3 Replies Latest reply on Dec 10, 2009 10:13 AM by Andy McBride

    Migrated to a new Polling server, having issues/errors after move!



      We are in the process of moving a majority of our network to a new data center.  The first server we moved was our NPM.  Here is what has been done:

      • VPN Tunnel setup between new (10.x.x.x) and old (192.168.x.x) data center
      • backed up db on original server
      • installed NPM on new server (New Data Center)
      • installed a new DB server (New Data Center)
      • Restored the DB to new server
      • All nodes seem to transfer over
      • Reset nodes to point to new polling server by editing the Engines table, Pollers Table, and Node Table
      • Verified most things are working as we can see our VMWare Environment and all virtual system status from the individual node, most networking equipment, drive spaces seem to math, etc.

      There are a couple nodes that are not working.  One is an Environment sensor.  I readded this node to the NODE list.  I get a warning when trying to add this "Could not reach JobScheduler service after 3 retries" but the node shows up in the list.  Where the Temp/Humidity gauges should be they display "No pollers assigned or no statistic results. Click edit button to change universal device poller assignment!" 

      I get into Universoal Device Poller, find the specific pollers that have been created, the seem to show the assigned node.  One of them does toss an error when selected "Object reference not set to an instance of an object". 

      The odd thing is I can edit the poller, browse the MIB and force test it on the particular node and it yields the correct results.

      What could be causing all this to happen?