[MXS-73] MaxScale uses nearly 100% CPU Created: 2015-03-27  Updated: 2015-06-17  Resolved: 2015-04-27

Status: Closed
Project: MariaDB MaxScale
Component/s: None
Affects Version/s: 1.1.0
Fix Version/s: 1.1.1

Type: Bug Priority: Major
Reporter: Kolbe Kegel (Inactive) Assignee: Timofey Turenko
Resolution: Cannot Reproduce Votes: 0
Labels: None
Environment:

CentOS 6


 Description   

I'm not too sure what's going on here, but maxscale seems to be using nearly 100% CPU even when totally idle.

[root@max1 ~]# service maxscale restart
 
Stopping MaxScale:                                         [  OK  ]
 
Starting MaxScale: maxscale (pid 12519) is running...      [  OK  ]
 
[root@max1 ~]# ps up $(pgrep maxscale)
 
USER       PID %CPU %MEM    VSZ   RSS TTY      STAT START   TIME COMMAND
 
root     11366 98.8  4.6 462088 47028 ?        Rsl  10:43 148:37 /usr/local/mariadb-maxscale/bin/maxscale
 
root     12519  1.0  4.5 861844 46388 ?        Ssl  13:13   0:00 /usr/local/mariadb-maxscale/bin/maxscale
 
[root@max1 ~]# ps up $(pgrep maxscale)
 
USER       PID %CPU %MEM    VSZ   RSS TTY      STAT START   TIME COMMAND
 
root     11366 98.8  4.6 462088 47028 ?        Rsl  10:43 148:40 /usr/local/mariadb-maxscale/bin/maxscale
 
root     12519  1.0  4.5 861844 46388 ?        Ssl  13:13   0:00 /usr/local/mariadb-maxscale/bin/maxscale
 
[root@max1 ~]# ps up $(pgrep maxscale)
 
USER       PID %CPU %MEM    VSZ   RSS TTY      STAT START   TIME COMMAND
 
root     11366 98.8  4.6 462088 47028 ?        Rsl  10:43 148:41 /usr/local/mariadb-maxscale/bin/maxscale
 
root     12519  1.0  4.5 861844 46388 ?        Ssl  13:13   0:00 /usr/local/mariadb-maxscale/bin/maxscale



 Comments   
Comment by Kolbe Kegel (Inactive) [ 2015-03-27 ]

This happens even with the very most basic possible config that does nothing but start the CLI service & listener.

[root@max1 ~]# service maxscale stop
 
Stopping MaxScale:                                         [  OK  ]
 
[root@max1 ~]# cat /usr/local/mariadb-maxscale/etc/MaxScale.cnf
 
[maxscale]
 
[CLI]
 
type=service
 
router=cli
 
[CLI Listener]
 
type=listener
 
service=CLI
 
protocol=maxscaled
 
port=6603
 
[root@max1 ~]# service maxscale start
 
Starting MaxScale: maxscale (pid 12843) is running...      [  OK  ]
 
[root@max1 ~]# ps up $(pgrep maxscale)
 
USER       PID %CPU %MEM    VSZ   RSS TTY      STAT START   TIME COMMAND
 
root     11366 98.8  4.6 462088 47028 ?        Rsl  10:43 163:25 /usr/local/mariadb-maxscale/bin/maxscale
 
root     12843  0.4  4.2 548100 43720 ?        Ssl  13:28   0:00 /usr/local/mariadb-maxscale/bin/maxscale
 
[root@max1 ~]# /usr/local/mariadb-maxscale/bin/maxadmin -pmariadb <<EoF
 
> show servers
 
> show monitors
 
> show filters
 
> EoF
 
[root@max1 ~]#

Comment by Kolbe Kegel (Inactive) [ 2015-03-27 ]

I've got no idea what caused this. There was some runaway maxscale process that was just sitting there burning CPU, and I couldn't kill it. Definitely not repeatable, nor was I even looking at the right thing when I restarted the service.

Comment by Kolbe Kegel (Inactive) [ 2015-03-27 ]

[root@max1 ~]# strace -f -p 11366
Process 11366 attached with 3 threads - interrupt to quit
[pid 11371] restart_syscall(<... resuming interrupted call ...> <unfinished ...>
[pid 11369] futex(0x385e1d4, FUTEX_WAIT_PRIVATE, 11, NULL

Comment by Timofey Turenko [ 2015-04-27 ]

can't reproduce.

Comment by Timofey Turenko [ 2015-05-12 ]

can't be verified

Generated at Thu Feb 08 03:56:35 UTC 2024 using Jira 8.20.16#820016-sha1:9d11dbea5f4be3d4cc21f03a88dd11d8c8687422.