DX Unified Infrastructure Management

  • 1.  How to set up an alert when a server reaches an uptime threshold

    Posted Mar 31, 2016 05:56 AM

    I need to set up a simple alert which triggers when a server reaches a defined uptime threshold (eg over 90 days uptime).

    We have known issue with some older (Win 2008) servers where we can no longer RDP to them after they've been up for more than around 90 days due to TCP port exhaustion, so until we can either replace or upgrade these servers, we need to proactively reboot them before they reach this point, so I thought a simple alert triggering when they reach a certain uptime would allow us to do this before the server stops responding.

    I can't seem to see an easy way of doing this with the basic probes such as CDM, can anyone suggest a way of doing this? What additional probes (if any) would I need? We have the basic server and application probe packs.

     

    Thanks,

    Chris.



  • 2.  Re: How to set up an alert when a server reaches an uptime threshold
    Best Answer

    Posted Mar 31, 2016 11:48 AM

    Search for this technote. It has some good information.

     

    ==================================

    How can I monitor system uptime?

    Document ID:  TEC000004793
    Last Modified Date:  11/20/2014

    ==================================

     

    I have found that the ntperf probe has been useful. It is configured with a profile to report and alarm on the Object = System; Counter = System Up Time performance object. 



  • 3.  Re: How to set up an alert when a server reaches an uptime threshold

    Posted Mar 31, 2016 11:54 AM

    Thanks for the reply. I'd actually come to the same conclusion myself shortly after posting this question, typically!

    I've tested using the NTPerf probe for this this afternoon and it works a treat.

     

    Cheers again,

    Chris.



  • 4.  Re: How to set up an alert when a server reaches an uptime threshold

    Posted Apr 05, 2016 12:00 PM

    Whether any possibilities to make it monitor from hub ? Again it a task to push the probe to the robots .



  • 5.  Re: How to set up an alert when a server reaches an uptime threshold

    Posted Apr 05, 2016 12:25 PM

    I will assume the question is remotely monitoring devices for uptime.

     

    If you have a query or utility that can do this then the hub server can run the logmon probe that watches the contents of an output file that the uptime query task would produce.

     

    It does not have to be the hub server. This would work on any server that is tasked with querying other servers for uptime information.



  • 6.  Re: How to set up an alert when a server reaches an uptime threshold

    Posted May 27, 2016 09:03 AM

    I use the ntevl probe, creating a watcher profile for Event ID 6013.

    Log Name:      System

    Event ID:      6013

    Task Category: None

    Level:         Information

    Description:     The system uptime is 7776000 seconds.

     

    For a 90 day alarm, take 90 x 24 x 60 x 60 (days x hours per day x minutes per hour x seconds per minute) to get the number of seconds.

    When the 90 day (7776000 seconds) threshold is reached, you get an alarm.