Server crash in 5.5.49

[navigator93]

Server version: 5.5.49-37.9-log Percona Server (GPL), Release 37.9, Revision efa0073

Final stmt before crash:
INSERT INTO user_available_scope_sets (user_id, scope_set_id) VALUES (1, 'invalid') ON DUPLICATE KEY UPDATE scope_set_id=scope_set_id, user_id=user_id;

Haven't been able to reproduce this using just that query. But the crash happens during unit tests of internal code that does a lot of building and tearing down of schemas to test all our data abstraction API endpoints.

Table structure:
DROP TABLE IF EXISTS `scope_set_masks`;
/*!40101 SET @saved_cs_client = @@character_set_client */;
/*!40101 SET character_set_client = utf8 */;
CREATE TABLE `scope_set_masks` (
  `id` int(11) NOT NULL AUTO_INCREMENT,
  `user_id` int(11) NOT NULL,
  `scope_set_id` varchar(255) NOT NULL,
  `bucket_0` bigint(20) DEFAULT NULL,
  `bucket_1` bigint(20) DEFAULT NULL,
  `bucket_2` bigint(20) DEFAULT NULL,
  `bucket_3` bigint(20) DEFAULT NULL,
  `bucket_4` bigint(20) DEFAULT NULL,
  `bucket_5` bigint(20) DEFAULT NULL,
  PRIMARY KEY (`id`,`scope_set_id`),
  UNIQUE KEY `scope_set_id_1` (`scope_set_id`)
) ENGINE=InnoDB DEFAULT CHARSET=utf8;
/*!40101 SET character_set_client = @saved_cs_client */;

DROP TABLE IF EXISTS `user_available_scope_sets`;
/*!40101 SET @saved_cs_client = @@character_set_client */;
/*!40101 SET character_set_client = utf8 */;
CREATE TABLE `user_available_scope_sets` (
  `user_id` int(11) NOT NULL AUTO_INCREMENT,
  `scope_set_id` varchar(255) NOT NULL,
  PRIMARY KEY (`user_id`,`scope_set_id`),
  UNIQUE KEY `scope_set_id` (`scope_set_id`),
  CONSTRAINT `fk_user_available_scope_set_scope_set` FOREIGN KEY (`scope_set_id`) REFERENCES `scope_set_masks` (`scope_set_id`)
) ENGINE=InnoDB DEFAULT CHARSET=utf8;
/*!40101 SET character_set_client = @saved_cs_client */;

Crash log:

160520 23:09:40 Percona XtraDB (http://www.percona.com) 5.5.49-37.9 started; log sequence number 21857081
160520 23:09:40 [Note] Event Scheduler: Loaded 0 events
160520 23:09:40 [Note] /usr/sbin/mysqld: ready for connections.
Version: '5.5.49-37.9' socket: '/var/lib/mysql/mysql.sock' port: 3306 Percona Server (GPL), Release 37.9, Revision efa0073
23:11:08 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 Server better by reporting any
bugs at http://bugs.percona.com/

key_buffer_size=8388608
read_buffer_size=131072
max_used_connections=4
max_threads=153
thread_count=3
connection_count=3
It is possible that mysqld could use up to
key_buffer_size + (read_buffer_size + sort_buffer_size)*max_threads = 343009 K bytes of memory
Hope that's ok; if not, decrease some variables in the equation.

Thread pointer: 0x263f2e0
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 = 7f7b54480d78 thread_stack 0x40000
/usr/sbin/mysqld(my_print_stacktrace+0x35)[0x7ce145]
/usr/sbin/mysqld(handle_fatal_signal+0x4b4)[0x6a2ab4]
/lib64/libpthread.so.0(+0xf790)[0x7f7b5b22e790]
/usr/sbin/mysqld[0x7ed0fb]
/usr/sbin/mysqld(_ZN7handler11get_dup_keyEi+0x49)[0x6a3149]
/usr/sbin/mysqld(_ZN7handler11print_errorEii+0x14a)[0x6a6d5a]
/usr/sbin/mysqld(_Z12write_recordP3THDP5TABLEP12st_copy_info+0xcb)[0x58207b]
/usr/sbin/mysqld(_Z12mysql_insertP3THDP10TABLE_LISTR4ListI4ItemERS3_IS5_ES6_S6_15enum_duplicatesb+0x7d4)[0x588614]
/usr/sbin/mysqld(_Z21mysql_execute_commandP3THD+0xbc6)[0x597fd6]
/usr/sbin/mysqld(_Z11mysql_parseP3THDPcjP12Parser_state+0x33b)[0x59c39b]
/usr/sbin/mysqld(_Z16dispatch_command19enum_server_commandP3THDPcj+0x1dae)[0x59ea1e]
/usr/sbin/mysqld(_Z24do_handle_one_connectionP3THD+0x13f)[0x63f39f]
/usr/sbin/mysqld(handle_one_connection+0x51)[0x63f481]
/lib64/libpthread.so.0(+0x7a51)[0x7f7b5b226a51]
/lib64/libc.so.6(clone+0x6d)[0x7f7b5993f93d]

Trying to get some variables.
Some pointers may be invalid and cause the dump to abort.
Query (7f7b14004b90): is an invalid pointer
Connection ID (thread ID): 5
Status: NOT_KILLED

You may download the Percona Server operations manual by visiting
http://www.percona.com/software/percona-server/. You may find information
in the manual which will help you identify the cause of the crash.

[Vasya Akulov (vasikby)]

I have the same problem on multiple servers :/

For example:

Server version: 5.5.49-37.9-log Percona Server (GPL), Release 37.9, Revision efa0073

22:12:05 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 Server better by reporting any
bugs at http://bugs.percona.com/

key_buffer_size=1098907648
read_buffer_size=131072
max_used_connections=8
max_threads=202
thread_count=8
connection_count=8
It is possible that mysqld could use up to
key_buffer_size + (read_buffer_size + sort_buffer_size)*max_threads = 27578050 K bytes of memory
Hope that's ok; if not, decrease some variables in the equation.

Thread pointer: 0x7faab80288b0
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 = 7faabfd09e60 thread_stack 0x30000
/usr/sbin/mysqld(my_print_stacktrace+0x2e)[0x7d6ebe]
/usr/sbin/mysqld(handle_fatal_signal+0x491)[0x6bd351]
/lib/x86_64-linux-gnu/libpthread.so.0(+0xf0a0)[0x7fabb56770a0]
/usr/sbin/mysqld[0x84a501]
/usr/sbin/mysqld(_ZN7handler11get_dup_keyEi+0x49)[0x6c1c19]
/usr/sbin/mysqld(_ZN7handler11print_errorEii+0x1da)[0x6c1e2a]
/usr/sbin/mysqld(_Z12write_recordP3THDP5TABLEP12st_copy_info+0x15b)[0x5adb5b]
/usr/sbin/mysqld(_Z12mysql_insertP3THDP10TABLE_LISTR4ListI4ItemERS3_IS5_ES6_S6_15enum_duplicatesb+0x904)[0x5aff54]
/usr/sbin/mysqld(_Z21mysql_execute_commandP3THD+0x2978)[0x5c3d78]
/usr/sbin/mysqld(_Z11mysql_parseP3THDPcjP12Parser_state+0x2e1)[0x5c8061]
/usr/sbin/mysqld(_Z16dispatch_command19enum_server_commandP3THDPcj+0x1e2b)[0x5c9f7b]
/usr/sbin/mysqld(_Z24do_handle_one_connectionP3THD+0x13f)[0x6663bf]
/usr/sbin/mysqld(handle_one_connection+0x51)[0x6664b1]
/lib/x86_64-linux-gnu/libpthread.so.0(+0x6b50)[0x7fabb566eb50]
/lib/x86_64-linux-gnu/libc.so.6(clone+0x6d)[0x7fabb3b70fbd]

Trying to get some variables.
Some pointers may be invalid and cause the dump to abort.
Query (6feb430): INSERT INTO `clients_visits` (id_client,last_visit) VALUES (9730481,NOW()) ON DUPLICATE KEY UPDATE last_visit=NOW()
Connection ID (thread ID): 1583
Status: NOT_KILLED

You may download the Percona Server operations manual by visiting
http://www.percona.com/software/percona-server/. You may find information
in the manual which will help you identify the cause of the crash.
160701 01:12:05 mysqld_safe Number of processes running now: 0
160701 01:12:05 mysqld_safe mysqld restarted
160701 1:12:05 [Note] /usr/sbin/mysqld (mysqld 5.5.49-37.9-log) starting as process 12641 ...
160701 1:12:06 [Note] Plugin 'FEDERATED' is disabled.
160701 1:12:06 InnoDB: The InnoDB memory heap is disabled
160701 1:12:06 InnoDB: Mutexes and rw_locks use GCC atomic builtins
160701 1:12:06 InnoDB: Compressed tables use zlib 1.2.7
160701 1:12:06 InnoDB: Using Linux nat...

[Shahriyar Rzayev (rzayev-sehriyar)]

Percona now uses JIRA for bug reports so this bug report is migrated to: https://jira.percona.com/browse/PS-3445