3 Replies Latest reply on Nov 24, 2014 8:08 AM by mharvey

    Advanced Alerting Help


      I apologize in advance for this wall of text, but I feel like I have a pretty complicated issue I am trying to deal with.

      To start, the computers we have deployed have huge problems with dropping hard drives; they will basically disappear somehow unless the computer restarts and they reappear, so there's that.

      So keeping that in mind, I am attempting to configure an alert that notifies my users if a hard drive in one of their machines:

      a) stops responding
      b) reaches less than 20% free space.

      The alerts work great, but the problem I am having, is that generally when the free space alert is sent out, my users will remove the full drive and place a new one (standard practice for what we do), with the same drive name. This creates some confusion for Solarwinds, as a hard drive that could have been removed and replaced knowingly will still generate an alert based on the criteria in the trigger conditions listed below, and confuses my users.

      The alert we have configured for hard drive space is below:

      WHERE    (    (Nodes.City IS NOT NULL) AND     (Nodes.Status <> '2') AND     (Nodes.Mute = 0 )  AND
      (      (Volumes.VolumeDescription LIKE 'R:\%') AND       (100-NullIf(VolumePercentUsed,-2) <= 20)) OR
      (      (Volumes.VolumeDescription LIKE 'S:\%') AND       (100-NullIf(VolumePercentUsed,-2) <= 20)) OR
      (      (Volumes.VolumeDescription LIKE 'T:\%') AND       (100-NullIf(VolumePercentUsed,-2) <= 20)) OR
      (      (Volumes.VolumeDescription LIKE 'U:\%') AND       (100-NullIf(VolumePercentUsed,-2) <= 20)) OR
      (      (Volumes.VolumeDescription LIKE 'V:\%') AND       (100-NullIf(VolumePercentUsed,-2) <= 20)) OR
      (      (Volumes.VolumeDescription LIKE 'X:\%') AND       (100-NullIf(VolumePercentUsed,-2) <= 20)) OR
      (      (Volumes.VolumeDescription LIKE 'Y:\%') AND       (100-NullIf(VolumePercentUsed,-2) <= 20)) OR
      (      (Volumes.VolumeDescription LIKE 'Z:\%') AND       (100-NullIf(VolumePercentUsed,-2) <= 20))


      The alert we have configured for non-responsive hard drive is below:

        (Nodes.City IS NOT NULL) AND
        (Nodes.Vendor = 'Windows') AND
        (Nodes.Mute = 0) AND
        (Volumes.VolumeResponding = 'N') AND
        (Nodes.Status <> '2')

      My main question is how can I tweak these alerts to essentially "ignore" hard drives that are replaced? I still need the alerts to be sent out when the hard drive disappears and reaches 80% capacity, but not when a user removes and replaces a drive.

      Again, sorry for the long post, but I am at a loss as to how to mitigate this confusion.


        • Re: Advanced Alerting Help

          I'm actually surprised you are seeing the new drives.  Typically SolarWinds will hang on to the new drive and just hold it's existing last polled value and not know to monitor the new drive until a list resources is done on the device.  Then the new volume has to be selected.  Once done, you'd then have two volumes of the same or close to the same name, one of which would have to be removed via manage nodes.

          1 of 1 people found this helpful
            • Re: Advanced Alerting Help

              Thanks for your response, mharvey. I also thought Solarwinds would exhibit the same behavior, but upon rediscovering the drives, the old "unresponsive" drives show alongside the new "good" drives. Even if I unmanage drives and re-manage drives, they still persist to show up.


              As you can see, the nearly full drives are the ones that have been dropped, and the new one is displayed below.

              Any thoughts?

                • Re: Advanced Alerting Help

                  You'll have to go into Manage Nodes, expand the node, select the old drives and delete them.  The system won't show them on a list resources as they aren't there, and won't know to remove them.  You would need to delete them out of Manage Nodes.