Closed marmijo closed 1 month ago
Offhand perusing the dracut commits, this looks like a possibly a regression from https://github.com/dracut-ng/dracut-ng/commit/377d52cb9867dc4074fe78965e1bfcdf80be14a1 and https://github.com/dracut-ng/dracut-ng/commit/4957ffa935ea8c393c8fed867f6051da4ca43d58, and which was then fixed by https://github.com/dracut-ng/dracut-ng/commit/1e802f15fee3d6402d83e0efefef8bb88c5a33e3. We could verify that it fixes the tests, and propose to have it backported to the dracut maintainers in Fedora.
The /var/log
message in the output above is likely non-fatal and unrelated to the hard failure here. It's probably a systemd generator that needs to be adapted to work with latest systemd (see https://github.com/coreos/fedora-coreos-config/pull/3014, esp. https://github.com/coreos/fedora-coreos-config/pull/3014#issuecomment-2148213529).
The
multipath.day1
andmultipath.day2
kola tests are failing after updating to the latestdracut
package in F42 rawhide.The following is seen in the console output of the two tests:
Full console output: multipath.day2.console.txt
pinning
dracut-102-2.fc41
in rawhide, as well as the dependentclevis
package, allows these tests to pass without issue.EDIT: opened the issue before filling out all information:
The recent changes to the dracut package can be found here:
544ea5d
: update to dracut 103b052294
: Update from source-git