Conversation
|
|
githubzilla
added a commit
that referenced
this pull request
Mar 21, 2026
Bug #3 (root cause of assertion crash): CleanBucketData/CleanRangeData could free CcEntries with BeingCkpt=true, causing dirty count double-decrement when the checkpoint callback (UpdateCceCkptTsCc) later runs on the freed entry. Fix: CanBeCleaned() now returns !GetBeingCkpt() for CleanBucketData, CleanRangeData, and CleanRangeDataForMigration. Entries being checkpointed are skipped and retried later. Bug #1: TemplateCcMap::BackFill called SetCkptTs() before SetCommitTsPayloadStatus(), which overwrites commit_ts_and_status_ and clears the flush bit, leaving the entry dirty without incrementing the counter. Also missing OnCommittedUpdate in ReadOutsideCc backfill path. Fix: Reorder to SetCommitTsPayloadStatus first, then SetCkptTs, and add OnCommittedUpdate in both BackFill and ReadOutsideCc paths. Bug #2: ClusterConfigCcMap called SetCommitTsPayloadStatus() at two sites without OnCommittedUpdate(), making entries dirty without counting them. Fix: Add OnCommittedUpdate after both SetCommitTsPayloadStatus calls. Also relax UpdateCceCkptTsCc assertions to allow IsPersistent() being true, since concurrent BackFill/ReadOutsideCc can legitimately mark an entry persistent before the checkpoint callback runs.
githubzilla
added a commit
that referenced
this pull request
Mar 21, 2026
Bug #3 (root cause of assertion crash): CleanBucketData/CleanRangeData could free CcEntries with BeingCkpt=true, causing dirty count double-decrement when the checkpoint callback (UpdateCceCkptTsCc) later runs on the freed entry. Fix: CanBeCleaned() now returns !GetBeingCkpt() for CleanBucketData, CleanRangeData, and CleanRangeDataForMigration. Entries being checkpointed are skipped and retried later. Bug #1: TemplateCcMap::BackFill called SetCkptTs() before SetCommitTsPayloadStatus(), which overwrites commit_ts_and_status_ and clears the flush bit, leaving the entry dirty without incrementing the counter. Also missing OnCommittedUpdate in ReadOutsideCc backfill path. Fix: Reorder to SetCommitTsPayloadStatus first, then SetCkptTs, and add OnCommittedUpdate in both BackFill and ReadOutsideCc paths. Bug #2: ClusterConfigCcMap called SetCommitTsPayloadStatus() at two sites without OnCommittedUpdate(), making entries dirty without counting them. Fix: Add OnCommittedUpdate after both SetCommitTsPayloadStatus calls. Also relax UpdateCceCkptTsCc assertions to allow IsPersistent() being true, since concurrent BackFill/ReadOutsideCc can legitimately mark an entry persistent before the checkpoint callback runs.
githubzilla
added a commit
that referenced
this pull request
Mar 24, 2026
Bug #3 (root cause of assertion crash): CleanBucketData/CleanRangeData could free CcEntries with BeingCkpt=true, causing dirty count double-decrement when the checkpoint callback (UpdateCceCkptTsCc) later runs on the freed entry. Fix: CanBeCleaned() now returns !GetBeingCkpt() for CleanBucketData, CleanRangeData, and CleanRangeDataForMigration. Entries being checkpointed are skipped and retried later. Bug #1: TemplateCcMap::BackFill called SetCkptTs() before SetCommitTsPayloadStatus(), which overwrites commit_ts_and_status_ and clears the flush bit, leaving the entry dirty without incrementing the counter. Also missing OnCommittedUpdate in ReadOutsideCc backfill path. Fix: Reorder to SetCommitTsPayloadStatus first, then SetCkptTs, and add OnCommittedUpdate in both BackFill and ReadOutsideCc paths. Bug #2: ClusterConfigCcMap called SetCommitTsPayloadStatus() at two sites without OnCommittedUpdate(), making entries dirty without counting them. Fix: Add OnCommittedUpdate after both SetCommitTsPayloadStatus calls. Also relax UpdateCceCkptTsCc assertions to allow IsPersistent() being true, since concurrent BackFill/ReadOutsideCc can legitimately mark an entry persistent before the checkpoint callback runs.
liunyl
added a commit
that referenced
this pull request
Jun 13, 2026
Two distinct bugs were chained under the quarantined LargeObjLRU-Test (the FastMetaDataMutex tls_shard_idx spin was fixed earlier in this branch): 1. Engine bug (template_cc_map.h, FindEmplace, LO_LRU policy): when the large-object page is the LAST page in the map and a key greater than it is inserted, the `target_it == ccmp_.end()` branch creates the new next-page but never repoints `target_page` to it (the parallel `else` branch does). The key is then Emplaced back into the large-object page, and TryUpdatePageKey calls FirstKey() on the still-empty new page -> assert(!keys_.empty()) (TC-FE-01). In NDEBUG this is worse: FirstKey() reads keys_.front() on an empty vector (UB) and the large-object "alone on its page" invariant is violated. Fix: add the missing `target_page = target_it->second.get();`, mirroring the else branch. 2. Test-setup bug (LargeObjLRU-Test.cpp, 4 sites): tests create partial- commit dirty entries and compensated dirty_data_key_count_ via f.shard.AdjustDataKeyStats(...) -- the SHARD counter only. OnCommittedUpdate normally bumps both the shard and the MAP counters, so the map's dirty_data_key_count_ was left at 0 while the entries are dirty, and Terminate()'s decrement underflowed it (assert in TemplateCcMap::AdjustDataKeyStats). This surfaced only once bug #1 stopped killing the suite earlier. Fix: drive the real clean->dirty API (cc_map.OnCommittedUpdate) so both counters stay in step. (This is unrelated to the production dirty-count underflow paths in #465.) With both fixes the full LargeObjLRU-Test (23 cases, 3701 assertions) passes deterministically, so it is un-gated in CI (only ClusterCrossNg-Test remains non-gating). Full ctest suite: 40/40 green. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
liunyl
pushed a commit
that referenced
this pull request
Jun 15, 2026
Bug #3 (root cause of assertion crash): CleanBucketData/CleanRangeData could free CcEntries with BeingCkpt=true, causing dirty count double-decrement when the checkpoint callback (UpdateCceCkptTsCc) later runs on the freed entry. Fix: CanBeCleaned() now returns !GetBeingCkpt() for CleanBucketData, CleanRangeData, and CleanRangeDataForMigration. Entries being checkpointed are skipped and retried later. Bug #1: TemplateCcMap::BackFill called SetCkptTs() before SetCommitTsPayloadStatus(), which overwrites commit_ts_and_status_ and clears the flush bit, leaving the entry dirty without incrementing the counter. Also missing OnCommittedUpdate in ReadOutsideCc backfill path. Fix: Reorder to SetCommitTsPayloadStatus first, then SetCkptTs, and add OnCommittedUpdate in both BackFill and ReadOutsideCc paths. Bug #2: ClusterConfigCcMap called SetCommitTsPayloadStatus() at two sites without OnCommittedUpdate(), making entries dirty without counting them. Fix: Add OnCommittedUpdate after both SetCommitTsPayloadStatus calls. Also relax UpdateCceCkptTsCc assertions to allow IsPersistent() being true, since concurrent BackFill/ReadOutsideCc can legitimately mark an entry persistent before the checkpoint callback runs.
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.
Here are some reminders before you submit the pull request
fixes eloqdb/tx_service#issue_id./mtr --suite=mono_main,mono_multi,mono_basic