3 Replies Latest reply on May 14, 2013 4:01 PM by rob.hock

    How do I remove bogus hardware sensors after a server rebuild?

    mmoore.work

      The basic story is that the server disks died and the was rebuilt from install media with new disks.  Now some of the snmp indices for devices are different and SW sees 2 of the same device - for example, DIMM_B2 is listed twice - one is unknown state and is the original dimm, whereas the new one is readable and all is green.  how do I remove the items that are no longer addressed/indexed correctly?  I can remove the server altogether, but would like to keep historical data if I can.  Also, if I remove and re-add the server to NPM, I think I will have to adjust any maps that used that node?

       

      Thanks,

      Matt

        • Re: How do I remove bogus hardware sensors after a server rebuild?
          Chrystal Taylor

          Are you referring to volumes or the hardware health sensors?  For volumes, you can delete them from the database.  You would want to delete the volumes in question directly from the volumes table.  I just ran into this issue with one of my clients.  The drives were renamed and then orphaned by SolarWinds.  You could not see them in list resources or by expanding out the node in Manage Nodes.  Deleting them from the database resolved the issue.  I wound up creating a report to show all of the orphaned volumes, so that we can keep an eye on it in future, and then deleted them from the database based on the volume IDs which are unique to each one.  If it is hardware health sensors though, I am not sure what all would be affected by deleting them from the database.  In that case, you may have to just list resources and deselect it.  Run a new poll, list resources again and add it back.

           

          Hope this helps.

           

          Chrystal Taylor

          http://www.loop1systems.com