-
Notifications
You must be signed in to change notification settings - Fork 36
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
rhel10 and daily-iso timeout with "WARNING anaconda:anaconda: display: Wayland startup failed: systemd exited with status 1" #1261
Comments
lorax-packages.log diff (compared to daily 2 days before):
|
So I've checked the logs & the main issue is that there is not much to work with - there is a slog message about Wayland being started & then a bit later a message about Wayland filing to start. :P Hard to do something about this unless we have a reproducer - but I have a theory. Maybe this is related to the issue we are seeing for quite a while, where the output from GNOME Kiosk clobbers the TTY1 console. Maybe that is where all the relevant error message send up in this case ? Might be a good idea to re-check if we can't run the GNOME tools via systemd units again, as that could possibly result in all the log messages ending up in Journal instead of all over the place on TTY1. :) |
0 occurrences last week. |
4 occurences 06-08-2024 - 12-08-2024 |
Actually there were one occurence 08-05-2024 |
After porting the Wayland migration to Fedora (rhinstaller/anaconda#5829) we are seeing the flake also on daily-iso. |
Hi @jexposit, any idea how we could debug this Wayland failure? |
There is not much information in the log. It looks like run-in-new-session starts correctly:
A systemd warning is reported:
But GNOME Kiosk seems to start:
I'm not sure if there could be more information in the TTY or if there is a way to check it. The only errors I see in the logs are:
But I'm not sure if those could be causing the problem. Checking if that error is present in a successful log could help. I checked in a local VM, latests |
Hi @vojtechtrefny, I have a feeling I saw these errors regularly but I might be wrong. Do you know something about the udisksd errors? |
@jexposit based on the code we are getting this error here: because socket for Wayland wasn't created: which means that the Wayland socket wasn't created. I don't think udisks is can have much of an impact to this. Is there a way how to get debug data from this? We can create Anaconda PR and run all the tests there to find out the cause. What we can do is to enabled some debug mode when inst.debug is used or something like that for these cases. |
Yes, that's a 6.10 kernel bug. We also have a workaround in the latest udisks: storaged-project/udisks#732 |
@vojtechtrefny could that kernel bug prevent the creation of files? In this case, it looks like the Wayland socket can not be created. |
No. This is a problem with response to some ATA commands and technically speaking the kernel is not broken, it just changed the format of the answer which the standard allows, but some userspace tools (like udisks) were not ready for that. It has no effect on actually reading/writing data from/to the disk. |
@jkonecny12 mutter includes several debug options: From a quick look at
I think that |
First occurence 07/20/2024
3 occurrences last week
virt-install.log
virt-install.log
virt-install.log
The text was updated successfully, but these errors were encountered: