[c-nsp] STP in L2TPv3

[Chris Flav]

>I had a case open with Cisco on this same issue pending for 6 months or so then I finally closed the ticket;  what I saw was basically the STP pkts was >arriving out-of-order due to fragmentation; the remote end never get STP updates as such.
>
>'ve been meaning to test this with EoMPLS over GRE to see if STP behaves the same way; one funny thing I did see though that if I turned off the PXF engine >on the head-end, then STP gets transmiitted then frrag/defrag properly; down side is that PXF engine provides for enhanced throughput of L2TPv3 (which is >very CPU intensive if process switched); in the end Cisco TAC advised us to run VTP domain in transparent mode so as not to pass STP pkts; that's what we >are doing now.
>
>If I get around to testing STP updates on EoMPLS over GRE then I'll post my results on this distro list; good luck.

Was your issue on the 800 series specifically or using L2TPv3 on any hardware platform?

This is somewhat annoying....

Thanks,

C.


      __________________________________________________________________
Looking for the perfect gift? Give the gift of Flickr! 

http://www.flickr.com/gift/