Monitor IP interfaces

Hi There

We are slowly migrating from Nagios XI into Solarwinds HCO and Id like to talk about monitoring as we are drowning in Alerts at the moment.
I can work on our SNMP based monitoring, but one thing we did do in Nagios was monitor every IP interface at both ends of the link which we found very useful.
This was especially useful in a failure scenario as it gave us a clue where the issue might have started based on the first alert received.
How can I replicate this in HCO, I understand it would be alot of work but for our team it is something we want to do.|
So if I have a switch with 30 IP interfaces, how do I set up a ICMP monitor for each interface?

Thanks


  • If I understand your question you want to tie together a link down condition and a specific SNMPTrap message to generate an alert? This is possible through using primary and secondary conditions in the alert trigger conditions. The primary section would be used to define the link down part of the alert and then the secondary section would do the matching for your SNMPTrap. Set up the Primary condition and then use the AND to add the additional condition for the SNMP part.

    Top part is the interface down part and bottom part is the SWQL query to match the SNMPTrap

  • HCO or NPM in general will monitor the status of your monitored  interfaces by default and can trigger alerts. I believe they have a default alert (Interface is down) which should be able to accomplish what you're looking to do. Default status intervals I believe are every 120 seconds? This can be adjusted as needed in your settings.

    One other feature I'd like to highlight is NetPath Services. Especially if it's a critical route within your infrastructure, you could monitor both ends of that path and gives you a nice hop by hop analysis along the way. It's like a tcp traceroute on steroids that gives you a nice map of the route.

    If I missed anything you were looking for .. please don't hesitate to reply so we can converse!

  • HI
    I did see netpath but with 20 comms rooms and hundreds of switches with multiple IP interfaces it didn't look like the correct tool to use for monitoring the LAN? I can see how it would be useful for monitoring a cloud based service though.
    I was hoping to set a simple ping monitor for every IP interface we have that wasn't tied in to the Node down trap.


  • Hmm .. I haven't tried it myself, but you might be able to go through the network discovery wizard and specify those individual ip addresses. Specify no SNMP credentials (so it doesn't know how to talk to the switch) and check the box to allow ICMP only monitoring.

    Though the only downside to that is you'd have a bunch of IP's in your inventory with 'Unknown' vendor, but you could probably group them together for the specific offices and keep an eye on each group?