[MDEV-9509] Server crash with 1.4.0 audit plugin Created: 2016-02-02  Updated: 2016-12-20  Resolved: 2016-12-20

Status: Closed
Project: MariaDB Server
Component/s: Plugin - Audit
Affects Version/s: 5.5.48
Fix Version/s: 5.5.54

Type: Bug Priority: Critical
Reporter: Kyle Joiner (Inactive) Assignee: Alexey Botchkov
Resolution: Duplicate Votes: 0
Labels: None
Environment:

Linux
MySQL 5.7


Sprint: 10.1.12, 10.2.0-6, 10.2.1-3, 10.2.1-4, 5.5.50, 10.2.2-1, 10.2.2-2, 10.2.2-3, 5.5.54

 Description   

22:08:01 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.
Attempting to collect some information that could help diagnose the problem.
As this is a crash and something is definitely wrong, the information
collection process might fail.
 
key_buffer_size=10485760
read_buffer_size=131072
max_used_connections=67
max_threads=480
thread_count=66
connection_count=66
It is possible that mysqld could use up to
key_buffer_size + (read_buffer_size + sort_buffer_size)*max_threads = 200961 K bytes of memory
Hope that's ok; if not, decrease some variables in the equation.
 
Thread pointer: 0x7f66f4012160
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 = 7f67f006ae28 thread_stack 0x40000
xxxx/bin/mysqld(my_print_stacktrace+0x35)[0xebb645]
xxxx/bin/mysqld(handle_fatal_signal+0x4a4)[0x76e6e4]
/lib64/libpthread.so.0(+0xf790)[0x7f683163e790]
xxxx/lib/plugin/server_audit-1.4.0-linux-64.so(+0x4a28)[0x7f67ffd9da28]
xxxx/lib/plugin/server_audit-1.4.0-linux-64.so(logger_write+0x68)[0x7f67ffd9db48]
xxxx/lib/plugin/server_audit-1.4.0-linux-64.so(+0x4be7)[0x7f67ffd9dbe7]
xxxx/lib/plugin/server_audit-1.4.0-linux-64.so(auditing+0x829)[0x7f67ffd9ed79]
xxxx/lib/plugin/server_audit-1.4.0-linux-64.so(+0x709d)[0x7f67ffda009d]
xxxx/bin/mysqld[0x76f50f]
xxxx/bin/mysqld(_Z18mysql_audit_notifyP3THD33mysql_event_connection_subclass_tPKci+0x191)[0x76fe11]
xxxx/5.7.9/bin/mysqld(handle_connection+0x79)[0xd75289]
xxxx/5.7.9/bin/mysqld(pfs_spawn_thread+0x174)[0xf18334]
/lib64/libpthread.so.0(+0x7a51)[0x7f6831636a51]
/lib64/libc.so.6(clone+0x6d)[0x7f683039f93d]

For the test I'm just using Sysbench with parallel_prepare.lua, 40 tables,
1 million rows per table, 10 minute tests. Thread levels run
8/16/32/64/128, and I run 10 tests for each. The crash seems to happen
"randomly", so can't narrow it down more than that. I've done a lot of
tests, and seen the crash just a few times so far but it's definitely
hitting something.



 Comments   
Comment by Elena Stepanova [ 2016-02-03 ]

kjoiner,

Please paste or attach the config file.
Are you only getting it with MySQL 5.7? Did you try MariaDB with the same load on the same machine?

Comment by Axel Schwenke [ 2016-02-27 ]

Kyle,

does that happen in the prepare phase (populating tables) or in the run phase (the actual benchmark)? Please give the full sysbench command line.

Comment by Scott Nemes [ 2016-03-29 ]

Hi Axel;

Kyle submitted this report for me, so I'll go ahead and answer as I did not realize this was still pending.

This happens while running the tests, but it is intermittent. I run an array of tests, with thread levels ranging from 8-64, 10 tests a piece, each at 10 minutes in length. The server then crashes randomly with the mentioned error.

Here is the base command:

sysbench \
--test=oltp.lua \
--oltp-tables-count=40 \
--mysql-host="${DB_HOST}" \
--mysql-port=$DB_PORT \
--mysql-user="${DB_USER}" \
--mysql-password="${DB_PASS}" \
--mysql-db="${DB_NAME}" \
--mysql-table-engine=innodb \
--num-threads=8 \
--max-requests=0 \
--max-time=600 \
--oltp-table-size=1000000 \
run

Comment by Scott Nemes [ 2016-05-02 ]

Any update on this? It still says feedback is needed, but you should have what you need at this point.

Thank you!

Comment by Alexey Botchkov [ 2016-06-06 ]

Waiting for Kyle.

Comment by Scott Nemes [ 2016-06-08 ]

Why is this stalled?

Comment by Alexey Botchkov [ 2016-12-20 ]

It duplicates this https://jira.mariadb.org/browse/MDEV-11510

Generated at Thu Feb 08 07:35:12 UTC 2024 using Jira 8.20.16#820016-sha1:9d11dbea5f4be3d4cc21f03a88dd11d8c8687422.