server crashes with userstats ON

Bug #1040614 reported by Marc Castrovinci
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
Percona Server moved to https://jira.percona.com/projects/PS
New
Undecided
Unassigned

Bug Description

With userstat=1, mysql attempts to start and dies immediately. MySQL had been restarted before with userstat=1 without a problem, but now it is having problems. Only thing different is the box has swap now. I've verified this on multiple servers. Commenting out userstat=1 causes the server to start fine. It looks like this bug, https://bugs.launchpad.net/percona-server/+bug/1008278, which was supposed to be fixed in 5.5.25. I'm not using any type of replication setup. My version is: Server version: 5.5.25a-27.1-log Percona Server (GPL), Release 27.1

Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: 20:46:43 UTC - mysqld got signal 11 ;
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: This could be because you hit a bug. It is also possible that this binary
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: or one of the libraries it was linked against is corrupt, improperly built,
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: or misconfigured. This error can also be caused by malfunctioning hardware.
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: We will try our best to scrape up some info that will hopefully help
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: diagnose the problem, but since we have already crashed,
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: something is definitely wrong and this may fail.
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: Please help us make Percona Server better by reporting any
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: bugs at http://bugs.percona.com/
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld:
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: key_buffer_size=26214400
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: read_buffer_size=131072
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: max_used_connections=0
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: max_threads=1200
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: thread_count=0
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: connection_count=0
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: It is possible that mysqld could use up to
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: key_buffer_size + (read_buffer_size + sort_buffer_size)*max_threads = 793909 K bytes of memory
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: Hope that's ok; if not, decrease some variables in the equation.
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld:
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: Thread pointer: 0x7fff9932e600
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: Attempting backtrace. You can use the following information to find out
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: where mysqld died. If you see no messages after this, something went
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: terribly wrong...
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: stack_bottom = 7fff99331710 thread_stack 0x30000
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: /usr/sbin/mysqld(my_print_stacktrace+0x35)[0x7d0b75]
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: /usr/sbin/mysqld(handle_fatal_signal+0x4a4)[0x69d004]
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: /lib/libpthread.so.0(+0xf8f0)[0x7f3ba90ff8f0]
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: /usr/sbin/mysqld(my_hash_insert+0x331)[0x7b84d1]
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: /usr/sbin/mysqld(_ZN7handler25update_global_table_statsEv+0x2f5)[0x6a3755]
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: /usr/sbin/mysqld(_Z18close_thread_tableP3THDPP5TABLE+0xad)[0x55d6dd]
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: /usr/sbin/mysqld(_Z19close_thread_tablesP3THD+0x18b)[0x55dccb]
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: /usr/sbin/mysqld(_Z18close_mysql_tablesP3THD+0x11)[0x55dd51]
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: /usr/sbin/mysqld[0x5a43b5]
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: /usr/sbin/mysqld(_Z11plugin_initPiPPci+0x9ca)[0x5a75da]
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: /usr/sbin/mysqld[0x52127a]
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: /usr/sbin/mysqld(_Z11mysqld_mainiPPc+0x5ed)[0x524e3d]
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: /lib/libc.so.6(__libc_start_main+0xfd)[0x7f3ba82c1c4d]
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: /usr/sbin/mysqld[0x51a515]

Aug 22 16:46:43 DEVELOP-DB daemon err mysqld:
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: Trying to get some variables.
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: Some pointers may be invalid and cause the dump to abort.
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: Query (0): is an invalid pointer
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: Connection ID (thread ID): 0
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: Status: NOT_KILLED
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld:
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: You may download the Percona Server operations manual by visiting
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: http://www.percona.com/software/percona-server/. You may find information
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: in the manual which will help you identify the cause of the crash.
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld:
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: The "--memlock" argument, which was enabled, uses system calls that are
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: unreliable and unstable on some operating systems and operating-system versions (notably, some versions of Linux).
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: This crash could be due to use of those buggy OS calls.
Aug 22 16:46:43 DEVELOP-DB daemon err mysqld: You should consider whether you really need the "--memlock" parameter and/or consult the OS distributer about "mlockall" bugs.
Aug 22 16:46:43 DEVELOP-DB daemon notice mysqld_safe: mysqld from pid file /var/lib/mysql/DEVELOP-DB.pid ended

Revision history for this message
yinfeng (yinfeng-zwx) wrote :
Revision history for this message
Marc Castrovinci (marc-castrovinci) wrote :

It is very similar, however I didn't add any plugins.

Revision history for this message
Raghavendra D Prabhu (raghavendra-prabhu) wrote :

@marc,
This looks quite similar.

Can you tell us if plugin.MYD is empty? It is hard to check it now without crashing the server. So

strings $datadir/mysql/plugin.MYD should show something.

In this case like:

[root@archie mysql]# strings plugin.MYD
daemon_example
libdaemon_example.so

Revision history for this message
Marc Castrovinci (marc-castrovinci) wrote :

You are correct. Guess I did add plugins.

root@DEVELOP-DB:/var/lib/mysql/mysql# strings plugin.MYD
handlersocket
handlersocket.so

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.