This topic contains 5 replies, has 4 voices, and was last updated by  Robert James 10 months, 3 weeks ago.

Using Nagios to monitor Iguana status

  • Our Support team is monitoring most of the company devices, but none of the Iguana instances. At present, this is because the mechanism being used in Nagios, which generates the alerts and auto-creates a support ticket for the team. Is anyone using Nagios to monitor and alert on channel and server status?

    -Robert James,
    Interface Analyst,
    GetWellNetwork, Inc.

    Iguana has a robust web api for monitoring overall and individual interface statuses; you can get a look at the data returned by clicking on the Web API links in the lower right hand corner of the Iguana dashboard.

    Nagios has a variety of web monitoring agents that should be adaptable to the Iguana API.

    Jeff Drumm ◊ VP and COO ◊ HICG, LLC. ◊ http://www.hicgrp.com

    We use Nagios internally – and I think Scott Ripley gave a talk about using Nagios with Iguana at our user conference last year. I’ll see I can get him to post something.

    Eliot, that would be very helpful. I don’t control our internal monitoring, so if I can show them something that’s already done, it will make the idea of doing that monitoring of Iguana with the existing tools that much easier.

    -Robert James,
    Interface Analyst,
    GetWellNetwork, Inc.

    the Nagios check_http (standard) plugin:
    http://nagios-plugins.org/doc/man/check_http.html

    could be used with the Iguana status URL
    http://<iguana-host&gt;:<iguana-port>/monitor_query?Compact=N

    to check for an expected string or regex:
    check_http -I <iguana-host> -p <iguana-port> -a <iguana-username>:<iguana-password> -s “NumberOfChannels=’20′” -u /monitor_query?Compact=N

    i.e. in this case the check would fail if the expected 20 channels weren’t running…

    if a simple string match isn’t sufficient:
    -> a custom script could be written that again fetches the Iguana Status URL / parses the XML / determines if the status is OK (or not) and returns the result to Nagios…

    Thanks, Scott. I’ll have to look at something more custom, as the simple “http check” isn’t going to suffice. I need to be able to provide alerting on a per-channel basis.

    -Robert James,
    Interface Analyst,
    GetWellNetwork, Inc.

Tagged: 

You must be logged in to reply to this topic.