Details
-
Bug
-
Status: Closed (View Workflow)
-
Major
-
Resolution: Duplicate
-
10.3.37, 10.6.11
-
VPS server on Centos 7 x64 with 8bg, cpanel/whm
Description
Since the server auto updated from 10.3.36 to 10.3.37, as well as after updating from 10.3.37 to 10.6.11, there is a memory leak.
on 10.3.37, top showed 1.3g use
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
|
14406 mysql 20 0 3179620 1.3g 6132 S 1.2 17.6 152:58.38 mysqld |
Which didn't match the 162MB shown by SHOW ENGINE INNODB STATUS;
----------------------
|
BUFFER POOL AND MEMORY
|
----------------------
|
Total large memory allocated 170590208 |
Dictionary memory allocated 340032 |
Buffer pool size 8192 |
Free buffers 1409 |
Database pages 6100 |
Old database pages 2255 |
Modified db pages 0 |
Percent of dirty pages(LRU & free pages): 0.000 |
Max dirty pages percent: 75.000 |
Pending reads 0 |
Pending writes: LRU 0, flush list 0, single page 0 |
Pages made young 6, not young 0 |
0.00 youngs/s, 0.00 non-youngs/s |
Pages read 5379, created 721, written 4228656 |
0.00 reads/s, 0.00 creates/s, 0.00 writes/s |
No buffer pool page gets since the last printout
|
Pages read ahead 0.00/s, evicted without access 0.00/s, Random read ahead 0.00/s |
LRU len: 6100, unzip_LRU len: 0 |
I/O sum[0]:cur[0], unzip sum[0]:cur[0] |
Moving to 10.6.11:
Some px-aux | grep mariadb log points
USER PID %CPU %MEM VSZ RSS TTY STAT START TIME COMMAND
|
mysql 29577 1.5 7.2 2715668 577652 ? Ssl Nov23 21:56 /usr/sbin/mariadbd |
mysql 29577 1.5 7.2 2699668 580328 ? Ssl Nov23 22:09 /usr/sbin/mariadbd |
mysql 29577 1.5 7.2 2708100 584600 ? Ssl Nov23 22:27 /usr/sbin/mariadbd |
mysql 29577 1.5 7.3 2708232 585152 ? Ssl Nov23 22:34 /usr/sbin/mariadbd |
mysql 29577 1.5 7.3 2708232 585416 ? Ssl Nov23 22:36 /usr/sbin/mariadbd |
mysql 29577 2.1 11.8 2720628 945832 ? Ssl Nov23 59:49 /usr/sbin/mariadbd |
mysql 29577 2.8 17.9 3318176 1437704 ? Ssl Nov23 136:12 /usr/sbin/mariadbd |
I have done a couple of consecutive memory dumps using and while I may not understand what to look for, it seems like the growing files show data relating to these tables and queries.
SELECT * FROM aercon0_aerweb.webaxess_sessions
|
WHERE id = ?
|
SELECT webfolder, webpassword
|
FROM aercon0_aerweb.webaxess
|
WHERE webusername = ?
|
UPDATE aercon0_aerweb.webaxess
|
SET aerware = ?,
|
last_comms = UNIX_TIMESTAMP()
|
WHERE webfolder = ?
|
Memory dump script from here https://serverfault.com/questions/173999/dump-a-linux-processs-memory-to-file
#!/bin/bash
|
 |
grep rw-p /proc/$1/maps \ |
| sed -n 's/^\([0-9a-f]*\)-\([0-9a-f]*\) .*$/\1 \2/p' \ |
| while read start stop; do \ |
gdb --batch --pid $1 -ex \ |
"dump memory $1-$start-$stop.dump 0x$start 0x$stop"; \ |
done
|
I'll try to migrate everything, but here is my original issue - https://dba.stackexchange.com/questions/320126/memory-leak-mariadb-10-3-37
Attachments
Issue Links
- duplicates
-
MDEV-29988 Major performance regression with 10.6.11
-
- Closed
-
- relates to
-
MDEV-30848 Memory leak in MariaDB 10.6
-
- Closed
-