Repository navigation
HBASE-30352 Tolerate stale recovered.edits below durable seqid in split/merge (branch-2.4 backport) - #8706
Closed
nirdosh0110 wants to merge 1 commit into
Closed
HBASE-30352 Tolerate stale recovered.edits below durable seqid in split/merge (branch-2.4 backport)#8706nirdosh0110 wants to merge 1 commit into
nirdosh0110 wants to merge 1 commit into
Conversation
…it/merge Backport of apache#8599 to branch-2.4. (cherry picked from commit cc1876e)
Contributor
|
@nirdosh0110 I realized branch-2.4 doesn't seem active anymore, we can leave this backport and close the PR. |
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
2 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Backport of #8599 to
branch-2.4(JIRA: HBASE-30352).Problem
On split/merge,
AssignmentManagerUtil.checkClosedRegionaborts the procedure whenever anyrecovered.editsfile exists for a parent region, to prevent data loss. In corner cases a region can be left withrecovered.editswhose max seqid is already covered by the region's durable (last-flushed) seqid — those edits are provably redundant, yet their mere presence wedges the split/merge into a stuck RIT that requires manual cleanup.Change
checkClosedRegionnow tolerates such stale files: before aborting, it consults the master's last-flushed seqid for the region and, only when that watermark is authoritative (> 0), dropsrecovered.editsfiles whose max seqid is<= durableSeqId. If any file's seqid is above the watermark (or the watermark is unknown), it still aborts as before — no behavior change for the genuinely-unsafe case.WALSplitUtil.getSplitEditFilesSorted.durableSeqId; a failed delete or an unparseable/higher seqid falls back to the safe abort.Test
TestAssignmentManagerUtilcovers the new path (3 tests, all green locally).Verification (local)
mvn -pl hbase-server -am clean test -Dtest=TestAssignmentManagerUtil -Dhadoop.profile=3.0(JDK8) → Tests run: 3, Failures: 0, Errors: 0mvn -pl hbase-server spotless:check -Dhadoop.profile=3.0→ BUILD SUCCESSBackport notes
AssignmentManagerUtil.java) applied cleanly, including the addedLogger/FS imports.org.junit.*/@Category/@ClassRule, addedassertTrue, and switched the new assertion to JUnit4 argument order (assertFalse(message, condition)).