Comment 9 for bug 1365093

Revision history for this message
Milos Blazevic (milos-blazevic) wrote :

Hi,

I'd like to report another occurrence of this particular bug, in a 3-node PXC cluster running on IO optimized VMs with 64GB of RAM, and 32 CPU cores each...

One of the nodes crashed for no apparent reason:

****
12:43:55 UTC - mysqld got signal 11 ;
This could be because you hit a bug. It is also possible that this binary
or one of the libraries it was linked against is corrupt, improperly built,
or misconfigured. This error can also be caused by malfunctioning hardware.
We will try our best to scrape up some info that will hopefully help
diagnose the problem, but since we have already crashed,
something is definitely wrong and this may fail.
Please help us make Percona XtraDB Cluster better by reporting any
bugs at https://bugs.launchpad.net/percona-xtradb-cluster

key_buffer_size=8388608
read_buffer_size=131072
max_used_connections=1256
max_threads=4002
thread_count=38
connection_count=5
It is possible that mysqld could use up to
key_buffer_size + (read_buffer_size + sort_buffer_size)*max_threads = 1602318 K bytes of memory
Hope that's ok; if not, decrease some variables in the equation.

Thread pointer: 0x37e70130
Attempting backtrace. You can use the following information to find out
where mysqld died. If you see no messages after this, something went
terribly wrong...
160217 16:43:55 mysqld_safe Number of processes running now: 0
160217 16:43:55 mysqld_safe WSREP: not restarting wsrep node automatically
160217 16:43:55 mysqld_safe mysqld from pid file /var/lib/mysql/pxc-01.pid ended
160217 17:00:11 mysqld_safe Starting mysqld daemon with databases from /var/lib/mysql
160217 17:00:11 mysqld_safe WSREP: Running position recovery with --log_error='/var/lib/mysql/wsrep_recovery.QF2aWv' --pid-file='/var/lib/mysql/pxc-01-recover.pid'
2016-02-17 17:00:12 0 [Warning] TIMESTAMP with implicit DEFAULT value is deprecated. Please use --explicit_defaults_for_timestamp server option (see documentation for more details).
2016-02-17 17:00:12 0 [Note] /usr/sbin/mysqld (mysqld 5.6.26-74.0-56-log) starting as process 128623 ...
160217 17:00:20 mysqld_safe WSREP: Recovered position 8a89896c-a304-11e5-bed2-1e4a9bc0b937:14497921
Log of wsrep recovery (--wsrep-recover):
2016-02-17 17:00:12 128623 [Note] Plugin 'FEDERATED' is disabled.
2016-02-17 17:00:12 128623 [Note] InnoDB: Using atomics to ref count buffer pool pages
****

The cluster had been running stabily for ~6 months on a less powerful VMs, only to be upgraded to a newer version of XtraDB cluster and more powerful VMs two months ago... and this is the first and only crash of the kind that had happened.

Unfortunately, there's not much more to add - we didn't have oprofile or gdb attached to mysql server process on any of the nodes, and other than "pt-summary" output (dating from 10 days later) for all 3 nodes, which I can provide on request, I doubt I can give any more useful info.

Interestingly enough, the segfault message is the same:

Feb 17 16:43:55 pxc-01 kernel: mysqld[123325]: segfault at 55 ip 00007f6ad0e81867 sp 00007f61cb6d9b00 error 4 in libgcc_s-4.4.7-20120601.so.1[7f6ad0e72000+16000]

Regards,
Milos.