[MDEV-32660] [signal 11]Running simple query with many input values in MariaDB results in signal 11 Created: 2023-11-02  Updated: 2023-12-04  Resolved: 2023-12-04

Status: Closed
Project: MariaDB Server
Component/s: Data Manipulation - Subquery
Affects Version/s: 10.6.15
Fix Version/s: 10.4.32, 10.5.23, 10.6.16, 10.11.6, 11.0.4, 11.1.3, 11.2.2

Type: Bug Priority: Major
Reporter: 이종명 Assignee: Unassigned
Resolution: Duplicate Votes: 0
Labels: crash-recovery, signal
Environment:

[OS]
Ubuntu 22.04.2 LTS
CPU 48 Core (Intel(R) Xeon(R) Silver 4310 CPU @ 2.10GHz)
MEMORY 263752616 KB

[DB]
MariaDB 10.6.15
table_open_cache = 4096
join_buffer_size = 512K
read_buffer_size = 512K
read_rnd_buffer_size = 1M
sort_buffer_size = 8M
bulk_insert_buffer_size = 32M
thread_stack = 512K
max_heap_table_size = 256M
tmp_table_size = 256M
max_prepared_stmt_count = 256K
innodb_buffer_pool_size = 175G
innodb_data_file_path = ibdata1:10M:autoextend
innodb_log_buffer_size = 64M
innodb_log_file_size = 512M

Attachments: File mysql_error.log.crash.20231102    
Issue Links:
Duplicate
duplicates MDEV-32064 Crash when searching for the best spl... Closed

 Description   

Hello
A signal 11 crash occurred when I received about 1220 input values from the BI tool called Tableau and executed a simple query.
I don't know what the cause is.
Attach the mysql_error.log where the crash occurred.
Give them a hand.

[OS]
Ubuntu 22.04.2 LTS
CPU 48 Core (Intel(R) Xeon(R) Silver 4310 CPU @ 2.10GHz)
MEMORY 263752616 KB

[DB]
MariaDB 10.6.15
table_open_cache = 4096
join_buffer_size = 512K
read_buffer_size = 512K
read_rnd_buffer_size = 1M
sort_buffer_size = 8M
bulk_insert_buffer_size = 32M
thread_stack = 512K
max_heap_table_size = 256M
tmp_table_size = 256M
max_prepared_stmt_count = 256K
innodb_buffer_pool_size = 175G
innodb_data_file_path = ibdata1:10M:autoextend
innodb_log_buffer_size = 64M
innodb_log_file_size = 512M

 Comments   
Comment by Daniel Black [ 2023-11-02 ]

Thank you for the bug report.

The crash information you have provided looks like the bug MDEV-32064, and a fix for this is due out any day now in version 10.6.16.

When this is released, can you please with the same query confirm if it is fixed.

Comment by Daniel Black [ 2023-11-02 ]

In the mean time, setting in_predicate_conversion_threshold either globally or session to a higher value than the default 1000 (or whatever you are using) might avoid the crash successfully.

Comment by 이종명 [ 2023-11-02 ]

Thank you for the cause of the problem and for your help to resolve it temporarily

Generated at Thu Feb 08 10:33:01 UTC 2024 using Jira 8.20.16#820016-sha1:9d11dbea5f4be3d4cc21f03a88dd11d8c8687422.