-
Type:
Task
-
Resolution: Unresolved
-
Priority:
Major - P3
-
None
-
Affects Version/s: None
-
Component/s: None
-
None
-
Storage Engines
-
None
-
None
Issue Summary
16 out of 66 BFG configurations encountered a checksum mismatch issue during testing, even after applying Ivan's fix and setting ops.reserve=0. Subsequent reruns of the 16 failed BFG configs using Chenhao's WT-16955 PR branch did not resolve the checksum mismatch issue.
Context
- The tests were run as part of WT-16483.
- Ivan's fix and the ops.reserve=0 parameter were applied to all BFG configs.
- 16 specific BFG configs failed with checksum mismatch (see list below).
- Some BFGs also hit a stable timestamp issue tracked in WT-16577.
- Applying Chenhao's WT-16955 branch did not resolve the checksum mismatch for these configs.
Failed BFG Configs
BFG-2936972 BFG-2941549 BFG-2943810 BFG-2943813 BFG-2945545 BFG-2947243 BFG-2947244 BFG-2949069 BFG-2951649 BFG-2957917 BFG-2957919 BFG-2958556 BFG-2960142 BFG-2960146 BFG-2962211 BFG-2962213
Proposed Solution
- Further investigation is required to identify the root cause of the checksum mismatch in these 16 BFG configs.
- Review logs and test output from the failed runs for additional clues.
- Coordinate with the team to determine if additional instrumentation or debugging is needed.
Original Slack thread: Slack Thread
This ticket was generated by AI from a Slack thread.
- is related to
-
WT-16955 (Disagg=switch) Test/format data mismatch during step-down
-
- In Code Review
-
-
WT-16483 test/format (multi-node disagg) data mismatch b/w leader and follower
-
- In Code Review
-
-
WT-16577 test/format (multi-node disagg) stable timestamp must not be older than current stable timestamp
-
- Investigating
-