possible memory leak

Description

The apparent leak seems to be proportional to innodb_buffer_pool_size

With innodb_buffer_pool_size = 128M it takes about 2-3 days for the kernel to run out of memory and kill/crash the mariadb node.

With innodb_buffer_pool_size = 1024M it takes about 2-3 hours for the kernel to run out of memory and kill/crash the mariadb node.

Please ask specifically for any/all details, logs, further tests etc. required to assist in investigation.

I'd also be happy to be directed to any related issues discussions etc. that googling has so far not revealed to me.

Environment

3 x gentoo linux vmware vm's with 8GB RAM allocted to each

Status

Assignee

Unassigned

Reporter

bmernz

Labels

External issue ID

None

External issue ID

None

Fix versions

Affects versions

10.0.15-galera

Priority

Critical