Richard A Steenbergen wrote: > On Mon, Nov 09, 2009 at 09:51:40AM -0800, Jared Gillis wrote: >> While I agree with these statements, our issue is not tree >> recalculation/convergence. Our issue and driving need for IS-IS >> multiarea is the fact that we have 3750ME's which can only hold ~2k >> routes in the TCAM in our IS-IS domain, and we'll very rapidly exhaust >> the TCAM unless we can do route summarization (i.e. upstream L2's send >> default/ATT only). > > So why can't you put the the routes into iBGP, use your IGP only for the > loopbacks, and learn a default route from your upstream devices?
That's exactly what we are doing, but still expect to exhaust the TCAM before too long, especially if we add support for IPv6 (which is a long-term goal). -Jared _______________________________________________ cisco-nsp mailing list [email protected] https://puck.nether.net/mailman/listinfo/cisco-nsp archive at http://puck.nether.net/pipermail/cisco-nsp/
