Due to https://bugzilla.redhat.com/show_bug.cgi?id=1403343 , I'm experimenting with video adapters other than 'qxl' on our staging openQA deployment. I first tried 'virtio', but it ran into a problem. In several tests, what looks like the color scheme from the Plymouth bootsplash persists on consoles, which makes openQA's screen match fail - the color scheme isn't reset to 'light grey on absolute black' but remains 'white on dark grey': https://openqa.stg.fedoraproject.org/tests/63643#step/_console_wait_login/6 https://openqa.stg.fedoraproject.org/tests/63631#step/disk_guided_encrypted_postinstall/7 https://openqa.stg.fedoraproject.org/tests/63630#step/_console_wait_login/9 this doesn't ever happen with the 'qxl' driver. It doesn't *always* fail with the virtio driver, in some tests it works, but it went wrong in several of them. The tests have videos too, e.g. https://openqa.stg.fedoraproject.org/tests/63643/file/video.ogv .
gerd, any thoughts on if this is something on the qemu or kernel side?
(In reply to Cole Robinson from comment #1) > gerd, any thoughts on if this is something on the qemu or kernel side? I'd bet on guest userspace, color is terminal settings, resetting is just sending ESC[0m I guess. Any chance the difference is just the qxl driver being in the initramfs and virtio not? I've seen plymouth doing strange things now and then in case the gfx driver is loaded late (i.e. while plymouth is running already).
If it helps, the next thing I tried after virtio is std, and I've seen what looks like the same thing a few times with std, though it doesn't seem to happen as often. I'm not sure what you mean by 'the qxl driver being in the initramfs'. *What* driver are we talking about here? No X drivers are in the initramfs, and the only kernel video-related drivers I can see are i915 and lcd. Please let me know what info I can provide to help :) Thanks.
(In reply to Adam Williamson from comment #3) > If it helps, the next thing I tried after virtio is std, and I've seen what > looks like the same thing a few times with std, though it doesn't seem to > happen as often. > > I'm not sure what you mean by 'the qxl driver being in the initramfs'. > *What* driver are we talking about here? No X drivers are in the initramfs, > and the only kernel video-related drivers I can see are i915 and lcd. kms drivers: qxl.ko, bochs-drm.ko (for std), virtio-gpu.ko > Please let me know what info I can provide to help :) Thanks. Is it possible interact with the guest in "failed" state? login, run /usr/bin/reset, see how the terminal looks then? The first case linked above clearly is kms driver loaded while plymouth runs. The third case seems to be the same, only with ovmf instead of seabios. The second case is really puzzling though: The display adapter apparently is still in vga text mode, so no kms driver loaded yet (which matches your observation that the initramfs has the i915 driver only). I have absolutely no clue why the qemu vga text mode emulation should behave differently depending on the display adapter, they all use the very same code for text mode ...
OK, so I may have missed one thing here: both the 'second case' and the 'same thing a few times with std' may not actually be the same thing at all. Those cases may rather be some different behaviour in plymouth, showing a text decryption prompt instead of a graphical one. I've just noticed that the test we have which starts from an encrypted F25 install and tries to upgrade it *always* fails at a screen which looks just like the 'second case', and that's the thing I've seen with 'std' as well. So you can probably ignore the claim that it affects 'std' as well (it doesn't seem to) and the 'second case'.
> > Please let me know what info I can provide to help :) Thanks. > > Is it possible interact with the guest in "failed" state? > login, run /usr/bin/reset, see how the terminal looks then? Question still stands.
Ah, sorry, I missed that. I should be able to do it one way or another, yeah. I'll try and get you that info soon.
This bug appears to have been reported against 'rawhide' during the Fedora 26 development cycle. Changing version to '26'.
well, I just tried like 15 times to reproduce this manually and couldn't, and we switched over to 'std' for openqa graphics a while back and that's been working fine. So I'm gonna close this as INSUFFICIENT_DATA for now.
I had to switch our UEFI tests back to virtio temporarily as UEFI boots with 'std' graphics currently fail to work in Rawhide, and this is still happening, e.g. https://openqa.fedoraproject.org/tests/181868 . Haven't tried to reproduce manually yet, but I'll see if I can do something about it tomorrow.
This definitely still happens, a lot, in openQA. So I'm re-opening it. Still haven't had time to try and reproduce manually.
Fedora 26 changed to end-of-life (EOL) status on 2018-05-29. Fedora 26 is no longer maintained, which means that it will not receive any further security or bug fix updates. As a result we are closing this bug. If you can reproduce this bug against a currently maintained version of Fedora please feel free to reopen this bug against that version. If you are unable to reopen this bug, please file a new report against the current release. If you experience problems, please add a comment to this bug. Thank you for reporting this bug and we are sorry it could not be fixed.
This message is a reminder that Fedora 28 is nearing its end of life. On 2019-May-28 Fedora will stop maintaining and issuing updates for Fedora 28. It is Fedora's policy to close all bug reports from releases that are no longer maintained. At that time this bug will be closed as EOL if it remains open with a Fedora 'version' of '28'. Package Maintainer: If you wish for this bug to remain open because you plan to fix it in a currently maintained version, simply change the 'version' to a later Fedora version. Thank you for reporting this issue and we are sorry that we were not able to fix it before Fedora 28 is end of life. If you would still like to see this bug fixed and are able to reproduce it against a later version of Fedora, you are encouraged change the 'version' to a later Fedora version prior this bug is closed as described in the policy above. Although we aim to fix as many bugs as possible during every release's lifetime, sometimes those efforts are overtaken by events. Often a more recent Fedora release includes newer upstream software that fixes bugs or makes them obsolete.
Fedora 28 changed to end-of-life (EOL) status on 2019-05-28. Fedora 28 is no longer maintained, which means that it will not receive any further security or bug fix updates. As a result we are closing this bug. If you can reproduce this bug against a currently maintained version of Fedora please feel free to reopen this bug against that version. If you are unable to reopen this bug, please file a new report against the current release. If you experience problems, please add a comment to this bug. Thank you for reporting this bug and we are sorry it could not be fixed.