This should also trigger in fedora-40 already. Let's make an image right away...

image-refresh fedora-40 done: https://github.com/cockpit-project/bots/commits/image-refresh-fedora-40-20241125-101752

Success. Log: https://cockpit-logs.us-east-1.linodeobjects.com/image-refresh-fedora-40-1d76dfbd-20241125-095709/log.html

(Copied from the parallel image refresh #7126 -- there is no reason to believe that this refresh magically fixes any of these):

Phew, this is a big one. There's the python crashes from #21301, files needs a pixel update due to the new cockpit version enabling "open in terminal", sub-man causes an oops in the same manner as in #7123, and composer failed on an internal npm mirror failure (retried that one).

(Copied from the parallel image refresh #7126 -- there is no reason to believe that this refresh magically fixes any of these):

Oops, I wasn't aware yet of that PR... sorry.

TestMultiCPU.testCPUUsage not caught by naughty?

TestMultiCPU.testCPUUsage not caught by naughty?

My mistake. Man, I am bad at multi-tasking...

sudo dnf info cockpit-bridge shows 329.1 now, so refreshing. And yup, this picked up the new version

image-refresh fedora-40 done: https://github.com/cockpit-project/bots/commits/image-refresh-fedora-40-20241126-124557

Success. Log: https://cockpit-logs.us-east-1.linodeobjects.com/image-refresh-fedora-40-fdac239d-20241126-122414/log.html

Argh, still that machines failure. @jelly and I collided here, I sent cockpit-project/cockpit-machines#1925 to clean that up. Tested locally.

cockpit-project/cockpit-machines#1925 landed, so retrying the machines tests. cockpit and -files only fail the pixel tests. After machines goes green, I'll land the pixel tests and then do a final rebase to land this fully green -- otherwise we'll ruin the flake statistics. It's evening and fairly quiet, so I think we can afford to do it this way.