the container image updated and this fix deployed to Fedora prod? The update gating proposal is waiting on it.

I can simply tag the new image (with prod-fedora) in quay. [1]

But I have currently problems logging into stage OpenShift instance ("500 - Internal Server Error") to verify if there is no problem before deploying to prod.

[1] https://quay.io/repository/factory2/greenwave?tab=tags

Edit: The stage login issue: https://pagure.io/fedora-infrastructure/issue/9585

Checking the greenwave stage logs; resultsdb stage is down: https://taskotron.stg.fedoraproject.org/resultsdb_api/api/v2.0/results

Any idea what's up?

Not sure. @tflink , @kevin ?

So, there was a proxy issue (it was trying to proxy to the old phx2 stg instance ;) and I fixed that...

Then the db didn't have any contents. I got sidetracked trying to fix that. ;(

If someone else wants to take over great, or I can try and do it soon...

Then the db didn't have any contents.

Fixed https://taskotron.stg.fedoraproject.org/resultsdb/results :)

ping re updating the production image? thanks!

I've updated the prod-fedora image tag (to master-b4a9ada).

❯ curl https://greenwave.stg.fedoraproject.org/api/v1.0/about
{"version":"1.7.1.dev3+git.b4a9ada"}

Thanks a lot!

Thanks. prod has been updated too, and decisions look correct.