Ad Widget

Collapse

Zabbix alerter process, 100%

Collapse
X
Collapse
 
  • Page of 1
  • Time
  • Show
Clear All
new posts
Previous template Next
  • 1inux
    Junior Member
    • Dec 2014
    • 2
    #1

    Zabbix alerter process, 100%

    Hello, I'm running zabbix 2.2.6 on gentoo
    Sometimes zabbix alerter process gets at 100% and I can't receive alarm messages.

    I can solve this problem temporary with restart of zabbix-server.

    Tags: None
    • eduardobertobr
      Member
      • Jul 2014
      • 30
      #2
      Hi! First sorry my english!

      Second make a backup from zabbix_server.conf. After seek and edit valors:

      StartPollers=80
      StartPingers=10
      StartPollersUnreachable=80
      StartIPMIPollers=10
      StartTrappers=20
      StartDBSyncers=8

      Restart zabbix_server.conf and make sure ok.

        Comment

        • tchjts1
          Senior Member
          • May 2008
          • 1605
          #3
          Originally posted by eduardobertobr
          Hi! First sorry my english!

          Second make a backup from zabbix_server.conf. After seek and edit valors:

          StartPollers=80
          StartPingers=10
          StartPollersUnreachable=80
          StartIPMIPollers=10
          StartTrappers=20
          StartDBSyncers=8

          Restart zabbix_server.conf and make sure ok.
          While I think this idea is on the right track, the values you show above may not be the correct values for the OP's situation.

          StartDBSyncers should be left at the default value of 4 unless you are monitoring thousands of devices and have the actual need to increment that.

          I think having a matching number of StartPollersUnreachable to StartPollers is excessive.

          As an example, and for comparison purposes only, I am currently monitoring 1500 hosts with 95,000 items and 23,000 triggers. I have an NVPS of 1346. My NVPS is higher than I would like, but it is what it is.

          And these are my pertinent zabbix_server.conf settings to handle the above:
          StartPollers=375
          StartPollersUnreachable=40
          StartTrappers=80
          StartPingers=50 (I do a lot of ICMP pings)
          CacheSize=1G
          StartDBSyncers=4
          HistoryCacheSize=512M
          TrendCacheSize=512M
          HistoryTextCacheSize=512M
          Timeout=15

          But again, that is just for comparison. Everyone's own environment will dictate what settings will be best for them.That is best determined by looking at ALL of your Zabbix internal process graphs and cache usage graphs, making adjustments in reasonable increments and re-evaluating the graphs to see if you are getting the desired results.

            Comment

            • 1inux
              Junior Member
              • Dec 2014
              • 2
              #4
              holy Zeus.
              Changing media type for jabber from native zabbix to user script (sendxmpp)

              thnx 2 all

                Comment

                Previous template Next
                Working...