[MDEV-14400] Innodb_row_lock_current_waits increases to 18446744073709551615 Created: 2017-11-15 Updated: 2017-11-22 |
|
| Status: | Open |
| Project: | MariaDB Server |
| Component/s: | Locking, Storage Engine - InnoDB |
| Affects Version/s: | 10.1.22 |
| Fix Version/s: | None |
| Type: | Bug | Priority: | Major |
| Reporter: | M.Lauterbach | Assignee: | Unassigned |
| Resolution: | Unresolved | Votes: | 0 |
| Labels: | None | ||
| Environment: |
mysql Ver 15.1 Distrib 10.1.22-MariaDB, for debian-linux-gnu (x86_64) using readline 5.2 |
||
| Attachments: |
|
| Description |
|
Innodb_row_lock_current_waits increases rapidly from 0 to 18446744073709551615 and does not lower again. As you can see, this happens very fast and it has been happening more than once. Monitoring last hour Monitoring current week Monitoring last weeks Monitoring last months
Here, I dumped my environment.
|
| Comments |
| Comment by Elena Stepanova [ 2017-11-20 ] | ||||||||||||||||
|
SelectNameFromDual, on some reason show status that you pasted only contains a few wsrep variables. Please attach the whole output, and also your cnf file(s). Thanks. | ||||||||||||||||
| Comment by M.Lauterbach [ 2017-11-22 ] | ||||||||||||||||
|
_elenst said:
I created a new status and attached it as a file. show_status_20171122-0941 _elenst said:
Your assumption is correct. I am using Galera cluster . And currently this problem only happens on one of the cluster nodes. Since the server entered this state, I did not restart the node. Up to now, every night this state flickers.
_elenst said:
Due to my monitoring, I got several insights to the time, when the state changes. Innodb data pending fsyncs
Threads connected Threads running Process runtime Entries in mysqld-error.log
I asked my developers colleagues if they might have a clue, which kind of statements run at this time. Up to now, they do not have a guess. |