[j-nsp] OSPFv3 interop/tuning recommendations?

[Justin M. Streiner]

On Fri, 10 Jun 2011, Jonathan Lassoff wrote:

> Perhaps look at the neighbor state through the 10 and 30 second
> windows and see if you can see what state is taking the most amount of
> time.
> Are you exchanging a lot of routes? Maybe the back and forth of many
> DBD packets are faster to be acknowledged on one host, but not the
> other?
>
> If you have the hardware, or are willing to do SVIs on the Cisco, try
> and get a tap of the OSPF traffic and analyze protocol timings in
> tshark or wireshark.

I am moving a decent number of IGP routes in OSPFv3 - about 2,200 IPv6 
routes.  I'm set up to do a series of stress tests at the moment, to 
determine how much the number of IGP routes impacts convergence times. 
I've also tested using a summary prefix on one of my routers that's 
acting as a route injector, and the adjacency times did not change 
significantly.  The summarization cut the number of v6 LSAs in the 
backbone area by about half.

>From a non-scientific evaluation, it looks like most, if not all phases of 
establishing the adjacency take longer on one of the Cisco<->Juniper 
adjacencies.  Init->2Way, ExStart, Exchange, passing DBDs, etc...

I can do SVIs, but that starts to get away from the topology that's in 
production today.  If I need to, I can set up a span port on the 6509s 
to a machine that's running Wireshark.

I also messed around with IS-IS, in place of OSPFv3, but IS-IS's constant 
complaining about address family mismatches where there were none got 
annoying pretty quickly ;)

jms