2024-10-24T01:43:14.729Z

<Milind Changire> will take a look
fyi - @Venky Shankar

2024-10-24T15:00:38.059Z

<gregsfortytwo> logging (thread IDs) is broken right now: <https://tracker.ceph.com/issues/68691> due to <https://github.com/ceph/ceph/pull/60058#issuecomment-2434164429>
Squid backport <https://github.com/ceph/ceph/pull/60279> got merged really fast so it’s also busted
(CCs: @Kamoltat (Junior) Sirivadhna @yuriw @Laura Flores)

2024-10-24T15:05:27.502Z

<Laura Flores> Ok. Junior and I were asked to review this in the rados suite, and we didn't see any issues. Was this bug caught in a teuthology test?

@Venky Shankar @Milind Changire can you guys take a look?

2024-10-24T15:06:33.032Z

<gregsfortytwo> well, I don’t think our debug log output is covered in any of the automated tests 🤷‍♂️

2024-10-24T15:06:46.858Z

<gregsfortytwo> so Patrick was looking at logs for a different change and saw that things didn’t look right

2024-10-24T15:06:47.312Z

<Laura Flores> Oh I see the teuthology link in the tracker. But yeah looks like it didn't fail a test

2024-10-24T15:07:04.071Z

<Laura Flores> I see.

2024-10-24T15:15:05.460Z

<Brett Niver> do I detect new test opportunities?

2024-10-24T15:16:30.319Z

<gregsfortytwo> it might be kind of silly since they are **debug** logs and likely to change whenever any relevant code does, but we could add some regexes that you need to deliberately update I guess?

2024-10-24T15:19:15.636Z

<Brett Niver> yeah, there could be some kind of deliberate/fixed output to check

2024-10-24T15:19:35.034Z

<Brett Niver> at least that would be minimal proof that something works

2024-10-24T15:20:26.166Z

<yuriw> It was reviewed and approved 🤷‍♂️

2024-10-24T15:21:32.984Z

<Laura Flores> I requested Radek's review for when he's back from vacation.

2024-10-24T15:21:35.409Z

<gregsfortytwo> it’s an fyi, don’t cut a surprise squid release @yuriw 😉

2024-10-24T15:22:14.502Z

<yuriw> no plans for it as of now 🙂

2024-10-24T15:43:41.214Z

<Rost Khudov> Hello people, I have a question
I am running Ceph unit tests by running following command:
```which cmake3
ctest3 -j3 --rerun-failed --output-on-failure```
but looks like because of parallel running some of the tests are interfiring with each other and failing, but works fine on the second run
Example of failed test:
```129 - unittest_mon_monmap (Failed)```
Is there something that I can do with it?

2024-10-24T16:39:24.958Z

<Æmerson> Do we have a new mandatory PR check that doesn't work?

2024-10-24T16:40:08.860Z

<Æmerson> Because we already have a lot of mandatory PR checks and I'd like fewer.

2024-10-24T17:03:15.787Z

<rzarzynski> Hi @Patrick Donnelly! I was looking on <https://github.com/ceph/ceph/pull/60058/files#diff-12955fa50220234ca6a6ac45287d69871dc082c446b54ad544af715195308e86> and I don't see where it's using already freed memory

2024-10-24T17:06:55.370Z

<Milind Changire> @rzarzynski The EventVector in Log::dump_recent() which moves the events goes out of scope and the `const char *` copied from the Events remain dangling

2024-10-24T17:13:15.979Z

<rzarzynski> ouch, you meant:
```  {
    EntryVector t;
    t.insert(t.end(), std::make_move_iterator(m_recent.begin()), std::make_move_iterator(m_recent.end()));```
I see now, my bad

2024-10-24T17:13:25.011Z

<Milind Changire> yeah, that

2024-10-24T17:13:51.600Z

<rzarzynski> thanks

2024-10-24T17:36:34.420Z

<Joseph Mundackal> should we fix the PR check? (**assuming its useful)

2024-10-24T17:37:46.316Z

<Æmerson> Fix yes. But I'd like these things to not be made required until they've been in operation for a few months.

2024-10-24T17:38:04.965Z

<Æmerson> Just so we don't end up with more required fixes that have to be rerun/whatever constantly

2024-10-24T17:38:16.430Z

<Æmerson> Just so we don't end up with more required checks that have to be rerun/whatever constantly.

2024-10-24T17:38:46.277Z

<Joseph Mundackal> ah - that makes perfect sense 🙂

2024-10-24T17:54:05.367Z

<Hutch> Hey im running into this bug issue, is there any specific MGR channel i should post this in and is there any logs or inof i can provide that might be helpfull

2024-10-24T17:54:17.670Z

<Hutch> Hey im running into this bug issue, is there any specific MGR channel i should post this in and is there any logs or info i can provide that might be helpful
<https://tracker.ceph.com/issues/68550>

2024-10-24T19:17:35.149Z

<gregsfortytwo> that’s wild. I see Neha commented an hour ago, so get the info she asked for there?