Add Fedora CoreOS to CI tests - #1806
Conversation
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
There was a problem hiding this comment.
Code Review
This pull request introduces Fedora CoreOS system test jobs in Zuul and refactors the dependency playbooks by extracting shared tasks into a new common playbook (dependencies-common.yaml). Feedback on the changes highlights that the rpm-ostree install command in dependencies-fedora-coreos.yaml uses an unsupported --idempotent flag which will cause failures, and recommends adding the missing flatpak-session-helper package to prevent potential failures in Toolbx system tests.
Important
The consumer version of Gemini Code Assist on GitHub is being sunset. Starting June 18, 2026, new organization installations will be blocked, and all code review activity will officially cease on July 17, 2026.
For more details on the timeline and next steps, please review the Help Documentation.
debarshiray
left a comment
debarshiray
left a comment
There was a problem hiding this comment.
Thanks for soldiering on with the Fedora CoreOS tests, @Rolv-Apneseth ! This is exciting.
It seems like we can land the first commit independently from the rest of the work and the CI turning green. It seems useful on its own, and it will keep things moving. Is it sufficiently well settled already?
No worries! It will be cool to finally have this done.
Sure, would you want it split off into a separate PR? Still working with the SF maintainers to get the CoreOS nodes working so this PR may take a while longer. |
|
By the way, are those timeouts in the |
88ee98e to
6d0f743
Compare
|
Looks like the next and testing nodes are passing now. The jobs on stable nodes fail due to a version mismatch when trying to layer in |
|
So the updating didn't work since it only updates to the latest available image (which it already is in this case). I'll keep that step anyway as it's good to avoid similar issues when the node image is older than the latest available one. I'll try see if there's any other workaround we can take here. The failures in |
|
The failure described in the previous comment probably needs to be investigated separately, maybe a separate issue? As for the timeouts - any suggestions? @debarshiray Edit: started looking into it but I see this is being investigated in #1805 |
ffe3d61 to
1063903
Compare
|
Hostname issue is actually on the Zuul node side of things, opened another PR with SF to try and resolve that. Should be the final step here. |
|
Hostname checks pass now, but re-running to see if we can get the ones that hit the retry limit to pass (the non-rawhide ones at least). |
|
Rebased after #1824 |
debarshiray
left a comment
debarshiray
left a comment
There was a problem hiding this comment.
This looks good to me! Just one question below.
containers#1806 Assisted-by: Claude (claude-opus-4-6) Signed-off-by: Rolv Apneseth <rolv.apneseth@gmail.com>
|
Rebased on |
debarshiray
left a comment
debarshiray
left a comment
There was a problem hiding this comment.
Oops! I hit "submit review" too soon.
Could you please remove the Assisted-by: Claude (claude-opus-4-6) trailers from the Git commit messages?
I never wrote Assisted-by: Stack Overflow or Assisted-by: #gnome-hackers or similarly some API documentation, books, debuggers, other code bases, etc. in my Git commit messages in all my years of copy-pasting code from the Internet. Why care about copying any and every snippet of code written by artificial intelligence when we barely bate an eyelid over doing the same for human intelligence? :)
It seems to me that you are taking responsibility for the correctness of the code, and I am doing the same as the reviewer, so the generative AI is irrelevant for correctness. There's a big question mark about copyright, and that would be impossible to work out from a trailer in the Git commit message.
Beyond the aesthetic or philosophical nitpicking, in concrete terms, I don't want the Git commit messages of free software projects to turn into billboards offering no-cost advertising for big AI companies.
|
Sure yes, I'll remove them. It was more about disclosure rather than wanting to give credit, but I understand your points. |
Add system tests across the stable, next and testing Fedora CoreOS streams, with some minor refactoring for shared dependency setup with Fedora. Also added a cat /etc/os-release to the common dependency setup as it may also be helpful on Fedora runs. containers#714 Signed-off-by: Rolv Apneseth <rolv.apneseth@gmail.com>
containers#1806 Signed-off-by: Rolv Apneseth <rolv.apneseth@gmail.com>
Not really sure what changed to cause this though. Any ideas @debarshiray? |
Me neither - I don't know why the tests running on Ubuntu 22.04 hosts started failing. I am guessing that it's something to do with GitHub and/or Ubuntu's infrastructure. The failed job from this PR has: ... and the last successful job from #1829 has: |
Restarting the failed job brought it somewhat in line with the last successful one: Let's hope it works out. :) |
Well that's odd. Hopefully it's not a recurring flake then |
Yeah. I have seen something like this once or twice before on rare occasions. I am relatively clueless about the internals of Ubuntu, so I restart the job and hope for the best. :) |
2 participants
Closes #714, which also includes the context for this PR.
This is also testing whether the images added to the SF nodepool in https://gitlab.com/softwarefactory-project/config/-/merge_requests/63#note_3445923313 actually work as intended so there may be some experimentation.
Let me know if there are any obvious issues, but otherwise I will let you know if/when tests are passing @debarshiray.
Assisted-by: Claude (claude-opus-4-6)