Phil Mayers <[email protected]> wrote: Hello Phil,
>> Also a good guess, but it's a fresh installation of 6.1(2) and I've >> verified that there is indeed a CoPP entry for "match protocol mpls". >> The class is not dropping a lot of traffic. > I'm assuming you're aware of the way traceroute works for P-routers, > specifically that the generated ICMP packet is sent onwards down the > original LSP then turned around by the far-end PE. Have you (or can you) > verify that this packet is being generated (SPAN on the next-hop input > interface)? Perhaps the far-end PE is dropping the packet because it > dislikes something about it, something specific to the NX-OS boxes? > > Determining whether this ICMP is being generated seems a good first step. Yes, I'm aware of how traceroute works. Will take me a day to get a sniffer there, but I'll check. Problem happens with all combinations of NX-OS/IOS head-end and far-end PEs and regardless of the ttl-expiration pop setting. Bernhard _______________________________________________ cisco-nsp mailing list [email protected] https://puck.nether.net/mailman/listinfo/cisco-nsp archive at http://puck.nether.net/pipermail/cisco-nsp/
