Ad Widget

Collapse

Zabbix Pollers and service issue

Collapse
X
 
  • Time
  • Show
Clear All
new posts
  • Yousif282
    Junior Member
    • Aug 2020
    • 9

    #1

    Zabbix Pollers and service issue

    Hello
    I Have a problem with my zabbix environment.
    My setup consist of two physical zabbix server running as active passive using pacemaker
    the servers specs are

    40 Core Intel(R) Xeon(R) CPU E5-2680 v3 @ 2.50GHz
    126 GB RAM

    My database is a master-master mysql.
    it also consist of two physical server, their specs are

    40 Core Intel(R) Xeon(R) CPU E5-2680 v2 @ 2.80GHz
    256 GB of RAM
    18 TB of SSD storage

    I have about 13,000 hosts and over 780,000 items
    my NVPS is 5300

    My problem is that my data pollers and unreachable pollers are always High
    My unreachable pollers are stuck at 100% and data pollers are above 95 %

    the other problem is that when i try ot fail over to my second zabbix server with pacemaker
    the zabbix-server service on the fist one does not shut down properly and it get stuck and the only way to fail over to the second server is to shut down the server so the pacemaker can start the service on the other server.

    Here's the My config file for zabbix server and mysql
    also you can view the attached images for the performance graphs for a 1 month period

    My zabbix version is 4.4.4 and i have one proxy server for testing.
    Attached Files
  • isaqueprofeta
    Senior Member
    Zabbix Certified SpecialistZabbix Certified Professional
    • Aug 2020
    • 154

    #2
    I'd break up this two zabbix servers physical machines into VMs and have myself a bunch of proxies to separate the collect and preprocessing workload.

    If anything more it helps to find issues in data collection with some segmentation.

    Comment

    • kloczek
      Senior Member
      • Jun 2006
      • 1771

      #3
      I'd break up this two zabbix servers physical machines into VMs and have myself a bunch of proxies to separate the collect and preprocessing workload.
      Such change does not make any sense. All wjhat needs to be done is:
      1) stop monitoring anything relse than zabbix server themeselv over zabbix server and add proxies
      2) swiitch from zabbix agent passive monitoring to active one
      http://uk.linkedin.com/pub/tomasz-k%...zko/6/940/430/
      https://kloczek.wordpress.com/
      zapish - Zabbix API SHell binding https://github.com/kloczek/zapish
      My zabbix templates https://github.com/kloczek/zabbix-templates

      Comment

      • isaqueprofeta
        Senior Member
        Zabbix Certified SpecialistZabbix Certified Professional
        • Aug 2020
        • 154

        #4
        Originally posted by kloczek

        Such change does not make any sense. All wjhat needs to be done is:
        1) stop monitoring anything relse than zabbix server themeselv over zabbix server and add proxies
        2) swiitch from zabbix agent passive monitoring to active one
        What I said, mean the same as what you write about "stop monitoring anything relse than zabbix server themeselv over zabbix server and add proxies", but I took into account the size of the servers he has, and his need to keep a HA environment. I'm just trying to make myself clear, you don't need to reply to this.

        Comment

        • Yousif282
          Junior Member
          • Aug 2020
          • 9

          #5
          Thanks for the reply,
          I will try to move on to zabbix proxies
          and change the zabbix-agent to passive checks

          Comment

          Working...