[MDEV-10276] Crash signal 11 during logrotate Created: 2016-06-23  Updated: 2020-08-25  Resolved: 2017-12-13

Status: Closed
Project: MariaDB Server
Component/s: Galera, Replication
Affects Version/s: 10.1.13
Fix Version/s: 10.1.30

Type: Bug Priority: Critical
Reporter: Maciej Radzikowski Assignee: Sachin Setiya (Inactive)
Resolution: Duplicate Votes: 2
Labels: None
Environment:

Debian 8

Issue Links:
Duplicate
is duplicated by MDEV-11550 Crash when running flush logs Closed
Relates
relates to MDEV-9510 Segmentation fault in binlog thread c... Closed
relates to MDEV-11610 Logrotate to only FLUSH LOCAL ERROR L... Closed

 Description   

We have Galera multi-master cluster with 3 nodes. We are experiencing random crashes of mysqld process on some nodes when logrotate is executed.

Daily logrotate is executed at 6:25. At exactly this time database sometimes crashes. After update from 10.0 to 10.1.13 for few days everything was ok. Then database crashed on Node1 two times in a row, day after day. At this time I've disabled logrotate for mysql for a few days on Node1 and there were no crashes. I've separated this node from others, disabled replication and executed logrotate manually few times. Nothing happend. I've restored node back to the cluster and enabled logrotate. Next day crash happend, but on Node2. Next day - nothing. Day after that - another crash, but on both Node1 and Node2.

On similar testing environment those errors don't occure.

Logs are configured to go to /var/log/mysql/mariadb-error.log and /var/log/mysql/mariadb-slow.log. General query log is not enabled and cannot be.

/etc/logrotate.d/mysql-server content:

/var/log/mysql/mariadb-error.log /var/log/mysql/mariadb-slow.log {
 
        daily
 
        rotate 7
 
        missingok
 
        create 640 mysql adm
 
        compress
 
        sharedscripts
 
        postrotate
 
          test -x /usr/bin/mysqladmin || exit 0
 
 
 
          if [ -f `my_print_defaults --mysqld | grep -oP "pid-file=\K[^$]+"` ]; then
 
            # If this fails, check debian.conf!
 
            mysqladmin --defaults-file=/etc/mysql/debian.cnf flush-logs
 
          fi
 
        endscript
 
}

(standard except logs location, but those files exists and paths are correct)

I don't know if it's related to Galera or not.

Logs after restoration from crash:

160623  6:25:03 [ERROR] 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.
 
 
 
To report this bug, see https://mariadb.com/kb/en/reporting-bugs
 
 
 
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.
 
 
 
Server version: 10.1.13-MariaDB-1~jessie
 
key_buffer_size=134217728
 
read_buffer_size=2097152
 
max_used_connections=7
 
max_threads=102
 
thread_count=23
 
It is possible that mysqld could use up to
 
key_buffer_size + (read_buffer_size + sort_buffer_size)*max_threads = 759828 K  bytes of memory
 
Hope that's ok; if not, decrease some variables in the equation.
 
 
 
Thread pointer: 0x0x7fd4c3c77008
 
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...
 
stack_bottom = 0x7fd571ee91f8 thread_stack 0x48400
 
/usr/sbin/mysqld(my_print_stacktrace+0x2e)[0x7fd576ce6e3e]
 
/usr/sbin/mysqld(handle_fatal_signal+0x34d)[0x7fd57682870d]
 
/lib/x86_64-linux-gnu/libpthread.so.0(+0xf8d0)[0x7fd575e568d0]
 
/usr/sbin/mysqld(_ZN13MYSQL_BIN_LOG21do_checkpoint_requestEm+0x98)[0x7fd5768ddbd8]
 
/usr/sbin/mysqld(_ZN13MYSQL_BIN_LOG20checkpoint_and_purgeEm+0x11)[0x7fd5768ddc01]
 
/usr/sbin/mysqld(_ZN13MYSQL_BIN_LOG16rotate_and_purgeEb+0x7e)[0x7fd5768e009e]
 
/usr/sbin/mysqld(_Z20reload_acl_and_cacheP3THDyP10TABLE_LISTPi+0x131)[0x7fd57678ce51]
 
/usr/sbin/mysqld(_Z21mysql_execute_commandP3THD+0x1405)[0x7fd57669eab5]
 
/usr/sbin/mysqld(_Z11mysql_parseP3THDPcjP12Parser_state+0x26e)[0x7fd5766a6dbe]
 
/usr/sbin/mysqld(+0x4205b9)[0x7fd5766a75b9]
 
/usr/sbin/mysqld(_Z16dispatch_command19enum_server_commandP3THDPcj+0x1a24)[0x7fd5766a9614]
 
/usr/sbin/mysqld(_Z10do_commandP3THD+0x16e)[0x7fd5766aa40e]
 
/usr/sbin/mysqld(_Z24do_handle_one_connectionP3THD+0x182)[0x7fd576773752]
 
/usr/sbin/mysqld(handle_one_connection+0x40)[0x7fd576773910]
 
/lib/x86_64-linux-gnu/libpthread.so.0(+0x80a4)[0x7fd575e4f0a4]
 
/lib/x86_64-linux-gnu/libc.so.6(clone+0x6d)[0x7fd573ffa87d]
 
 
 
Trying to get some variables.
 
Some pointers may be invalid and cause the dump to abort.
 
Query (0x7fd55f1f3020): flush logs
 
Connection ID (thread ID): 1507663
 
Status: NOT_KILLED
 
 
 
Optimizer switch: index_merge=on,index_merge_union=on,index_merge_sort_union=on,index_merge_intersection=on,index_merge_sort_intersection=off,engine_condition_pushdown=off,index_condition_pushdown=on,derived_merge=on,derived_with_keys=on,firstmatch=on,loosescan=on,materialization=on,in_to_exists=on,semijoin=on,partial_match_rowid_merge=on,partial_match_table_scan=on,subquery_cache=on,mrr=off,mrr_cost_based=off,mrr_sort_keys=off,outer_join_with_cache=on,semijoin_with_cache=on,join_cache_incremental=on,join_cache_hashed=on,join_cache_bka=on,optimize_join_buffer_size=off,table_elimination=on,extended_keys=on,exists_to_in=on



 Comments   
Comment by Reto [ 2016-09-20 ]

Hi guys, we have the same issue with version 10.1.12. Is this issue in a future release aready fixed?

We have exactly the same setup.

Comment by Mark Anstice [ 2017-02-07 ]

Hi, we have the same issue with 10.1.13 but with Ubuntu 14.04.4 LTS, the same node of our three node Galera cluster has crashed 3 times roughly 3 weeks apart.

Comment by Patrick Rono [ 2017-02-17 ]

Experiencing the same issue on 10.1.17 running on red-hat RHEL 6.6. The stack:

170216 0:00:01 [ERROR] 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.

To report this bug, see https://mariadb.com/kb/en/reporting-bugs

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.

Server version: 10.1.17-MariaDB
key_buffer_size=16777216
read_buffer_size=262144
max_used_connections=32
max_threads=502
thread_count=22
It is possible that mysqld could use up to
key_buffer_size + (read_buffer_size + sort_buffer_size)*max_threads = 412210 K bytes of memory
Hope that's ok; if not, decrease some variables in the equation.

Thread pointer: 0x0x7f52f0bb4008
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...
stack_bottom = 0x7f54bc440db8 thread_stack 0x40000
//opt/app/mysql/product/mariadb-10.1.17/bin/mysqld(my_print_stacktrace+0x2e)[0xc0717e]
//opt/app/mysql/product/mariadb-10.1.17/bin/mysqld(handle_fatal_signal+0x4ba)[0x7667ea]
/lib64/libpthread.so.0[0x380520f710]
//opt/app/mysql/product/mariadb-10.1.17/bin/mysqld(_ZN13MYSQL_BIN_LOG16rotate_and_purgeEb+0x101)[0x8428d1]
//opt/app/mysql/product/mariadb-10.1.17/bin/mysqld(_Z20reload_acl_and_cacheP3THDyP10TABLE_LISTPi+0x668)[0x6bbff8]
//opt/app/mysql/product/mariadb-10.1.17/bin/mysqld(_Z21mysql_execute_commandP3THD+0x1343)[0x5b2673]
//opt/app/mysql/product/mariadb-10.1.17/bin/mysqld(_Z11mysql_parseP3THDPcjP12Parser_state+0x2b4)[0x5bac44]
//opt/app/mysql/product/mariadb-10.1.17/bin/mysqld[0x5bacd8]
//opt/app/mysql/product/mariadb-10.1.17/bin/mysqld(_Z16dispatch_command19enum_server_commandP3THDPcj+0x14e5)[0x5bcdb5]
//opt/app/mysql/product/mariadb-10.1.17/bin/mysqld(_Z10do_commandP3THD+0x2a7)[0x5be087]
//opt/app/mysql/product/mariadb-10.1.17/bin/mysqld(_Z24do_handle_one_connectionP3THD+0x183)[0x68d2f3]
//opt/app/mysql/product/mariadb-10.1.17/bin/mysqld(handle_one_connection+0x42)[0x68d512]
//opt/app/mysql/product/mariadb-10.1.17/bin/mysqld[0xb09699]
/lib64/libpthread.so.0[0x38052079d1]
/lib64/libc.so.6(clone+0x6d)[0x3804ee88fd]

Trying to get some variables.
Some pointers may be invalid and cause the dump to abort.
Query (0x7f54b64f3020): flush logs
Connection ID (thread ID): 12787
Status: NOT_KILLED

Optimizer switch: index_merge=on,index_merge_union=on,index_merge_sort_union=on,index_merge_intersection=on,index_merge_sort_intersection=off,engine_condition_pushdown=off,index_condition_pushdown=on,derived_merge=on,derived_with_keys=on,firstmatch=on,loosescan=on,materialization=on,in_to_exists=on,semijoin=on,partial_match_rowid_merge=on,partial_match_table_scan=on,subquery_cache=on,mrr=off,mrr_cost_based=off,mrr_sort_keys=off,outer_join_with_cache=on,semijoin_with_cache=on,join_cache_incremental=on,join_cache_hashed=on,join_cache_bka=on,optimize_join_buffer_size=off,table_elimination=on,extended_keys=on,exists_to_in=on,orderby_uses_equalities=off

The manual page at http://dev.mysql.com/doc/mysql/en/crashing.html contains
information that should help you find out what is causing the crash.
170216 00:00:01 mysqld_safe Number of processes running now: 0
170216 00:00:01 mysqld_safe WSREP: not restarting wsrep node automatically
170216 00:00:01 mysqld_safe mysqld from pid file //opt/app/mysql/data/datadir/servername.pid ended

Comment by Daniel Black [ 2017-03-01 ]

Folks, In light of the number of log-rotation based stalls, crashes I created the workaround in MDEV-11610. The code for it is here https://github.com/MariaDB/server/pull/323. In essence it removes rotation of binary and relays logs from the logrotate script and injects a SET LOCAL SQL_BIN_LOG=0. Error, general, engine and slow query logs are still rotated.

If you have any issues with this patch please note it on the github PR or in MDEV-11610.
If you approve of this intermediary solution please vote for issue MDEV-11610.

Comment by Andrei Elkin [ 2017-12-13 ]

Fixed by MDEV-9510.

Generated at Thu Feb 08 07:40:59 UTC 2024 using Jira 8.20.16#820016-sha1:9d11dbea5f4be3d4cc21f03a88dd11d8c8687422.