On Sun, 9 May 2004, Masataka Ohta wrote:
Back to the original problem, PMTUD depends on the capabilities
of intermediate systems on a path to generate certain ICMP,
generation of which is as complex as fragmentation itself,
that it is not very end to end.
That is, PMTUD is a broken concept.
This is the whole point of reopening the original pmtud working group.
http://www.ietf.org/html.charters/pmtud-charter.html
See draft-ietf-pmtud-method-01.txt I quote:
This document describes a robust new method for Path MTU Discovery
that relies on TCP or other Packetization Layer to probe an Internet
path with progressively larger packets. This method is described as
an extension to RFC 1191 and RFC 1981.
...
The general strategy of the new algorithm is to start with a small
MTU and probe upward, testing successively larger MTUs by probing
with single packets. If the probe is successfully delivered, then
the MTU is raised. If the probe is lost, it is treated as an MTU
limitation and not as a congestion signal.
This does not require any messages from the network, and works just fine with
arbitrary middleboxes that either toss over-sized frames or ICMP
messages.
Comments, suggestions and additions to the text are welcome. The document is
not well baked, but there is running code. There is some (slightly out of
date) background info at http://www.psc.edu/~mathis/MTU/#pmtud
Thanks,
--MM--
-------------------------------------------
Matt Mathis http://www.psc.edu/~mathis
Work:412.268.3319 Home/Cell:412.654.7529
-------------------------------------------
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/ietf