fix: handle startup-probe-phase pods in rolling update categorization#435
Merged
Conversation
julienmancuso
previously approved these changes
Feb 12, 2026
Ronkahn21
reviewed
Feb 17, 2026
There was a problem hiding this comment.
Pull request overview
Fixes rolling update categorization so pods in the startup-probe phase (Running + Started=false/nil + Ready=false) are no longer silently dropped from computeUpdateWork, preventing updates from getting stuck or ending prematurely (issue #400).
Changes:
- Add
HasAnyContainerNotStartedto detect pods whose containers have not passed startup probe. - Expand
computeUpdateWorkwitholdTemplateHashStartingPodsandoldTemplateHashUncategorizedPodsbuckets, and ensure old-hash pods always land in a bucket. - Delete all non-ready old-hash pods (pending/unhealthy/starting/uncategorized) immediately; add unit tests for the new utility and categorization.
Reviewed changes
Copilot reviewed 4 out of 4 changed files in this pull request and generated no comments.
| File | Description |
|---|---|
| operator/internal/utils/kubernetes/pod.go | Adds HasAnyContainerNotStarted pod utility. |
| operator/internal/utils/kubernetes/pod_test.go | Adds unit tests for HasAnyContainerNotStarted. |
| operator/internal/controller/podclique/components/pod/rollingupdate.go | Extends rolling update bucketing and expands non-ready old-hash deletion behavior. |
| operator/internal/controller/podclique/components/pod/rollingupdate_test.go | Adds unit tests covering computeUpdateWork bucketing, including startup-probe-phase pods. |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
2f5130d to
00f79c4
Compare
Contributor
Author
|
/ok to test 00f79c4 |
Ronkahn21
previously approved these changes
Feb 17, 2026
julienmancuso
previously approved these changes
Feb 19, 2026
fbccbdf
fbccbdf to
d97f86b
Compare
Contributor
Author
|
/ok to test d97f86b |
Ronkahn21
approved these changes
Feb 20, 2026
d97f86b to
30c7ebe
Compare
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
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.
/kind bug
What this PR does / why we need it:
Fixes a bug where rolling updates get permanently stuck or prematurely complete when an old-hash pod is in the startup probe phase.
computeUpdateWorkhad no category for pods withStarted=false(startup probe not yet passed), causing them to silently fall through all classification branches.This PR:
HasAnyContainerNotStartedutility to detect pods still in the startup probe phaseoldTemplateHashStartingPodsandoldTemplateHashUncategorizedPodsbuckets toupdateWorkcomputeUpdateWorkcategorization andHasAnyContainerNotStartedWhich issue(s) this PR fixes:
Fixes #400
Special notes for your reviewer:
The root cause: when a second spec change arrives while a replacement pod from a previous change is still in its startup probe phase, the replacement pod (now old-hash) has
Phase=Running,Started=false,Ready=false. It fails all existing predicates (not Pending, not Started-but-not-Ready, not Ready) and is dropped from the update work entirely. This causes either an infinite requeue loop or prematureupdateEndedAtmarking.Does this PR introduce a API change?
Additional documentation e.g., enhancement proposals, usage docs, etc.: