1 Reply Latest reply on Aug 30, 2011 5:31 PM by derik.pfeffer

    Best Practices - Node w/Multiple Interfacese in NPM

    kpieckiel

      Greetings, friends!

      I have a large field network (Cisco) with redundant paths to nearly every node using a mix of SONET and DS1 circuits.  When I first installed Orion and did a network discovery many of my routers were found twice--once for each interface that had an IP address on one of the subnets on which that router resides.  (e.g. Router A connects to Router B via subet 1 and Router B connects to Router C via subnet 2.  Router B was found twice, once with its IP address on subnet 1 and again with its IP address on subnet 2.)

      When it comes to routers and situations like this, is it best practice to keep both nodes or to remove one of them?  Why?

      Thank you,
      Kevin

        • Re: Best Practices - Node w/Multiple Interfacese in NPM
          derik.pfeffer

          Kevin,

           

          The biggest reason to delete any duplicate device that was found during Network Scan on many different IP addresses is that you are duplicating the work for the collector, network between the collector and device, and the device itself being polled for data.

           

          Think of it in simple terms... It is like having a co-worker ask you what, where, etc.. for lunch while standing on your right side and then asking again from your left side.

           

          Most Network Engineers will use the Loopback address of a Router, Switch, etc.. which is the most reliable interface because it is up as long as any other interface is up. This provides some redundancy when a device is reachable via multiple paths. If you have any other question, please feel free to reach out to me directly.

           

          Derik