Hi!
Wondering how other NAV'ers are doing link surveillance?
I'm mostly interested in downed routing-ports, but also normal switch-ports, both host-access and switch-trunks, could of interest.
Any input is appreciated!
Cheers,
-Sigurd
Apart from using link traps and snmptrapd, I don't think NAV has any other way of doing it. You're going to have a problem ruling out "normal" switchports as well with doing it this way. So we're going to use LMS (old Ciscoworks) to do that until we can do the same with NAV :-)
-Vidar
-----Opprinnelig melding----- Fra: Sigurd Mytting [mailto:sigurd@mytting.no] Sendt: 8. desember 2010 18:20 Til: nav-users@uninett.no Emne: Link surveillance with NAV
Hi!
Wondering how other NAV'ers are doing link surveillance?
I'm mostly interested in downed routing-ports, but also normal switch-ports, both host-access and switch-trunks, could of interest.
Any input is appreciated!
Cheers,
-Sigurd
On Wed, Dec 08, 2010 at 06:20:28PM +0100, Sigurd Mytting wrote:
Hi!
Wondering how other NAV'ers are doing link surveillance?
I'm mostly interested in downed routing-ports, but also normal switch-ports, both host-access and switch-trunks, could of interest.
Any input is appreciated!
At the moment, NAV only features linkUp/linkDown SNMP trap handling, which means that you'll need to do fine-grained configuration of devices to send traps only for those ports you would be interested in. There's also the weakness that a trap message may get lost (it's unreliable UDP, after all).
Also, the interface link status that NAV displays is only collected every 6 hours.
We are talking, however, of writing a plugin to ipdevpoll that will exclusively poll link status more often, and create alerts when uplink/downlink ports change status (doing so for all access ports would likely be infeasible, as it would quickly flood NAV's event system).
Anyone else find the latter idea interesting?