This scenario may be unusual in the ISP world where genuine AS re-use/overlaps are very rare but is much more common in the Enterprise world and I've faced it myself. e.g. I had to intregrate 2 disparate networks who both used an MPLS service from the same WAN provider.
As such the route tables on both networks had <Private AS>.<SP MPLS AS>.<Private AS> paths and a full dynamic exchange fell foul of the anti-looping. The SP didn't (and wouldn't) provide any of the SP based options (allow-as-in etc.) to get round it. In the end I had to dump all the routes from one network into an igp and then redistribute "clean" routes iback into BGP. Complexity and extra boxes I could have done without. Whilst the ability to remove specific AS numbers from an as-path might be perceived as undesirable or dangerous on the internet - it would have made many an enterprise engineer's life very much easier. Its not like there aren't other configuration options which can be equally dangerous in the wrong hands. Dean Smith ----- Original Message ----- From: "Gary Roberton" <[EMAIL PROTECTED]> To: "Peter Rathlev" <[EMAIL PROTECTED]> Cc: <[email protected]> Sent: Monday, April 14, 2008 10:25 AM Subject: Re: [c-nsp] as-override > Peter > > Thanks for your reply. Why can't Cisco put it as simply as that. > > I am only trying to replace the AS as the one being advertised by R1 is > used > again by another part of the network. i.e. R5 also uses the same AS > number. I need my network to be advertised through to R5 and R5 would > drop > the updates if it saw its own AS number inthe path, therefore I am trying > to > find various options to 'hide' the AS of R1. This was one of the > potential > options but not now as it doesnt do what I want it to do. > > Hope this makes sense. > > Gary _______________________________________________ cisco-nsp mailing list [email protected] https://puck.nether.net/mailman/listinfo/cisco-nsp archive at http://puck.nether.net/pipermail/cisco-nsp/
