[NANOG] Microsoft.com PMTUD black hole?

Wed May 7 19:38:32 UTC 2008

Valdis.Kletnieks at vt.edu wrote:

> The usual case where you get screwed over is when the router trying to toss
> the ICMP FRAG NEEDED is *behind* the ICMP-munching firewall.  And in case (2),
> you still can't assume that path MTU == local MTU, because your local MTU is
> likely 1500, and the fragging router often trying to stuff your 1500 byte
> packet down an PPPoE tunnel that's got an MTU of 1492....

Yes, but my point was precisely that one OR the other side (server OR 
client) is going to NOT have the ICMP-munching firewall in between 
itself and the "RITM" as I have affectionately been calling it (although 
it is definitely possible that there are two ICMP-munchers on either 
side of the RITM).

And case #2 is exactly what is occurring right now _anyway_: hosts 
assume that path MTU == local MTU even if there is already an active 
PMTU cache entry from a recent earlier communication with the remote 
host.  So I don't see how making that assumption _after_ making an 
honest attempt at actively determining whether or not it is actually the 
case is any more broken than they way things are already being done.

The problem is that, as I realized at the end of the message you quoted, 
there are potentially multiple paths between the same two hosts, and the 
path that the packet takes in one direction is not guaranteed to be the 
same path that the packet takes in the opposite direction.

-- 
Nathan Anderson
First Step Internet, LLC
nathana at fsr.com