All things related to multicast

[Bill Fenner <>]

I MSDP peered with cs.dng.vbns.net for 24 seconds, and got 49836
SA messages.  The top RP's advertised are:

48810 195.83.118.16
 467 138.18.100.1
 140 203.255.248.1
 109 204.147.128.141
 101 171.69.10.13

Note that 195.83.118.16 is the RP for 98% of the SA messages that I
observed.

I also peered with Ohio State, for about 30 seconds, and got 133210
SA messages.  The top RP's from this peering are:

130228 195.83.118.16
1006 193.166.5.252
 858 128.223.253.7
 447 203.255.248.1
 200 171.69.10.13

Note that 195.83.118.16 is the RP for 98% of the SA messages.

My guess is that there's a Peer-RPF-Flooding failure, possibly near
195.83.118.16 but not necessarily.  This failure is causing loops for SA's
originated by that router.  IOS's MSDP seems to reset the TCP connection
if the TCP receive window ever drops to 0, which could explain many
of the resets.  (I also saw connection resets where the window never
dropped to 0, so window closing is not the only reason).

  Bill