Percona server crashes during insert into mysql.servers table

[Vyacheslav Gritsenko]

Actually, this bug does exist in MySQL-5.1 and MySQL-5.5 servers as well, but just in case I decided to report it not only to MySQL but to Percona too.

How to reproduce.
Just insert some value into mysql.servers table, but using *host* column only. You should leave other columns empty:

mysql> select * from mysql.servers;
Empty set (0.00 sec)

mysql> insert into mysql.servers set host='sfdsfsfsdf';
Query OK, 1 row affected (0.00 sec)

mysql> flush privileges;
ERROR 2013 (HY000): Lost connection to MySQL server during query
mysql> status;
ERROR 2006 (HY000): MySQL server has gone away
No connection. Trying to reconnect...
ERROR 2002 (HY000): Can't connect to local MySQL server through socket '/tmp/mysql.sock' (2)
ERROR:
Can't connect to the server

mysql>

Server crashes and you are not able to use it anymore. It just keeps restarting until you fix(restore) mysql.servers table.
Error log entry is below:

120725 06:36:30 mysqld_safe Number of processes running now: 0
120725 06:36:30 mysqld_safe mysqld restarted
120725 6:36:30 [Note] Flashcache bypass: disabled
120725 6:36:30 [Note] Flashcache setup error is : ioctl failed

120725 6:36:30 [Note] Plugin 'FEDERATED' is disabled.
120725 6:36:30 InnoDB: The InnoDB memory heap is disabled
120725 6:36:30 InnoDB: Mutexes and rw_locks use GCC atomic builtins
120725 6:36:30 InnoDB: Compressed tables use zlib 1.2.3
120725 6:36:30 InnoDB: Using Linux native AIO
120725 6:36:30 InnoDB: Initializing buffer pool, size = 128.0M
120725 6:36:30 InnoDB: Completed initialization of buffer pool
120725 6:36:30 InnoDB: highest supported file format is Barracuda.
InnoDB: The log sequence number in ibdata files does not match
InnoDB: the log sequence number in the ib_logfiles!
120725 6:36:30 InnoDB: Database was not shut down normally!
InnoDB: Starting crash recovery.
InnoDB: Reading tablespace information from the .ibd files...
InnoDB: Restoring possible half-written data pages from the doublewrite
InnoDB: buffer...
120725 6:36:30 InnoDB: Waiting for the background threads to start
120725 6:36:31 Percona XtraDB (http://www.percona.com) 1.1.8-rel27.1 started; log sequence number 1597945
120725 6:36:31 [Note] Recovering after a crash using mysql-bin
120725 6:36:31 [Note] Starting crash recovery...
120725 6:36:31 [Note] Crash recovery finished.
120725 6:36:31 [Note] Server hostname (bind-address): '0.0.0.0'; port: 3306
120725 6:36:31 [Note] - '0.0.0.0' resolves to '0.0.0.0';
120725 6:36:31 [Note] Server socket created on IP: '0.0.0.0'.
10:36:31 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=402653184
read_buffer_size=2097152
max_used_connections=0
max_threads=151
thread_count=0
connection_count=0
It is possible that mysqld could use up to
key_buffer_size + (read_buffer_size + sort_buffer_size)*max_threads = 1013562 K bytes of memory
Hope that's ok; if not, decrease some variables in the equation.

Thread pointer: 0x20cbbc0
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 = 7fffc796eef8 thread_stack 0x40000
/usr/local/Percona-Server-5.5.25a-rel27.1-277.Linux.x86_64/bin/mysqld(my_print_stacktrace+0x35)[0x7d6755]
/usr/local/Percona-Server-5.5.25a-rel27.1-277.Linux.x86_64/bin/mysqld(handle_fatal_signal+0x3e1)[0x693621]
/lib64/libpthread.so.0(+0xf500)[0x7f7a264d9500]
/lib64/libc.so.6(+0x131aff)[0x7f7a259a2aff]
/usr/local/Percona-Server-5.5.25a-rel27.1-277.Linux.x86_64/bin/mysqld[0x61b3b9]
/usr/local/Percona-Server-5.5.25a-rel27.1-277.Linux.x86_64/bin/mysqld(_Z14servers_reloadP3THD+0x1c1)[0x61c2f1]
/usr/local/Percona-Server-5.5.25a-rel27.1-277.Linux.x86_64/bin/mysqld(_Z12servers_initb+0x126)[0x61c456]
/usr/local/Percona-Server-5.5.25a-rel27.1-277.Linux.x86_64/bin/mysqld(_Z11mysqld_mainiPPc+0x11ec)[0x509e2c]
/lib64/libc.so.6(__libc_start_main+0xfd)[0x7f7a2588fcdd]
/usr/local/Percona-Server-5.5.25a-rel27.1-277.Linux.x86_64/bin/mysqld[0x500299]

Trying to get some variables.
Some pointers may be invalid and cause the dump to abort.
Query (0): is an invalid pointer
Connection ID (thread ID): 0
Status: NOT_KILLED

[Raghavendra D Prabhu (raghavendra-prabhu)]

The upstream bug seems to have marked closed but I can still repeat on 5.5.27 , however, it links to Oracle internal bug Bug#14220942 whose status is not known (I also checked bzr log where it is not marked).

[Shahriyar Rzayev (rzayev-sehriyar)]

Percona now uses JIRA for bug reports so this bug report is migrated to: https://jira.percona.com/browse/PS-1927