[c-nsp] BGP and HSRP

[Patrick Greene]

Your iBGP routers must have a full mesh of peers.  Meaning you must setup a peering relationship between all iBGP routers within an AS unless using a Route-Reflector, which you don't need.  Additionally, best practices are to use a Loopback nterface for iBGP peerings.  HSRP is a gateway redundancy protocol not a routing protocol feature.

-----Original Message-----
From: "myNET NOC - Bernd Ueberbacher" <noc at mynet.at>
To: cisco-nsp at puck.nether.net
Sent: 5/9/2007 6:29 PM
Subject: [c-nsp] BGP and HSRP

Hi everyone!

I'm reading this list for a couple of months now and tonight I got my 
first question :-)

We get a new location with 2 internet upstreams and I'd like to run HSRP 
for fail-over. There is a bit of a strange topology though...
My carriers gave me 2x2 /30 for two BGP sessions so I can run on both 
routers a full table BGP session to each of them. The problem(?) is that 
behind those two routers, there is one router who wants to announce some 
iBGP stuff to them. If I run HSRP on the "LAN" side, is it possible to 
make a peering to the virtual HSRP IP? How would BGP handle this or 
wouldn't this work at all?

My "dream" would be that I can peer the internal router with one IP in 
the front, the active router announces those received routes to my 
upstreams and if the active router fails, the secondary takes over and 
starts to announce them over the other peering.

Should I completely wipe this idea and look for something else?


Somehow a stupid explanation and please excuse my poor English ;-)

Greets,
Bernd

_______________________________________________
cisco-nsp mailing list  cisco-nsp at puck.nether.net
https://puck.nether.net/mailman/listinfo/cisco-nsp
archive at http://puck.nether.net/pipermail/cisco-nsp/