Details
-
Bug
-
Status: Open (View Workflow)
-
Major
-
Resolution: Unresolved
-
10.3(EOL)
-
None
Description
We skipped 10.2 because of MDEV-13983. So we decided to test 10.3.8 and it crashes in the first few seconds after brining it online.
Server version: 10.3.8-MariaDB-1:10.3.8+maria~trusty-log
|
key_buffer_size=16777216
|
read_buffer_size=131072
|
max_used_connections=20
|
max_threads=502
|
thread_count=38
|
It is possible that mysqld could use up to
|
key_buffer_size + (read_buffer_size + sort_buffer_size)*max_threads = 1119930 K bytes of memory
|
Hope that's ok; if not, decrease some variables in the equation.
|
|
|
Thread pointer: 0x7fb2700009a8
|
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 = 0x7fb3b1769e40 thread_stack 0x80000
|
/usr/sbin/mysqld(my_print_stacktrace+0x2e)[0x7fbfb2b37f2e]
|
/usr/sbin/mysqld(handle_fatal_signal+0x357)[0x7fbfb25d60a7]
|
/lib/x86_64-linux-gnu/libpthread.so.0(+0x10330)[0x7fbfb0784330]
|
/lib/x86_64-linux-gnu/libc.so.6(gsignal+0x37)[0x7fbfafdd7c37]
|
/lib/x86_64-linux-gnu/libc.so.6(abort+0x148)[0x7fbfafddb028]
|
/usr/sbin/mysqld(+0x4c8d18)[0x7fbfb2321d18]
|
/usr/sbin/mysqld(+0x9e69c3)[0x7fbfb283f9c3]
|
/usr/sbin/mysqld(+0xa15f01)[0x7fbfb286ef01]
|
/usr/sbin/mysqld(+0x945f4a)[0x7fbfb279ef4a]
|
/usr/sbin/mysqld(_Z8closefrmP5TABLE+0x131)[0x7fbfb24aab41]
|
/usr/sbin/mysqld(_Z8tc_purgeb+0x69)[0x7fbfb2558c39]
|
/usr/sbin/mysqld(_Z19close_cached_tablesP3THDP10TABLE_LISTbm+0x87)[0x7fbfb23a8b97]
|
/usr/sbin/mysqld(_Z20reload_acl_and_cacheP3THDyP10TABLE_LISTPi+0x2ff)[0x7fbfb24f7abf]
|
/usr/sbin/mysqld(_Z21mysql_execute_commandP3THD+0x1379)[0x7fbfb24016d9]
|
/usr/sbin/mysqld(_Z11mysql_parseP3THDPcjP12Parser_statebb+0x202)[0x7fbfb2409562]
|
/usr/sbin/mysqld(_Z16dispatch_command19enum_server_commandP3THDPcjbb+0x1c38)[0x7fbfb240c068]
|
/usr/sbin/mysqld(_Z10do_commandP3THD+0x13e)[0x7fbfb240cefe]
|
/usr/sbin/mysqld(_Z24do_handle_one_connectionP7CONNECT+0x1aa)[0x7fbfb24dd62a]
|
/usr/sbin/mysqld(handle_one_connection+0x3d)[0x7fbfb24dd74d]
|
/usr/sbin/mysqld(+0x929ead)[0x7fbfb2782ead]
|
/lib/x86_64-linux-gnu/libpthread.so.0(+0x8184)[0x7fbfb077c184]
|
/lib/x86_64-linux-gnu/libc.so.6(clone+0x6d)[0x7fbfafe9f03d]
|
|
|
Trying to get some variables.
|
Some pointers may be invalid and cause the dump to abort.
|
Query (0x7fb2700116e0): FLUSH TABLES WITH READ LOCK
|
|
|
Connection ID (thread ID): 7916
|
Status: NOT_KILLED
|
|
|
Optimizer switch: index_merge=on,index_merge_union=on,index_merge_sort_union=on,index_merge_intersection=on,index_merge_sort_intersection=off,engine_condition_pushdown=off,index_condition_pushdown=on,derived_merge=on,derived_with_keys=on,firstmatch=on,loosescan=on,mat
|
erialization=on,in_to_exists=on,semijoin=on,partial_match_rowid_merge=on,partial_match_table_scan=on,subquery_cache=on,mrr=off,mrr_cost_based=off,mrr_sort_keys=off,outer_join_with_cache=on,semijoin_with_cache=on,join_cache_incremental=on,join_cache_hashed=on,join_cach
|
e_bka=on,optimize_join_buffer_size=off,table_elimination=on,extended_keys=on,exists_to_in=on,orderby_uses_equalities=on,condition_pushdown_for_derived=on,split_materialized=on
|
our slave server also crashes with 10.3.8 installed
180808 21:51:29 [ERROR] mysqld got signal 6 ;
|
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.
|
|
|
To report this bug, see https://mariadb.com/kb/en/reporting-bugs
|
|
|
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.
|
|
|
Server version: 10.3.8-MariaDB-1:10.3.8+maria~trusty-log
|
key_buffer_size=134217728
|
read_buffer_size=131072
|
max_used_connections=3
|
max_threads=502
|
thread_count=0
|
It is possible that mysqld could use up to
|
key_buffer_size + (read_buffer_size + sort_buffer_size)*max_threads = 1234618 K bytes of memory
|
Hope that's ok; if not, decrease some variables in the equation.
|
|
|
Thread pointer: 0x0
|
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 = 0x0 thread_stack 0x80000
|
/usr/sbin/mysqld(my_print_stacktrace+0x2e)[0x556816bdef2e]
|
/usr/sbin/mysqld(handle_fatal_signal+0x357)[0x55681667d0a7]
|
/lib/x86_64-linux-gnu/libpthread.so.0(+0x10330)[0x7f434b3a7330]
|
/lib/x86_64-linux-gnu/libc.so.6(gsignal+0x37)[0x7f434a9fac37]
|
linux/raise.c:56(__GI_raise)[0x7f434a9fe028]
|
/usr/sbin/mysqld(+0x4c8d18)[0x5568163c8d18]
|
/usr/sbin/mysqld(+0x9e69c3)[0x5568168e69c3]
|
/usr/sbin/mysqld(+0xa15f01)[0x556816915f01]
|
/usr/sbin/mysqld(+0x945f4a)[0x556816845f4a]
|
/usr/sbin/mysqld(_Z8closefrmP5TABLE+0x131)[0x556816551b41]
|
/usr/sbin/mysqld(_Z8tc_purgeb+0x69)[0x5568165ffc39]
|
/usr/sbin/mysqld(_Z19close_cached_tablesP3THDP10TABLE_LISTbm+0x87)[0x55681644fb97]
|
/usr/sbin/mysqld(+0x4ecd8f)[0x5568163ecd8f]
|
/usr/sbin/mysqld(_Z10unireg_endv+0x3b)[0x5568163ed07b]
|
/usr/sbin/mysqld(+0x4f09a9)[0x5568163f09a9]
|
/usr/sbin/mysqld(kill_server_thread+0xe)[0x5568163f09ee]
|
/lib/x86_64-linux-gnu/libpthread.so.0(+0x8184)[0x7f434b39f184]
|
/lib/x86_64-linux-gnu/libc.so.6(clone+0x6d)[0x7f434aac203d]
|
Attachments
Issue Links
- duplicates
-
-
- Closed
-
- relates to
-
-
- Open
-
-
-
- Closed
-
-
-
- Closed
-
Activity
I understand. Low traffic tests on our testing setup work fine. This is about the heart of the problem, as we would love to debug this with you but at the moment (as last year) this only occurs on production level traffic. We simply cannot afford to have our business offline for this purpose - unfortunately.
On a too quick look, it seemed that this failure could share a root cause with the following:
MDEV-12023 Assertion failure sym_node->table != NULL on startup
But, it is claimed that the failure only occurs when the system is heavily loaded. The MDEV-12023 scenario only applies on the very first startup when upgrading the server.
There was no core dump on all threads, so I cannot be sure whether this could duplicate MDEV-15488, which is reported for 10.2, but I think that it is plausible.
I think that my previous comment was in response to CrewOne’s message that included an InnoDB crash (which could have been fixed in MariaDB Server 10.3.11 by MDEV-12023), and a stack trace of a possibly unrelated thread:
2018-08-09 10:02:08 0x7fdbef2f9700 InnoDB: Assertion failure in file /home/buildbot/buildbot/build/mariadb-10.3.8/storage/innobase/que/que0que.cc line 563
|
The stack trace in the Description (crash during FLUSH TABLES WITH READ LOCK) could be unrelated to MDEV-12023. There have been some bugs fixed in the .frm file parsing. Some memory corruption due to such bugs might be the cause of the crash.
I'm very sorry about the trouble you experienced, and I understand the frustration.
Most users seem to migrate well, however there is something, either in the workflow, or in the data layout, or in their combination, that causes this problem for some setups or configurations. We are trying to hunt it down.