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?
| - hosts: all | ||
| tasks: | ||
| - include_tasks: dependencies-fedora.yaml | ||
| - include_tasks: "{{ dependencies_playbook | default('dependencies-fedora.yaml') }}" |
There was a problem hiding this comment.
Thanks! This is very handy and beyond my very rudimentary YAML skills. :)
| label: cloud-fedora-rawhide | ||
| pre-run: playbooks/setup-env-restricted.yaml | ||
| vars: | ||
| dependencies_playbook: dependencies-fedora-restricted.yaml |
There was a problem hiding this comment.
This reminds me of a suggestion from Gemini Code Assist, when I was doing some CVE backports, to parameterise the definitions of the jobs in .zuul.yaml.
I never managed to get around to it. Do you think it would be a good idea and aligned with your work here? If so, would you like to throw in a patch for that? :)
There was a problem hiding this comment.
Sure yeah, I can look into it
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 |
With this change, jobs for different platforms can select their own dependency playbook without needing a separate setup-env-* file. This allows us to eliminate setup-env-restricted.yaml, as it was identical to setup-env.yaml except for the include_tasks line. This will also be used for the FCOS jobs. Assisted-by: Claude (claude-opus-4-6) Signed-off-by: Rolv Apneseth <rolv.apneseth@gmail.com>
Rolv-Apneseth
force-pushed
the
fcos_ci
branch
2 times, most recently
from
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 |
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. Assisted-by: Claude (claude-opus-4-6) Signed-off-by: Rolv Apneseth <rolv.apneseth@gmail.com>
Assisted-by: Claude (claude-opus-4-6) Signed-off-by: Rolv Apneseth <rolv.apneseth@gmail.com>
Rolv-Apneseth
force-pushed
the
fcos_ci
branch
3 times, most recently
from
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. |
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)