TokuDB crashes after upgrade to 5.6.26-74.0 under heavy load

[Florian Preugschat]

After upgrading Percona MySQL Server and TokuDB to version 5.6.26-74.0, the TokuDB part crashes under medium to heavy load.

Our environment:
- Debian Wheezy 7.9
- Percona MySQL / TokuDB 5.6.26-74.0
- libc6 Version 2.13-38+deb7u8 (Package contains files libpthread.so.0 and libc.so.6)

Stack trace:
=========

16:32:54 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=104857600
read_buffer_size=2097152
max_used_connections=22
max_threads=2002
thread_count=10
connection_count=10
It is possible that mysqld could use up to
key_buffer_size + (read_buffer_size + sort_buffer_size)*max_threads = 6280245 K bytes of memory
Hope that's ok; if not, decrease some variables in the equation.

Thread pointer: 0x7f810f797000
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 = 7f86c8876df0 thread_stack 0x30000
/usr/sbin/mysqld(my_print_stacktrace+0x2e)[0x8b6b0e]
/usr/sbin/mysqld(handle_fatal_signal+0x491)[0x67ef31]
/lib/x86_64-linux-gnu/libpthread.so.0(+0xf0a0)[0x7f86cabbb0a0]
/usr/lib/mysql/plugin/ha_tokudb.so(+0x108a55)[0x7f8670d5ea55]
/usr/lib/mysql/plugin/ha_tokudb.so(_Z29toku_db_release_lt_key_rangesP13__toku_db_txnP17txn_lt_key_ranges+0x59)[0x7f8670d5f939]
/usr/lib/mysql/plugin/ha_tokudb.so(+0x135ab9)[0x7f8670d8bab9]
/usr/lib/mysql/plugin/ha_tokudb.so(+0x11b38b)[0x7f8670d7138b]
/usr/lib/mysql/plugin/ha_tokudb.so(+0x552f8)[0x7f8670cab2f8]
/usr/sbin/mysqld(_Z13ha_commit_lowP3THDbb+0x86)[0x5dd556]
/usr/sbin/mysqld(_ZN11TC_LOG_MMAP6commitEP3THDb+0xf5)[0x66f935]
/usr/sbin/mysqld(_Z15ha_commit_transP3THDbb+0x2f7)[0x5ddbd7]
/usr/sbin/mysqld(_Z12trans_commitP3THD+0x33)[0x76fc33]
/usr/sbin/mysqld(_Z21mysql_execute_commandP3THD+0x1a2a)[0x6f2bda]
/usr/sbin/mysqld(_Z11mysql_parseP3THDPcjP12Parser_state+0x598)[0x6f65a8]
/usr/sbin/mysqld(_Z16dispatch_command19enum_server_commandP3THDPcj+0xdf9)[0x6f7439]
/usr/sbin/mysqld(_Z24do_handle_one_connectionP3THD+0x15f)[0x6cb05f]
/usr/sbin/mysqld(handle_one_connection+0x47)[0x6cb147]
/usr/sbin/mysqld(pfs_spawn_thread+0x143)[0xb07243]
/lib/x86_64-linux-gnu/libpthread.so.0(+0x6b50)[0x7f86cabb2b50]
/lib/x86_64-linux-gnu/libc.so.6(clone+0x6d)[0x7f86c8e9d95d]

TokuDB error codes:
=================

TOKUDB_OUT_OF_LOCKS = -100000,
TOKUDB_SUCCEEDED_EARLY = -100001,
TOKUDB_FOUND_BUT_REJECTED = -100002,
TOKUDB_USER_CALLBACK_ERROR = -100003,
TOKUDB_DICTIONARY_TOO_OLD = -100004,
TOKUDB_DICTIONARY_TOO_NEW = -100005,
TOKUDB_DICTIONARY_NO_HEADER = -100006,
TOKUDB_CANCELED = -100007,
TOKUDB_NO_DATA = -100008,
TOKUDB_ACCEPT = -100009,
TOKUDB_MVCC_DICTIONARY_TOO_NEW = -100010,
TOKUDB_UPGRADE_FAILURE = -100011,
TOKUDB_TRY_AGAIN = -100012,
TOKUDB_NEEDS_REPAIR = -100013,
TOKUDB_CURSOR_CONTINUE = -100014,

TOKUDB_BAD_CHECKSUM = -100015,
        TOKUDB_HUGE_PAGES_ENABLED = -100016,
        TOKUDB_OUT_OF_RANGE = -100017,
        TOKUDB_INTERRUPTED = -100018,

[Sveta Smirnova (svetasmirnova)]

Thank you for the report.

Please try to collect more information about crash. You can consult this blog post as a guidance: https://www.percona.com/blog/2015/08/17/mysql-is-crashing-a-support-engineers-point-of-view/

[Launchpad Janitor (janitor)]

[Expired for Percona Server because there has been no activity for 60 days.]

[Shahriyar Rzayev (rzayev-sehriyar)]

Percona now uses JIRA for bug reports so this bug report is migrated to: https://jira.percona.com/browse/PS-3321