Archived community.zenoss.org | full text search
Skip navigation
1524 Views 1 Reply Latest reply: Dec 6, 2013 10:58 AM by Photon RSS
jaime franco Newbie 2 posts since
May 28, 2013
Currently Being Moderated

May 28, 2013 11:26 AM

How I can reduce the load average of my zenoss server?

hi

 

i am looking for a solution of this problem in a forums but i cannot find the answer or the cause of this problem, the zenoss server in this moment show a load average: 9.17, 9.34, 8.87, for this reason im not enable to see the zenoos web gui, this server is a centos 6.4 x64 with zenoss version 4.2.3 and

MySQL 5.5.28 (5.5.28)

 

show the top command of the server

 

top - 10:21:26 up 51 min,  2 users,  load average: 9.08, 9.27, 8.92

Tasks: 203 total,   7 running, 196 sleeping,   0 stopped,   0 zombie

Cpu(s): 84.5%us,  8.8%sy,  6.5%ni,  0.0%id,  0.0%wa,  0.0%hi,  0.2%si,  0.0%st

Mem:   8124368k total,  7779448k used,   344920k free,   115352k buffers

Swap:  8257528k total,        0k used,  8257528k free,   544092k cached

 

  PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND                                                                                                  

2514 zenoss    20   0 1162m 637m 8764 S 87.7  8.0  41:31.53 runzope                                                                                                  

21029 zenoss    20   0  496m  99m 9128 R 57.6  1.3   0:07.06 python                                                                                                   

21040 zenoss    20   0  496m 100m 9128 R 50.7  1.3   0:05.99 python                                                                                                   

2565 zenoss    20   0  560m 109m 4504 R 47.7  1.4  29:00.46 python                                                                                                   

21064 zenoss    20   0  454m  70m 8156 R 44.0  0.9   0:03.47 python                                                                                                   

20663 zenoss    25   5  569m 114m 9176 R 24.5  1.4   0:22.10 python                                                                                                   

21114 zenoss    20   0  215m  11m 3508 R  8.9  0.1   0:00.27 python                                                                                                   

2643 zenoss    20   0  775m 179m 3532 S  7.3  2.3   1:27.17 python                                                                                                   

1627 mysql     20   0 2805m 383m 5624 S  5.6  4.8   2:56.71 mysqld                                                                                                   

2639 zenoss    20   0  785m 178m 3540 S  3.3  2.2   1:27.43 python                                                                                                   

2641 zenoss    20   0  775m 179m 3540 S  3.3  2.3   1:27.32 python                                                                

  • Photon Rank: White Belt 44 posts since
    Apr 15, 2011

    Hello,

     

    You can try using a Distributed Collectors setup. Having the collector(s) do all the polling/writing performance data should reduce load on the master. This setup is normally used for 1000+ devices or when you want to "separate" networks for security purposes.

      There's also the tuning solution. There are plenty of posts about that.

     

    Good luck,

More Like This

  • Retrieving data ...

Legend

  • Correct Answers - 4 points
  • Helpful Answers - 2 points