-
Notifications
You must be signed in to change notification settings - Fork 257
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
github: require reef job to pass in mergify config #910
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.
Request to add missing check for quincy job in mergify configuration.
Thank you for the reminder. PR Updated. |
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.
So reef build is failing there's yet another nfs ganesha packaging related issue. I'm not sure how long this one will last. We can try to wait it out or we can manually merge before the nfs ganesha issue is fixed. We have to manually merge this PR at some time because it changes the mergify config. |
I noticed the failure and initiated a discussion with Kaleb as there was an update pushed yesterday for nfs-ganesha from CentOS Storage SIG. Is that OK if I give it a day or two for fixing the installation issues? |
Fine by me. |
Current installation issue with nfs-ganesha is due to the fact that we execute a Reef is configured to build images with nfs-ganesha v5 packages. They do have a set of clean up instructions within the Dockerfile where /etc/selinux is removed. With updated packages for nfs-ganesha-5.x/nfs-ganesha-selinux-5.x, an update fails to find /etc/selinux/config resulting in an installation error as seen in the GitHub CI logs.
Issue will be fixed once the Reef image gets rebuilt with latest nfs-ganesha-5.x packages. But if I understand correctly this rebuild will only happen with next update to v18.2. In this context I would like to rethink on the need for |
AFAIR we update the packages because the RPM repository doesn't necessarily have the same packages that are in the container image any more. Since the point is to get the devel packages we ended up failing the install because the matching dev/non-dev packages were not present.
So the yum/dnf update prevents failing the devel package install in this case. I'm not inclined to remove this step, which has been working for years at this point because of a bug we know will be fixed. I'd be willing to try and put in an nfs-ganesha specific workaround if we can. However, I'm willing to continue the discussion too. :-) |
Are you talking about Ceph related packages or in general? In any case I think DNF is capable of detecting updates in such situations. Out of my curiosity I created a similar situation on my Fedora 38 machine:
|
It was this fix: #510 It was definitely an issue. Some additional ceph packages couldn't be installed, because some other installed ceph packages were not matching. It did not always happen, but in certain times around releases IIRC. So we would need an alternative solution, we can't just remove the update. |
1eaf69f
to
5b7e1bd
Compare
Hm.. I'm not convinced completely but I am good to go with #913 for now. |
@Mergifyio rebase |
Signed-off-by: John Mulligan <[email protected]>
After adding the reef job, it was noticed that there was no quincy job. Apparently we (probably me) forgot to add it at the time we added quincy to our CI matrix, so we're adding it now. Signed-off-by: John Mulligan <[email protected]>
✅ Branch has been successfully rebased |
5b7e1bd
to
098c248
Compare
It does what it says on the tin.
This is a change to the mergify configuration and thus will require a manual merge upon approval.