Hi,
I’m running a special initrd generated for SystemImager with a special dracut module that relies on plyouth scripted theme. It works perfectly on CentOS-6, CentOS-7, OpenSUSE-42.3, Fedora-29 (not tested later) and used to work on CentOS-8.0
Since CentOS-8.1 was released, IT chrashes with the following error: Plymouthd crash 913: ply_event_loop_watch_for_timeout: Assertion `seconds > 0.0' failed.
I’m running quemu 5.0.0 with tianocore UEFI bios on MacOS (using accel=none or accel=hvf) (Bug also occurs with standard 256k bios)
When I strace, I see multiple reasons but I can’t tell wich one is fatal. 1/ plymouthd reads /etc/Plymouth/plymouthd.conf (ok), then tries to read them from /var/run/plymouth/themes/<themename> instead of /usr/share/plymouth/themes/<themename>/ . As it fails it tries to read /var/run/plymouth/plymouthd.defaults (fails) then /usr/share/plymouth//plymouthd.defaults (ok) then it tries to read the theme /var/run/plymouth/themes/<themename>/<themename>.plymouth (fails) but never tries to read in from /usr/share/plymouth/themes/<themename>/<themename.plymouth> ???
It still continues
2/ After a few things, it tries to read /sys/devices/virtual/drm/ttm/uevent (ok) and the it tries to read /run/dev/data/+drm:ttm (fail). Is it expectable or does this means a problem like a missing udev rule or a boch driver bug or something else?
3/ Same as step 2
4/ Crash with @lymouthd: ply-event-loop.c:913: ply_event_loop_watch_for_timeout: Assertion `seconds > 0.0' failed.
I’ve reported this in Centos-8.1 with no return and then RedHat-8.1 today. RedHat 8.1 bug report: https://bugzilla.redhat.com/show_bug.cgi?id=1836828 CentOS-8.1 bug report: https://bugs.centos.org/view.php?id=17140
Any idea where I could look for? More likely a plymouth bug or an udev missing rule or something else? (full dmesg and strace log in bug reports)
Thanks a lot for any tips.
PS: looking at google results, the only things I can find are very old bugs without resolution (not fixed and obsoleted due to new OS release)