Ad Widget

Collapse

Problem after upgrading to PG11+Timescladb and zabbix 4.4.1

Collapse
X
 
  • Time
  • Show
Clear All
new posts
  • sgaunet
    Junior Member
    • Nov 2019
    • 2

    #1

    Problem after upgrading to PG11+Timescladb and zabbix 4.4.1

    Hello,

    I have migrated our production zabbix from zabbix 4.0 + postgresql 10 to :
    * postgresql 11 with timescaledb
    * zabbix 4.4.1

    I'm using an ubuntu server :

    root@zbx:~# lsb_release -a
    No LSB modules are available.
    Distributor ID: Ubuntu
    Description: Ubuntu 18.04.3 LTS
    Release: 18.04
    Codename: bionic


    I have dit it on a test environment with a database with less values, no problem at all.

    The migration on the production environnement have been a success.
    But sometimes, and when I'm restarting the zabbix-server service, an UPDATE sql request is sent to postgresql which launch a process that takes all RAM, and swap and the server is completely lost... If I wait, zabbix can restart but it's a normal behaviour.
    In the logs, it's always when syncing trend datas.

    I have tried to tune postgresql without success. Do you have any solutions for me ?

    Our databases size is about 100GB
    The server has 4 cpu / 16 GB of RAM (before the upgrade we had 4GB)

  • Glencoe
    Zabbix developer
    • Oct 2019
    • 152

    #2
    Looks like https://support.zabbix.com/browse/ZBX-16347

    What version of TimescaleDB do you use?

    Comment

    • sgaunet
      Junior Member
      • Nov 2019
      • 2

      #3
      Exactly, thanks, I will follow the bug.

      My TimescaleDB version is timescaledb 1.5.1

      Comment

      Working...