Details
-
Bug
-
Status: Closed (View Workflow)
-
Critical
-
Resolution: Duplicate
-
10.11.10
Description
After upgrading to MariaDB 10.11.10 from 10.5.18, we are noticing a significant delay during mariadb shutdown more than 10 minutes and sometimes even more while there is nothing actively running on the database. It usually took around 10 seconds to shutdown mariadb in 10.5.
We use mariadb galera cluster.
To shutdown we use, systemctl stop mariadb
Shutdown Logs are given below:
2025-01-06 5:13:45 1 [Note] WSREP: Non-primary view
2025-01-06 5:13:45 1 [Note] WSREP: Server status change disconnecting -> disconnected
2025-01-06 5:13:45 1 [Note] WSREP: wsrep_notify_cmd is not defined, skipping notification.
2025-01-06 5:13:45 0 [Note] WSREP: recv_thread() joined.
2025-01-06 5:13:45 0 [Note] WSREP: Closing send queue.
2025-01-06 5:13:45 0 [Note] WSREP: Closing receive queue.
2025-01-06 5:13:45 1 [Note] WSREP: Applier thread exiting ret: 0 thd: 1
2025-01-06 5:25:06 0 [Note] Event Scheduler: Killing the scheduler thread, thread id 3
2025-01-06 5:25:06 0 [Note] Event Scheduler: Waiting for the scheduler thread to reply
2025-01-06 5:25:06 0 [Note] Event Scheduler: Stopped
2025-01-06 5:25:06 0 [Note] WSREP: dtor state: CLOSED
2025-01-06 5:25:06 0 [Note] WSREP: MemPool(TrxHandleSlave): hit ratio: 0, misses: 1, in use: 1, in pool: 0
2025-01-06 5:25:06 0 [Note] WSREP: mon: entered 1 oooe fraction 0 oool fraction 0
2025-01-06 5:25:06 0 [Note] WSREP: mon: entered 1 oooe fraction 0 oool fraction 0
2025-01-06 5:25:06 0 [Note] WSREP: mon: entered 4 oooe fraction 0 oool fraction 0
2025-01-06 5:25:06 0 [Note] WSREP: cert index usage at exit 5
2025-01-06 5:25:06 0 [Note] WSREP: cert trx map usage at exit 1
2025-01-06 5:25:06 0 [Note] WSREP: deps set usage at exit 0
2025-01-06 5:25:06 0 [Note] WSREP: avg deps dist 1
2025-01-06 5:25:06 0 [Note] WSREP: avg cert interval 0
2025-01-06 5:25:06 0 [Note] WSREP: cert index size 5
2025-01-06 5:25:06 0 [Note] WSREP: Service thread queue flushed.
2025-01-06 5:25:06 0 [Note] WSREP: wsdb trx map usage 0 conn query map usage 0
2025-01-06 5:25:06 0 [Note] WSREP: MemPool(LocalTrxHandle): hit ratio: 0, misses: 1, in use: 0, in pool: 1
2025-01-06 5:25:06 0 [Note] WSREP: Flushing memory map to disk...
2025-01-06 5:25:06 0 [Note] WSREP: Deinitializing allowlist service v1
2025-01-06 5:25:06 0 [Note] InnoDB: FTS optimize thread exiting.
2025-01-06 5:25:06 0 [Note] InnoDB: Starting shutdown...
2025-01-06 5:25:06 0 [Note] InnoDB: Dumping buffer pool(s) to /mysql/ib_buffer_pool
2025-01-06 5:25:06 0 [Note] InnoDB: Buffer pool(s) dump completed at 250106 5:25:06
2025-01-06 5:25:06 0 [Note] InnoDB: Removed temporary tablespace data file: "./ibtmp1"
2025-01-06 5:25:06 0 [Note] InnoDB: Shutdown completed; log sequence number 546868043122; transaction id 211382546
2025-01-06 5:25:06 0 [Note] /usr/local/mysql/bin/mariadbd: Shutdown complete
It took almost 13 minutes to shutdown. It did not print anything after the thread exiting log up to next 13 minutes.
Attachments
Issue Links
- is blocked by
-
MDEV-35710 cluster node hangs during shutdown if threadpool is used
-
- Closed
-
- is duplicated by
-
MDEV-35710 cluster node hangs during shutdown if threadpool is used
-
- Closed
-