On Sun, 27 Sep 2015 17:33:35 +0200 ken.livesey@gmail.com wrote:
[WARNING plugins.typeoid.typeoid] [inventory DA13_SW07.xxxx.xxx] Netbox has changed type from unknown to unknown (sysObjectID 1.3.6.1.4.1.9.1.367)
This warning message may be badly worded (and should, IMHO, be an INFO level message, rather than a WARNING).
What this means is that the device had no type entry, but that ipdevpoll has now created a new type entry based on the collected sysObjectID, .1.3.6.1.4.1.9.1.367. However, the new type entry still represents an unknown entity, so it should later be edited in SeedDB by an admin to make sense (This may be why the WARNING level was chosen for this message)..
The newly created type entry will be attached to the "unknown" vendor, using the sysObjectID itself as the type name, while the type description will be whatever ipdevpoll got from the device's sysDescr value.
[ERROR jobs.jobhandler] [inventory DA13_SW07.xxxx.xxx] Job 'inventory' for DA13_SW07.xxxx.xxx aborted: Plugin ciscovlan reported a timeout
This plugin uses the proprietary mibs CISCO-VTP-MIB, CISCO-VLAN-MEMBERSHIP-MIB and CISCO-VLAN-IFTABLE-RELATIONSHIP-MIB to retrieve VLAN configuration information. The plugin will only attempt to use the MIBs on a device with Cisco's enterprise number (9) in its sysObjectID.
A timeout here usually just means that this device is a bit on the slow side when building responses to queries in one of these MIBs. Some of the queried values are large octet strings representing lists of ports or vlans, which could be the cause of this.
You will likely fix your problem by tweaking the SNMP settings in `ipdevpoll.conf`. Either decrease the max-repetitions option, increase the timeout option, or both.
You can manually run an inventory job against your device to verify whether it helped:
/usr/lib/nav/ipdevpolld -J inventory -n da13_sw07