[MDEV-15312] Crash during OPTIMIZE TABLE Created: 2018-02-14  Updated: 2018-02-14  Resolved: 2018-02-14

Status: Closed
Project: MariaDB Server
Component/s: Data Definition - Alter Table, Storage Engine - InnoDB
Affects Version/s: 10.0.23, 10.1.10, 10.2.0, 10.1.19, 10.3.0
Fix Version/s: N/A

Type: Bug Priority: Minor
Reporter: Andrey Assignee: Unassigned
Resolution: Duplicate Votes: 0
Labels: None

Issue Links:
Duplicate
duplicates MDEV-12827 Assertion failure when reporting dupl... Closed

 Description   

Hello, I found task
https://jira.mariadb.org/browse/MDEV-12827
and hope you can help me.

I have very similar crash with signal 11 - here is error log:

Crash log

00:10:26 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=536870912
read_buffer_size=262144
max_used_connections=699
max_threads=100020
thread_count=411
connection_count=408
It is possible that mysqld could use up to
key_buffer_size + (read_buffer_size + sort_buffer_size)*max_threads = 435 821 357 K bytes of memory
Hope that's ok; if not, decrease some variables in the equation.

Thread pointer: 0x7f08753e3000
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 = 7f0abcf07be0 thread_stack 0x40000
/usr/sbin/mysqld(my_print_stacktrace+0x3b)[0x8e81db]
/usr/sbin/mysqld(handle_fatal_signal+0x471)[0x65f461]
/lib64/libpthread.so.0(+0xf370)[0x7f2e59417370]
/usr/sbin/mysqld(_Z10key_unpackP6StringP5TABLEP6st_key+0x38)[0x645e98]
/usr/sbin/mysqld(_Z18print_keydup_errorP5TABLEP6st_keyPKci+0x82)[0x5a0242]
/usr/sbin/mysqld[0x9adfcc]
/usr/sbin/mysqld(_Z17mysql_alter_tableP3THDPcS1_P24st_ha_create_informationP10TABLE_LISTP10Alter_infojP8st_orderb+0x1e66)[0x73b266]
/usr/sbin/mysqld(_Z20mysql_recreate_tableP3THDP10TABLE_LISTb+0x17c)[0x73ca8c]
/usr/sbin/mysqld[0x81f41c]
/usr/sbin/mysqld(_ZN22Sql_cmd_optimize_table7executeEP3THD+0xde)[0x81fc9e]
/usr/sbin/mysqld(_Z21mysql_execute_commandP3THD+0xfda)[0x6e3b4a]
/usr/sbin/mysqld(_Z11mysql_parseP3THDPcjP12Parser_state+0x5b8)[0x6e9e78]
/usr/sbin/mysqld(_Z16dispatch_command19enum_server_commandP3THDPcj+0x1002)[0x6eb622]
/usr/sbin/mysqld(_Z26threadpool_process_requestP3THD+0x108)[0x77a8f8]
/usr/sbin/mysqld[0x77b6ee]
/usr/sbin/mysqld(pfs_spawn_thread+0x146)[0x983b46]
/lib64/libpthread.so.0(+0x7dc5)[0x7f2e5940fdc5]
/lib64/libc.so.6(clone+0x6d)[0x7f2e575e476d]

Trying to get some variables.
Some pointers may be invalid and cause the dump to abort.
Query (7f0b2662e010): is an invalid pointer
Connection ID (thread ID): 8249845
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.

Test case is something like that:
DELETE FROM t5
OPTIMIZE TABLE t0
OPTIMIZE TABLE t1
OPTIMIZE TABLE t2
OPTIMIZE TABLE t3
OPTIMIZE TABLE t4
OPTIMIZE TABLE t5

Crash - always on OPTIMIZE TABLE t5
DB version is 5.6.38-83.0-log Percona Server


Can you provide parameters to run test case from MDEV-12827 ?
How many threads did you need back there?

You found, that crash do not repeat on some MySQL and MariaDB versions - do you have full list?
Maybe you have list of versions, where DB process was crashed?



 Comments   
Comment by Elena Stepanova [ 2018-02-14 ]

Options needed for the non-deterministic test case are mentioned in or before the test case.
For the deterministic test case suggested by Marko you need a debug build and a corresponding debug sync point in the code.
We don't know if any of the test cases will work as is for your version of Percona server, we didn't check and have no plans to.
All information that we have handy about affected/not affected versions is also in the bug report.

Generated at Thu Feb 08 08:20:19 UTC 2024 using Jira 8.20.16#820016-sha1:9d11dbea5f4be3d4cc21f03a88dd11d8c8687422.