https://github.com/osquery/osquery logo
Join Slack
Powered by
Re-triggering the CI by open and closing doesn't s...
# core
z
Re-triggering the CI by open and closing doesn't seem to fix the issue, even though the UI shows that it's now looking for a 
macos-12
runner. I'm going to try rebasing a PR.
s
If you refer to the apparent hanging, that's because closing/reopening a lot of PRs uses all the available jobs
And PR's jobs don't finish all sequentially by PR
also although we have I think 10 or 20 parallel jobs that we can run, the parallel jobs per OS are lower (and I think macOS is the lowest one), might be 4 max
which it seems what's still running now
s
I don’t love it, but we can also merge over CI issues
z
Ah okay so maybe it will self resolve
s
yeah it's just not enough runners available. test jobs are running
s
I think Stefano is correct — https://github.com/organizations/osquery/settings/actions/hosted-runners says we have a limit of 5 macOS workers, and they’re all doing stuff
s
yeah the weirdness happens because when you start let's say 10PRs runs, it first enqueues all the immediately starting jobs, in order of the PRs (when you have opened and closed them). Then since there's a dependency for the test job, those are not taking a slot. So first all of the main jobs have to finish, and only then the test jobs can run.
3 Views
Open in Slack
PreviousNext
Powered by