Hi, On Fri, Feb 06, 2009 at 04:32:03PM -0600, Justin Shore wrote: > Gert Doering wrote: > >I can only second this. If you have a dedicated point-to-point interface > >for things, tacking the route on the interface is usually more robust than > >pointing towards a gateway IP that might not be there, or might be learned > >recursively over another interface, etc. > > I'm going to have to think on this. This could be a good solution for > us, though again I don't ever foresee the need to re-IP a loopback. Can > anyone give me an example of such a scenario?
Well, one example would be "you currently have 3 routers with lots of
individual E1 cables. Your telco provider offers you to move all the
E1s into a STM-1, and you can serve the same customers from a single
new router".
Now of course you could move all 3 former loopback IPs to the new box and
migrate all customers at once - or assign a single new one, and migrate
them one by one.
(This might not be the most typical example, but "moving customers to a
different router" is happening out there :) - and then you either have
something that's not dependent on the loopback IP, or you have pains)
gert
--
USENET is *not* the non-clickable part of WWW!
//www.muc.de/~gert/
Gert Doering - Munich, Germany [email protected]
fax: +49-89-35655025 [email protected]
pgp0fZRYZPqUk.pgp
Description: PGP signature
_______________________________________________ cisco-nsp mailing list [email protected] https://puck.nether.net/mailman/listinfo/cisco-nsp archive at http://puck.nether.net/pipermail/cisco-nsp/
