PXC donor node got stuck during SST (using thread pool plugin)
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
Percona XtraDB Cluster moved to https://jira.percona.com/projects/PXC |
Fix Released
|
Undecided
|
Raghavendra D Prabhu |
Bug Description
- a PXC cluster with 1 active node (just bootstrapped)
- another node is SST'ing with xtrabackup
- Suddenly, (I think just after the backup has been taken) the master node locks up and you get:
130730 6:03:10 [Note] WSREP: Provider paused at a6280df4-
130730 6:04:40 [ERROR] Threadpool could not create additional thread to handle queries, because the number of allowed threads was reached. Increasing 'thread_
If 'extra_port' parameter is set, you can still connect to the database with superuser account (it must be TCP connection using extra_port as TCP port) and troubleshoot the situation. A likely cause of pool blocks are clients that lock resources for long time. 'show
processlist' or 'show engine innodb status' can give additional hints.
130730 6:04:40 [Note] Threadpool has been blocked for 30 seconds
- A few minutes later, the innobackupex prepare on the JOINER was finished
- I got TOO MANY CONNECTIONS on the donor, and --extra-port wasn't set
- I tried changing max_connections using Aurimas'es method: http://
- poor mans profiler output:
458 pthread_
40 epoll_wait,
10 ??,??,?
8 pthread_
3 pthread_
2 pthread_
2 pthread_
1 select,
1 read,_IO_
1 pthread_
1 pthread_
1 pthread_
1 pthread_
1 poll,handle_
1 epoll_wait,
1 do_sigwait,
1
In order to resolve this, I killed alll the SST processes and mysql processes and started mysql again.
Raghu mentioned it could be related to some FTWRL...
Related branches
Changed in percona-xtradb-cluster: | |
status: | Triaged → Fix Committed |
Changed in percona-xtradb-cluster: | |
status: | Fix Committed → Fix Released |
Will ask the customer to add --extra-port and get SHOW PROCESSLIST and 'SHOW ENGINE INNODB STATUS' if it happens again.