From: Aminul Siddiqui (aminul.siddiqui@gmail.com)
Date: Wed Oct 17 2007 - 18:07:58 ART
Hi Alex,
What do you mean by two MPLS circuits? Is it two interfaces connected to the
MPLS network. In that case, both interfaces may be in one ima group. So, you
will not notice until the last interface goes down and you have backup VPN
after that. You should be getting the information of interface down/up on
the log though.
Thanks.
Aminul
On 10/17/07, Alex Steer <alex.steer@eison.co.uk> wrote:
>
> I was looking at monitoring route changes for a customer.
> (Un)fortunately they have a very resilient network and devices don't
> become unreachable upon failure of pretty much anything. Typically they
> have 2 main MPLS circuits and a backup mesh VPN (not adjacent to
> customer equipment).
>
> I looked at using SNMP to report routing table changes, its pretty
> static unless we make changes. The best I could do was monitor a mib
> route change counter. This counter starts at 0 and just keeps going up.
> Never does it go back down to 0 after a certain period, so pretty
> useless. I'm sure there must be a way without doing something daft like
> rmon but I'm not a CCIE and don't know how.
>
> Used different loopback interfaces in the end to determine when the
> primary network was down. Still can't tell if we loose one of the main
> circuits though as they are connected behind CPEs. I hope the SP can.
>
> Ooow look, the lights off
>
> Alex
This archive was generated by hypermail 2.1.4 : Fri Nov 16 2007 - 13:11:15 ART