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
Fix: Recreate DaemonSet pods completed with Succeeded phase #117073
Fix: Recreate DaemonSet pods completed with Succeeded phase #117073
Conversation
Please note that we're already in Test Freeze for the Fast forwards are scheduled to happen every 6 hours, whereas the most recent run was: Mon Apr 3 16:34:40 UTC 2023. |
/sig apps Adding an integration test for DaemonSet to confirm that Succeeded pods are not restarted. Notably, the test for the Failed pod passes, so failed pods are restared. Not sure why the discrapancy exists, seems like a bug. Will look deeper tomorrow. |
4be55b6
to
96e56e7
Compare
/assign @alculquicondor @soltysh |
/retest |
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.
This looks good to me, but I'd like @smarterclayton to double check me, I looked in the history and couldn't find any documents about inner-workings of daemonset controller, so I'd fallback to wiser man to double check me 😅
/priority important-longterm |
LGTM label has been added. Git tree hash: 25221ea423a393d389236df41649fa168db4fe34
|
ACK-ing the PR, I pinged the Release Team Leads on Slack as they're responsible for the code freeze exceptions: https://kubernetes.slack.com/archives/C2C40FMNF/p1680636984022679 |
cc @kubernetes/release-team-leads |
Looks like the issue was introduced in 1.27 and the fix is critical. Since rc.1 was delayed to Thursday, I am +1 to merge this one before v1,27,0-rc.1. |
/lgtm nit: I would retitle PR title and the release note to "Recreate DaemonSet pods completed with Succeeded phase" rather than restart (since it is a slightly more accurate). |
/retitle Fix: Recreate DaemonSet pods completed with Succeeded phase /release-note-edit
|
/cc @kow3ns |
Since this is considered a release blocker and introduced in 1.27, adding label to merge it before 1.27.0-rc1 cut. |
/hold cancel @soltysh already approved and my LGTM is under my SIG Apps reviewer hat |
/remove-milestone v1.28 |
since 1.26 daemonset controllers can run against 1.27 API servers with 1.27 kubelets, shouldn't this be backported to 1.26? |
I think we should indeed, based on https://kubernetes.io/releases/version-skew-policy/. I will open the cherry-pick PR |
Ready: #117496 |
…73-upstream-release-1.26 Automated cherry pick of #117073: Fix: Recreate DaemonSet pods completed with Succeeded phase
What type of PR is this?
/kind bug
/kind regression
What this PR does / why we need it:
Which issue(s) this PR fixes:
Fixes #117018
Special notes for your reviewer:
It is a fix for the issue introduced in #115331
Does this PR introduce a user-facing change?
Additional documentation e.g., KEPs (Kubernetes Enhancement Proposals), usage docs, etc.: