kafka: reconcile returning 3.0 broker logs from full metadata - #583
Draft
bringhurst wants to merge 1 commit into
Draft
bringhurst wants to merge 1 commit into
bringhurst wants to merge 1 commit into
Conversation
A former replica can miss deletion while offline and reuse old bytes after reassignment. Under the cleanup gate, reconcile unhosted logs from a full initial image. Include leaderless assignments, preserve incremental updates and keep the old JVM cache methods.
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
A former replica can miss deletion while offline and reintroduce old records after name reuse. The targeted test failed after ISR recovery and promotion: offset 0 contained an old 64-byte record, not the new 128-byte record. It now passes with the fix.
Behind the existing default-off cleanup gate, use the complete initial metadata image to retire unhosted logs that are no longer assigned here. Keep assigned leaderless logs, reject stale controller epochs, and never treat an incremental update as a complete image. Preserve existing JVM methods.
Both versions pass the updated stray-log and metadata-cache tests. The 3.9 follow-up requires scenario revision 3 and both broker directions, and retries only bounded ZooKeeper startup probes. All 68 Python tests pass. The expanded full verifier is running; this remains a draft, not rollout approval.