Details
-
Bug
-
Status: Closed (View Workflow)
-
Major
-
Resolution: Fixed
-
23.08.6
-
None
-
CentOS Stream release 8
 4.18.0-448.el8.x86_64
MariaDB = 10.11.7
-
MXS-SPRINT-215, MXS-SPRINT-216, MXS-SPRINT-217, MXS-SPRINT-218, MXS-SPRINT-219, MXS-SPRINT-220, MXS-SPRINT-221
Description
I got next critical issue with maxcale binlogrouter and It repeated twice already
I have the schema:
masters ( 2 hosts ) --> maxscale binlog --> replicas (few hosts)
When I do switchover master to other host that cause problem that replicas after binlogrouter stop getting new updates . Replication process Slave_IO & Slave_SQL are active on these replicas and replicas lools like alive , but Gtid_IO_Pos not changing .
Only restart of replica (stop slave ; start slave ; ) helps to continue replication.
On maxscale binllog all looks fine . It's getting update from new master , Slave_IO & Slave_SQL are active and Gtid_IO_Pos position increases
I've attached log from maxscale binlog and few replicas