Ad Widget

Collapse

Show cpu load of router internal processes through Zabbix?

Collapse
X
 
  • Time
  • Show
Clear All
new posts
  • redes
    Junior Member
    • Jun 2014
    • 2

    #1

    Show cpu load of router internal processes through Zabbix?

    Hello all.

    It is possible to view a graph of cpu load of an internal router process?

    For example in a Cisco router you can view the internal processes and the cpu they are using with the command show processes cpu.

    Is there a way I can create an item that uses SNMP and get specific data from a specific internal process of a router?

    Thanks in advance.
  • jan.garaj
    Senior Member
    Zabbix Certified Specialist
    • Jan 2010
    • 506

    #2
    If you are able to collect data, then it's possible.
    I think that you can collect only total CPU metrics with SNMP. CPU utilization per process is not available trough SNMP way:
    This document describes how to collect CPU utilization on Cisco IOS® devices that use Simple Network Management Protocol (SNMP).


    But you can probably use telnet:
    This document describes common symptoms and causes of high CPU utilization on Cisco routers and provides guidelines and solutions for common issues.

    and then just parse output from "show processes cpu" and send it to Zabbix with zabbix_sender.

    Anyway read manual of your Cisco device, what is (not) possible :-)
    Devops Monitoring Expert advice: Dockerize/automate/monitor all the things.
    My DevOps stack: Docker / Kubernetes / Mesos / ECS / Terraform / Elasticsearch / Zabbix / Grafana / Puppet / Ansible / Vagrant

    Comment

    • redes
      Junior Member
      • Jun 2014
      • 2

      #3
      Hi.

      The only thing i did is collect total cpu via SNMP as you said.

      So you say use Telnet to connect to the router, type in the show processes command, export it as txt and finally send it with zabbix_sender right?
      And I must configure and new item using Zabbix trapper?

      Comment

      • jan.garaj
        Senior Member
        Zabbix Certified Specialist
        • Jan 2010
        • 506

        #4
        Yes, exactly.

        Eventually you can use telnet checks - https://www.zabbix.com/documentation.../telnet_checks
        , but efficiency will be worse, because you have to make telnet for every metric.
        Devops Monitoring Expert advice: Dockerize/automate/monitor all the things.
        My DevOps stack: Docker / Kubernetes / Mesos / ECS / Terraform / Elasticsearch / Zabbix / Grafana / Puppet / Ansible / Vagrant

        Comment

        Working...