Details
-
Bug
-
Status: Open (View Workflow)
-
Major
-
Resolution: Unresolved
-
11.4, 11.7
-
None
Description
CREATE TABLE t (a INT) ENGINE=MyISAM; |
INSERT INTO t VALUES (1); |
LOCK TABLE t READ; |
--connect (con1,localhost,root,,)
|
SET max_statement_time= 0.001; |
--error ER_STATEMENT_TIMEOUT
|
ALTER TABLE t FORCE; |
--error ER_STATEMENT_TIMEOUT
|
ALTER TABLE IF EXISTS t FORCE; |
|
# Cleanup
|
--disconnect con1
|
--connection default
|
UNLOCK TABLES;
|
DROP TABLE t; |
11.4 866a8ea6736d2edc0f6be552d1cdd6810c10d5ab |
mariadbd: /data/bld/11.4-asan/sql/sql_error.h:1087: uint Diagnostics_area::sql_errno() const: Assertion `m_status == DA_ERROR' failed.
|
241210 0:21:57 [ERROR] mysqld got signal 6 ;
|
|
#9 0x00007fd846e53e32 in __GI___assert_fail (assertion=0x55f50d4b2660 "m_status == DA_ERROR", file=0x55f50d4b26a0 "/data/bld/11.4-asan/sql/sql_error.h", line=1087, function=0x55f50d4b2700 "uint Diagnostics_area::sql_errno() const") at ./assert/assert.c:101
|
#10 0x000055f50afe9956 in Diagnostics_area::sql_errno (this=0x62c0000f6368) at /data/bld/11.4-asan/sql/sql_error.h:1087
|
#11 0x000055f50b5f9d95 in mysql_alter_table (thd=0x62c0000f0218, new_db=0x62c0000f4f98, new_name=0x62c0000f53f8, create_info=0x7fd83af59ce0, table_list=0x629000267358, recreate_info=0x7fd83af59a70, alter_info=0x7fd83af59b30, order_num=0, order=0x0, ignore=false, if_exists=true) at /data/bld/11.4-asan/sql/sql_table.cc:10513
|
#12 0x000055f50b7cf433 in Sql_cmd_alter_table::execute (this=0x629000267a70, thd=0x62c0000f0218) at /data/bld/11.4-asan/sql/sql_alter.cc:701
|
#13 0x000055f50b2ef9f0 in mysql_execute_command (thd=0x62c0000f0218, is_called_from_prepared_stmt=false) at /data/bld/11.4-asan/sql/sql_parse.cc:5864
|
#14 0x000055f50b2fcb2e in mysql_parse (thd=0x62c0000f0218, rawbuf=0x629000267238 "ALTER TABLE IF EXISTS t FORCE", length=29, parser_state=0x7fd83af5ba90) at /data/bld/11.4-asan/sql/sql_parse.cc:7893
|
#15 0x000055f50b2d3cbe in dispatch_command (command=COM_QUERY, thd=0x62c0000f0218, packet=0x62900025d219 "ALTER TABLE IF EXISTS t FORCE", packet_length=29, blocking=true) at /data/bld/11.4-asan/sql/sql_parse.cc:1905
|
#16 0x000055f50b2d09f6 in do_command (thd=0x62c0000f0218, blocking=true) at /data/bld/11.4-asan/sql/sql_parse.cc:1418
|
#17 0x000055f50b7b075b in do_handle_one_connection (connect=0x6080000036b8, put_in_cache=true) at /data/bld/11.4-asan/sql/sql_connect.cc:1429
|
#18 0x000055f50b7b02ba in handle_one_connection (arg=0x6080000036b8) at /data/bld/11.4-asan/sql/sql_connect.cc:1341
|
#19 0x000055f50c4299ee in pfs_spawn_thread (arg=0x617000005f18) at /data/bld/11.4-asan/storage/perfschema/pfs.cc:2201
|
#20 0x00007fd846ea8044 in start_thread (arg=<optimized out>) at ./nptl/pthread_create.c:442
|
#21 0x00007fd846f2861c in clone3 () at ../sysdeps/unix/sysv/linux/x86_64/clone3.S:81
|
No obvious problem on a non-debug build.
The failure is also happening with other variations of killing the query, e.g. upon KILL_SERVER, it is just more reliable with the statement timeout.
It started happening after this commit in 11.4.1:
commit 2f5174e556cd247133aa14d7b37372ae49fe83c5
|
Author: Vladislav Vaintroub
|
Date: Thu Feb 1 02:14:23 2024 +0100
|
|
MDEV-33075 Resolve server shutdown issues on macOS, Solaris, and FreeBSD
|
Note that this patch was recently backported from 11.4 to 10.5-10.11. The failure isn't happening on 10.5-10.11 with the backport, but I cannot say whether it's because the backport also fixes the issue (after all, it came in 4 commits, not just one), or because it just doesn't affect 10.5.
Attachments
Issue Links
- is caused by
-
MDEV-33075 Sending TERM to mariadb no longer gracefully terminates the process
- Closed