Orion NPM - Node/interfaces consistently report unknown states

Oristilli

We’ve had an increasingly high number of nodes reporting an unknown state in the Orion web interface. These devices are from various Vendors and run on different platforms(Cisco, Windows Server, Linux, Juniper). We have configured these devices no different from other devices . An example would be one of our Cisco WS-C3524-PWR-XL running system image c3500xl-c3h2s-mz.120-5.WC17.bin.We have configured the trunk interfaces to a firewall interface and another LAN switch (same model) but Orion is reporting the node as being in an unknown state for these monitored interfaces and they are also greyed out in the web console. This unknown status doesn’t seem to stop Orion from polling the switch and collecting SNMP data and as with most of our other devices we have configured the node with SNMPv2 .There are a lot more like this example with similar symptoms and there isn’t a pattern as we have similar devices with the same platform that work alright. Has anyone got a fix , workaround or explanation for this problem?

ecklerwr1

This is what I try first:

Usually when I get a node in this condition I go into manage nodes... do a rediscover on it... then list resources.  Usually something has changed on the device and I might need to select a new interface etc.  After that the node does back to normal.  On server this is often a change with a disk volume... On a router or switch possibly a module was added or something else changed.  Either way Orion gets confused about the current interfaces on the device and needs to have them updated.