This discussion has been locked. The information referenced herein may be inaccurate due to age, software updates, or external references.
You can no longer post new replies to this discussion. If you have a similar question you can start a new discussion in this forum.

Managing Resources On Large Network Devices In NPM

We have some large network devices Cisco 9k's and 10k's and some 7750's that we are trying to manage with Orion NPM.  The problem is that we are having a really had time managing the elements associated with these devices when they have thousands of interfaces.  For example, we had to add a 7750 with ~11,000 interfaces to NPM with the Sonar Discovery.  We were not able to use the 'Add Node' option on the Manage Nodes page because it would time out.  The problem now that we are managing the 7750 via the discovery is that we need to add new interfaces as business dictates and we are not able to successfully 'List Resources' on the node. 

In troubleshooting this, we noticed that a walk of the ifTable with the SNMP MIB Browser from EE Toolset, it will list the ifIndex numbers really pretty quickly but when you 'List Resources' for a node from Orion NPM, it is much much slower.  I understand that a 'List Resources' is collecting from several OID's for the interface from the ifTable rather than just listing the ifIndex.

These devices are critical to business operation and very high visibility.  Some of the interfaces affect very important business metrics that need to be monitored.  Can SolarWinds tell me if there any changes that we can expect that will make monitoring a network device of this caliber something that we can monitor in NPM?  If there is a config that I might be missing here to make this complete, I would appreciate any suggestions.  I've set my SNMP timeouts and retries out and still having trouble.  If we were to 'List Resources', if it is successful, at the rate it is running, our expectation is that the walk would complete somewhere in the neighborhood of 3 to 5 hours.  Any help/direction is appreciated.


Thanks,
Jason Henson
Loop1 Systems
www.Loop1Systems.com