Details
-
Bug
-
Status: Confirmed (View Workflow)
-
Critical
-
Resolution: Unresolved
-
10.5.24, 10.11.7, 10.5.25, 10.5.26
Description
To reproduce:
Requirements:
installed docker + docker compose + attached script
|
docker compose up -d
|
# wait until both cluster are up
|
|
# create async replication
|
./init.sh
|
|
# create traffic (can be repeated multiple times)
|
./flood.sh
|
|
Node 1 will be cleared after ~ 60 entries, but entries on node 2 and 3 persists
and will be never deleted.
Reproducable on 10.5.24, 10.5.25 and 10.5.26, but not on 10.6.19
Attachments
Issue Links
- causes
-
MDEV-33977 MariaDB hangs on --wsrep-recover phase
- Confirmed
- is caused by
-
MDEV-31413 Node has been dropped from the cluster on Startup / Shutdown with async replica
- Closed
- is duplicated by
-
MDEV-35627 mysql.gtid_slave_pos gets really big
- Open
- relates to
-
MDEV-31905 WSREP GTID MODE is inconsistent
- Closed
-
MDEV-34170 table gtid_slave_pos entries never been deleted with wsrep_gtid_mode = 0
- Closed