-
Notifications
You must be signed in to change notification settings - Fork 158
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
kola/kernel-replace: adapt for RHCOS and SCOS #3048
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Shellcheck will complain other than this it looks good.
Could you run the test locally on RHCOS ?
d95cceb
to
18d9d86
Compare
Just tried it now, it's failing. I'm now investigating:
|
Trying to replace the RHCOS kernel with a Fedora one in theory should work, but I think we can get closer to what users are likely to do by replacing with a CentOS Stream kernel instead. So basically:
For Fedora, the test currently pins to a specific Koji build, which is stale now. We should bump it. Once we have dnf in all the streams, we should probably change that actually to just dynamically always take the latest kernel that's != to the kernel embedded in the container image (similarly for RHCOS/SCOS too). |
1899c9f
to
3ec17ba
Compare
I ran this on both |
That's now done.
That's also done.
I can look into this in the future. |
Again, latest changes successfully passed test both on |
e99b318
to
e4c36a4
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Suggested commit title:
kola/kernel-replace: adapt for RHCOS and SCOS
For the commit message, I think in general we should avoid just linking to some other URL that has all the context. Let's inline that context, and provide the URL only as an additional reference in the footer.
5650a3a
to
03d962c
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM overall. Thoughts on suggested commit title in #3048 (review) ? I think mentioning OS_ID
there isn't really capturing it.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM overall. I am a bit concerned by the hard-coded kernel package URLs, won't they go stale eventually, breaking the test?
I don't have a good alternative to suggest though, except adding an item to the rebase checklist..
I agree, @jlebon also talked about it here: #3048 (comment). Let's wait with that until we have DNF in all of the streams. |
In one of the older issues, kola switch-kernel test was failing. In the discussion @jlebon suggested to remove named test and make it external, so that it can be utilized by FCOS, and also SCOS/RHCOS. Additionally, the kernel version for FCOS was updated in this PR to the latest stable version `6.9.8-200.fc40`. See: coreos/coreos-assembler#1245
There were instances where kola/switch-kernel test would fail.
It was recommended to remove it (PR) and generalize this test, so that it can work with FCOS, RHCOS and SCOS.