Details
-
Bug
-
Status: Open (View Workflow)
-
Major
-
Resolution: Unresolved
-
10.11.10
-
None
-
None
-
None
Description
Hi,
I am migrating a 10.4 galera cluster with replica's to a 10.11 cluster with replica's.
The replication from 10.4 to 10.11 works without issues. On the 10.11 cluster I have wsrep_gtid_mode on, the same wsrep_gtid_domain_id on each node, and a different gtid_domain_id on each node. On the 10.4 cluster I have nothing configured, I use the default gtid_domain_id (0) for all transactions and have wsrep_gtid_mode off.
The replication from 10.4 to 10.11 works as expected (from 10.11 back to 10.4 works as well, but that stream is not running anymore). When I add a 10.11 replica to the 10.11 cluster, I get these errors:
Could not execute Write_rows_v1 event on table mysql.gtid_slave_pos; Duplicate entry '0-64337585' for key 'PRIMARY', Error_code: 1062; handler error HA_ERR_FOUND_DUPP_KEY; the event's master log mariadb.000093, end_log_pos 848979524 |
I tried to skip a few of these transactions but they did not work. I tried to switch to gtid mode replication but that did not work either.
Any idea how to fix that? Maybe there is a work-around?
Attachments
Issue Links
- relates to
-
MDEV-33610 Circular replication breaks after upgrading
- Open