Details
-
Bug
-
Status: Closed (View Workflow)
-
Minor
-
Resolution: Not a Bug
-
10.11.5
-
None
Description
Hi Team,
We have a MariaDB Replication enabled in 2 Servers and we have observed that MariaDB Server is not releasing the cache/buffer memory after using it or even it also not getting reclaimed by the Server.
We have check the VM Stats:
total used free shared buff/cache available
Mem: 515464 140033 43544 2 331886 389750
Swap: 32767 10 32757
As in above command you can see used is only 140 GB out of 515 but still free is 44 GB because rest is occupied in cache and that is with MariaDB as mentioned below:
top - 15:49:53 up 356 days, 23:06, 2 users, load average: 0.96, 1.29, 1.52
Tasks: 376 total, 1 running, 375 sleeping, 0 stopped, 0 zombie
%Cpu(s): 2.8 us, 1.1 sy, 0.0 ni, 94.7 id, 0.5 wa, 0.5 hi, 0.3 si, 0.0 st
MiB Mem : 515464.4 total, 43427.3 free, 140039.3 used, 331997.8 buff/cache
MiB Swap: 32768.0 total, 32757.9 free, 10.0 used. 389750.3 avail Mem
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
679579 mysql 20 0 332.1g 134.2g 25656 S 117.6 26.7 5543:11 mariadbd
We have tried to do stop slave & start slave but no luck.
Please check & suggest some cache cleanup mechanism or fix for this