Very interesting -- a test failed, but I can't reproduce it locally. Seems like there is some sort of flakeyness

I've restarted the test suite. Let's see.

@wernerfred any way that I can get access to re-run or cancel actions stuff? I think I already asked that, but I wanted to double check.

@wernerfred any way that I can get access to re-run or cancel actions stuff? I think I already asked that, but I wanted to double check.

if you did so, i missed it!

imho nothing speaks against "promoting" you to be a maintainer which would include that permission. Another way would be to create a new team with write permission to the repository (kind of between triage and maintainer from a permission perspective) if you prefer not to activly contribute as a maintainer but are slowed down in your contributions through the current permission level.

cc @georglauterbach

@wernerfred any way that I can get access to re-run or cancel actions stuff? I think I already asked that, but I wanted to double check.

if you did so, i missed it!

imho nothing speaks against "promoting" you to be a maintainer which would include that permission. Another way would be to create a new team with write permission to the repository (kind of between triage and maintainer from a permission perspective) if you prefer not to activly contribute as a maintainer but are slowed down in your contributions through the current permission level.

cc @georglauterbach

Maintainer works for me. I don't see myself ever not using this repo to host my personal mail server which means I'll always be contributing. :)

I'll probably be doing some stress testing on the test suite too since the more I use it the more I see random failures that don't reoccur, so this is perfect time to be talking about this.

@wernerfred I'm fine with @NorseGaud becoming a maintainer :)

We should however keep the "write" permission level in mind - maybe it could come in handy some time in the future 👍🏼

All set 🚀

All set 🚀

Gracias, bros!

Tests are definitely flakey, and it's usually the same ones. I'll keep stress testing them and see if I can figure out why it's failing.

Tested just the set of tests around what is failing and can't get it to fail. Seems like it might be related to other tests interacting with these. Expanding the testing I'm doing to see if I can track down what specifically is require to get it to fail. I'd say we can move this to WIP

@georglauterbach I see we have a milestone for 10.2.0. Is it ok if I put it in there?

LGTM 👍🏼

Very nice additions!
I think this PRs paves the way for bringing some of the logic of setup.sh in the container itself to make setup.sh independent of the mail server version again. I will have a look after this PR is merged :D

Thanks!

The tests seem not to finish 😕

The tests seem not to finish 😕

Sorry, can you clarify what you're seeing?

The build-and-test action took incredible long (1h 28m 58s), but have finished successful 👍

The build-and-test action took incredible long (1h 28m 58s), but have finished successful 👍

Got it -- I see now.

First that that's happened. I'll add some debug output to the tests and see if we can reproduce it.

Hmm, it didn't happen before the commit you made. I wonder how much of this is the PR and not just a random/rare situation.

Can't reproduce it... Weird...

Glitch in the matrix or our CI was just busy with other stuff 😉

Glitch in the matrix or our CI was just busy with other stuff 😉

@NorseGaud can this be rebased and merged?

@NorseGaud can this be rebased and merged?

Yep, it's ready. Are we doing 10.2.0?

I drafted v10.2.0, but the milestone has one or two PRs I'd like to see merged :) I will merge this and test :edge until v10.2.0 is released.