Details
-
Bug
-
Status: In Review (View Workflow)
-
Critical
-
Resolution: Unresolved
-
10.11, 12.3
-
None
-
Can result in unexpected behaviour
-
Q1/2026 Server Development, Q1/2026 Server Maintenance
Description
MDEV-5804 implemented @@gtid_ignore_duplicates on serial replication by overwriting @@sql_slave_skip_counter to 1 regardless of its prior value (be it 0 or 100).
The result of the attached test currently shows a drop from 100 to 0.
Because SHOW SLAVE STATUS waits for a mutex that’s locked while @@gtid_ignore_duplicates takes effect, its Skip_Counter cannot show a temporary increase from 0 to 1 (at least for now).
@@sql_slave_skip_counter, however, does not require this mutex and so can independently reveal this increase.
MDEV-4937 almost banned @@sql_slave_skip_counter in GTID mode, but was revised to fix it.
Attachments
Issue Links
- blocks
-
MDEV-4698 With GTID replication, relay logs cannot be relied upon while purging binary logs on master
-
- In Progress
-
- causes
-
MDEV-30458 Consolidate Serial Replica to Parallel Replica with 1 Worker Thread
-
- Open
-
-
MDEV-38672 Consolidate replica-side event group filtering
-
- Open
-
- is caused by
-
MDEV-5804 If same GTID is received on multiple master connections in multi-source replication, the event is double-executed causing corruption or replication failure
-
- Closed
-
- relates to
-
MDEV-4937 sql_slave_skip_counter does not work with GTID
-
- Closed
-