SolarWinds Deployment - Discovering and Monitoring Cisco Switch Uplinks

Hello,

I'm part of a team that is deploying SolarWinds to a fairly large environment.  One of the questions we needed to ask ourselves is what interfaces on the switches/routers which we want to alert?  Our thought was that all our uplink interfaces should be monitored to indicate changes in the environment or loss of redundant paths.  The issue we are running into is an easy way with the SolarWinds tools to identify these interfaces.  Additionally, it would be nice to have it do this dynamically, where is new uplinks are created through the environment it would add them to monitoring without manual interactions.  I have been researching online and in this community, but having difficulty finding any documentation to complete this task.  Any help is greatly appreciated.  Thank you!

  • In your Network discovery profiles, you can automatically add devices with pre-defined settings. 


    If the ports are named a specific way, or have the properties you can look for, then use that to add what you need. 

    If you are looking for something else, tell us how you identified 'stuff' you want to see. 

  • Thank you for your reply as it's helpful,  .  We do have an interface naming standard we could use to catch a majority of them.  My fear is the ones done by those in the past which didn't use that standard.  I wasn't sure if there was a way with a Cisco MIB or NCM script to identify all the uplinks using CDP neighbor to catch the ones that may not been labeled correctly. 

  • So once you have added the device, yes, topology is calculated, and CDP is one of the things used. 

    You can use a report with SWQL (not SQL) like: 

    SELECT N.Caption AS [Local Device],

        CDP.IPAddress AS [IP Address],

        CDP.DevicePort AS [Local Interface],

        CDP.DeviceId AS Device,

        CDP.DevicePort AS [Remote Interface]

    FROM  Orion.NodeCdpEntry CDP

    INNER JOIN Orion.Nodes n ON CDP.NodeID = N.NodeID


    And the Network tab of the node has a topology widget (which is also a table that can be queried. 


    You can find topology where the device is added, and the interface is missing by query too.

    However, I tend to go the other direction. My license is per device, not element so I don't care how much gets added, so I do limit what I know is garbage, and then during discovery add in all the other 'Up' ports. (My client access ports are easily filtered out, if that wouldn't be true for you it's likely the primary thing to avoid). Then I report on "network interfaces that don't follow my standard naming" but are monitored. I look to see if I need to add a rule restricting something else, or if I need to get some interface named properly. I try to keep that list short, and it keeps the topology links 'complete' in most cases. FYI, the image above shows 2 switches that I need to get the interfaces named to our standards, these are just default. The topology looks great, but it's on my report for cleanup.