5 Replies Latest reply on Apr 11, 2017 12:25 PM by -luca-

    disk health monitoring


      We have started monitoring a few Geovision DVRs on NPM (they're OEM boxes, running Windows 7 embedded) via WMI.

      The server's volumes show up on the node (vital stats) page with the correct labels / utilization data. 

      Is there any way we can create an alert when a disk fails with NPM?



        • Re: disk health monitoring

          You could do a quick alert and structure it off of the volume and whether it's responding or not, or you could set it to fire off when it's in an "unreachable", "unknown", "down", etc. state.


          To test it you could make an alert and point it at an external USB disk or something, then just unplug it and wait, see what status the volume goes into and then alert off of that maybe?


          I think I'm answering what you're asking, think...


          This all of course wouldn't be one alert, but you see what I mean?



          2 of 2 people found this helpful
            • Re: disk health monitoring

              mbird, this seems to be exactly was I was looking for.    I created an alert and set the scope to the server's ip, and the trigger condition.  When I tested the alert in action manager, I got a list of all the servers we are monitoring, and had to drill down to the correct server and pick a specific volume. I was wondering if this is normal, or if the simulate/execute windows should have already picked the correct server based on the IP address I selected in the alert's scope. Hope this makes sense.



              • Re: disk health monitoring

                I got around to test this with an external USB drive:


                1. added USB drive

                2. list resources > force refresh

                3. added the new partition (volume) to monitoring

                4. checked that new volumes size / data usage were populated in vital stats


                5. after a while, I disconnected the external drive


                I can see that size and usage info is gone from Vital Stats for the disconnected volume, but no alert gets created. 

                Should I change the trigger condition?  I'm using NPM 12.0.1.