error monitoring linux servers

XMLWordPrintable

    • Type: Problem report
    • Resolution: Won't fix
    • Priority: Trivial
    • None
    • Affects Version/s: None
    • Component/s: Agent (G)
    • None
    • Environment:
      3 computers on the same network: 1 Windows Server (Windows Server 2008 R2 Standard) and 2 Linux servers (Ubuntu 18.04.6 LTS and Ubuntu 16.04.7 LTS). A firewall (SonicWall) is also available.

      Good morning,

      We currently have a Zabbix 6.2 server mounted on an Oracle Linux Server 8.6 where we monitor 3 computers that are on the same network protected with a firewall (Sonicwall). For 3 months they worked without problems. 3 weeks ago, suddenly the 2 Linux servers stopped monitoring each other and I get the error: Get value from agent failed: cannot connect to [[IP of our hosts]:10050]: [4] Interrupted system call

      The Windows server continues to be monitored without any problems.

      We carry out the following checks:

      The Zabbix-Agent service is running.
      Configuration file of correctly configured agents:
      Server= IP of our Zabbix server
      ListenPort=10050
      ServerActive= IP of our Zabbix server
      Hostname: we assign the same name that we have also put when creating a
      new hosts.
      The servers have access to the Internet
      The rules to listen to port 10050 and 10051 are active on our SonicWall
      We check that the Zabbix server is listening to port 10050 and 10051 with the netstat -tunlp | grep 10050/10051
      We reviewed the logs of the 2 Linux servers
      In the Ubuntu 18.04.6 LTS server, the last records that are repeated are:
      20086:20230518:183204.825 active check configuration update from [13.36.191.140
      :10051] started to fail (cannot connect to [[13.36.191.140]:10051]: [4] Interrupt
      ted system call)
      20086:20230519:073210.076 active check configuration update from [13.36.191.140
      :10051] is working again
      On the Ubuntu 16.04.7 LTS server the latest logs are:
      20086:20230518:183204.825 active check configuration update from [13.36.191.140
      :10051] started to fail (cannot connect to [[13.36.191.140]:10051]: [4] Interrupt
      ted system call)
      20086:20230519:073210.076 active check configuration update from [13.36.191.140
      :10051] is working again

      Things we've tried:

      Restart both zabbix agents
      Restart the Oracle Linux server where the zabbix sevrer is located.
      Change the configuration of the agent files specifically the ListenPort line, where we tried different options (ListenPort=10050, ListenPort=10051 and #ListenPort=0.0.0.0).
      Completely remove Zabbix-Agent from both servers and reinstall and configure them. At the same time, delete the hosts that we had in the Zabbix graphical interface.
      The only thing that gave us an error is that we tried to telnet to port 10050 and 10051 on the 2 Linux servers with the command* telnet (Zabbix server IP) 10050/10051*, both connect us without problems on port 10051, in change in 10050 we get an error.

      We reviewed the configuration of our SonicWall and we have verified that the rule that allows traffic from port 10050 and 10051 is activated.

      Is there any other configuration that we must take into account to be able to monitor Linux servers again?

      Thank you

            Assignee:
            Zabbix Support Team
            Reporter:
            Joan Hornos
            Votes:
            0 Vote for this issue
            Watchers:
            3 Start watching this issue

              Created:
              Updated:
              Resolved: