On Fri, 2025-07-11 at 08:50 -0600, Justin Forbes wrote: > > > > The openQA failures come along with a ton of warning kernel traces > > about DRM stuff, which are apparently fixed by > > https://lore.kernel.org/dri-devel/663110e1-3aa0-4f6f-8727-3a240bc96075@xxxxxxxxxx/T/#me2a0f3254a57cdd157ba79344ab8d51476ba18e5 > > , but I don't know if fixing those tracebacks will actually fix the > > test failures, or if that's some other issue. Also, that patch is not > > yet merged to mainline, so just doing a newer kernel snapshot build > > won't pull it in. > > I have been tracking that patch and it has been merged to linux-next. > I have it sitting here (and currently in rawhide) waiting for a build > to test it. > > > Kernel builds were failing anyway due to a dracut bug - > > https://bugzilla.redhat.com/show_bug.cgi?id=2379116 . I've just > > bisected that and sent a dracut build with the offending commit > > reverted, so we can build kernels again, but I'm not going to try and > > do any kernel builds (I can't anyway as I'm not on the SB signing > > list). > > I tried a kernel build this morning and we are still getting dracut > 107-1 in the build root. Yeah, it was a catch-22 - the dracut update failed tests because of the kernel problems :/ To catch everyone else up, I waived the failures so the dracut update went stable and Justin did a new kernel build, but sadly it doesn't seem to fix the problems. My graphics card workaround didn't actually fix them either. We're trying to figure out what the heck is going on with the failures but not getting anywhere fast. The tests that frequently fail on updates are kiwi image builds, podman tests, and FreeIPA tests (but a lot of other tests failed on the compose - https://openqa.fedoraproject.org/tests/overview?distri=fedora&version=Rawhide&build=Fedora-Rawhide-20250711.n.0&groupid=1 - and I haven't had time to check through all of them yet). There's no obvious common thread between all of those, and nothing juicy in the logs. Operations just seem to stall or fail for no immediately obvious reason. There is one really gross hack up my sleeve - I can build the last working kernel with an epoch bump and force it into openQA tests with the 'workaround' mechanism - but that's kind of a last resort. -- Adam Williamson (he/him/his) Fedora QA Fedora Chat: @adamwill:fedora.im | Mastodon: @adamw@xxxxxxxxxxxxx https://www.happyassassin.net -- _______________________________________________ devel mailing list -- devel@xxxxxxxxxxxxxxxxxxxxxxx To unsubscribe send an email to devel-leave@xxxxxxxxxxxxxxxxxxxxxxx Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/ List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/devel@xxxxxxxxxxxxxxxxxxxxxxx Do not reply to spam, report it: https://pagure.io/fedora-infrastructure/new_issue
