[MDEV-675] LP:851404 - Deadlocks occurring with sysbench against mariadb 5.3.1 beta + innodb Created: 2011-09-15 Updated: 2013-01-17 Resolved: 2013-01-17 |
|
| Status: | Closed |
| Project: | MariaDB Server |
| Component/s: | None |
| Affects Version/s: | None |
| Fix Version/s: | None |
| Type: | Bug | ||
| Reporter: | Mark Kirkwood (Inactive) | Assignee: | Axel Schwenke |
| Resolution: | Cannot Reproduce | Votes: | 0 |
| Labels: | Launchpad | ||
| Attachments: |
|
| Description |
|
sysbench 0.4.12 with options: for 2 threads get approx 1200 tps and 0 deadlocks, but for 4, 8, 16, 32 get ~200 tps and steadily increasing numbers of deadlocks (10 up to 297). Innodb status for deadlocks shows:
A run with identically configured mariadb 5.2.7 elicits no deadlocks for any of the 2-> 32 threads. |
| Comments |
| Comment by Mark Kirkwood (Inactive) [ 2011-09-15 ] |
|
Re: Deadlocks occurring with sysbench against mariadb 5.3.1 beta + innodb Mariadb 5.3.1 built from source: $ CFLAGS='-g -O2 -fno-strength-reduce' \ Non default options in my.cnf: datadir = /data0/madata/5.3 |
| Comment by Elena Stepanova [ 2012-03-22 ] |
|
Re: Deadlocks occurring with sysbench against mariadb 5.3.1 beta + innodb Do you think there is a problem? Is there anything we need to do about it? This does look somewhat strange: UPDATE sbtest set k=k+1 where id=? 2M row locks?.. But the report is pretty old, quite possibly something has changed since then. |
| Comment by Rasmus Johansson (Inactive) [ 2012-03-22 ] |
|
Launchpad bug id: 851404 |
| Comment by Mark Kirkwood (Inactive) [ 2012-03-22 ] |
|
Re: Deadlocks occurring with sysbench against mariadb 5.3.1 beta + innodb |