Details
-
Bug
-
Status: Closed (View Workflow)
-
Critical
-
Resolution: Fixed
-
10.11, 11.4, 11.8, 12.3, 11.8.9
-
Can result in hang or crash
-
InnoDB could crash if the parameter innodb_buffer_pool_size_auto_min was set and it failed to shrink the buffer pool in response to a Linux memory pressure event.
Description
We have a number of mariadb systems on which an overnight database-check script runs.
On one of them, which has a smaller amount of RAM but compensating larger swap size the mariadb process crashes overnight
Crash log attached
Attachments
Issue Links
- includes
-
MDEV-41330 missing a >> 20 in garbage_collect sql_print_information
-
- Closed
-
- is caused by
-
MDEV-29445 reorganise innodb buffer pool (and remove buffer pool chunks)
-
- Closed
-
-
MDEV-34863 RAM Usage Changed Significantly Between 10.11 Releases
-
- Closed
-