fix: exclude transient pod failures from controller status - #22
Merged
Conversation
relu
force-pushed
the
fix-kstatus-polling-pending-pods
branch
from
February 25, 2026 21:15
0f6d781 to
112f730
Compare
matheuscscp
reviewed
Feb 26, 2026
matheuscscp
left a comment
matheuscscp
left a comment
Member
There was a problem hiding this comment.
@relu Thanks for catching this so fast!
Pods that are unschedulable or being deleted during a rolling update are now excluded when determining if a pod controller has failed. This prevents false failure reports when cluster autoscalers have not yet provisioned nodes or when old pods are being terminated. The unschedulable check requires a PodScheduled condition with status=False and reason=Unschedulable, matching the existing check in the watcher package. Signed-off-by: Aurel Canciu <aurel.canciu@nexhealth.com>
relu
force-pushed
the
fix-kstatus-polling-pending-pods
branch
from
February 26, 2026 10:54
112f730 to
28dac90
Compare
matheuscscp
approved these changes
Feb 26, 2026
matheuscscp
left a comment
matheuscscp
left a comment
Member
There was a problem hiding this comment.
LGTM! 🚀
Thanks very much!
stefanprodan
approved these changes
Feb 26, 2026
Member
|
@relu This is shipped in Flux v2.8.1, can you pls upgrade and let us know? 🙏 |
Member
Author
|
@matheuscscp I've managed to upgrade and test the fix earlier today. So far, so good. I'm no longer seeing the transient failures. 🙌 |
TerryHowe
added a commit
to TerryHowe/helm
that referenced
this pull request
Mar 3, 2026
Bumps fluxcd/cli-utils from v0.37.1-flux.1 to v0.37.2-flux.1, which includes a fix to exclude transient pod failures from controller status (fluxcd/cli-utils#22). This resolves intermittent failures with `helm upgrade --install` when using the watcher wait strategy, where StatefulSets were prematurely reported as Failed when pods were unschedulable (e.g., waiting for cluster autoscaler/Karpenter to provision nodes). The upstream fix treats unschedulable pods and pods being deleted during rolling updates as transient states rather than terminal failures. Fixes helm#31896 Signed-off-by: Terry Howe <terrylhowe@gmail.com>
5 tasks
scottrigby
pushed a commit
to helm/helm
that referenced
this pull request
Mar 11, 2026
Bumps fluxcd/cli-utils from v0.37.1-flux.1 to v0.37.2-flux.1, which includes a fix to exclude transient pod failures from controller status (fluxcd/cli-utils#22). This resolves intermittent failures with `helm upgrade --install` when using the watcher wait strategy, where StatefulSets were prematurely reported as Failed when pods were unschedulable (e.g., waiting for cluster autoscaler/Karpenter to provision nodes). The upstream fix treats unschedulable pods and pods being deleted during rolling updates as transient states rather than terminal failures. Fixes #31896 Signed-off-by: Terry Howe <terrylhowe@gmail.com> (cherry picked from commit 1dfa77e)
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Pods that are pending scheduling or being deleted during a rolling update are now excluded when determining if a pod controller has failed. This prevents false failure reports when cluster autoscalers have not yet provisioned nodes or when old pods are being terminated.
Fixes #21