[c-nsp] 6500 redundant supervisors and non-NFS aware RIP

[Justin M. Streiner]

On Mon, 14 Jul 2014, Jeffrey G. Fitzwater wrote:

> Q1 Since BGP is NFS aware but only after the second SUP is installed, 
> will BGP peering stop and restart?
> We see that two of our ISP are currently sending GRACEFUL RESTART 
> messages but our end is not yet because no secondary sup.

We have been installing dual supervisors in several of our core 6500s over 
the past few months, and it's been my experience that you might lose a 
packet or two, but not enough to trigger BGP hold timers.  We don't have 
Cisco gear at our Internet borders, however I recall IBGP chugging right 
along with no major headaches.

> Q2 What happens with the RIP routes?
> We know we have to move to OSPF, but thats another day;  Soon I hope.

I can't speak to how RIP behaves, but OSPF and OSPFv3 run pretty much 
without a hitch.

I would definitely plan for a few brief outages as you install the second 
sups, however if anyone sees any blips, most likely they will resolve 
themselves before people can pick up the phone.

FWIW, this is also in a large .edu environment.

jms