Re: Last Call: Security Architecture for the Internet Protocol to Proposed Standard

[Valdis.Kletnieks at vt.edu]

On Wed, 01 Apr 1998 12:59:06 PST, Phil Karn said:
> Explicit congestion notification would help considerably.  This
> doesn't necessarily require the definition of new bits in the IP and
> TCP headers; you can actually do quite a lot with the oft-maligned
> ICMP Source Quench message. I've found it quite effective in the
> common case of a host on a private leaf Ethernet sending data through
> a dialup router to the outside world. It certainly reacts more quickly
> than a mechanism that requires a far-end turnaround of the congestion
> bit.

Wow.  Full circle time.  Didn't modern TCP congestion control come
from Van Jacobsen's observation that 98% of the time, a lost packet
was due to router buffer overflow, and thus should be treated as
an implicit ICMP Source Quench?

I guess the Jargon File was right about the Cycle of Reincarnation -
currently, we're just finishing up pulling 3 zillion 10-baseT runs
to replace the one thinwire run that replaced the 3 zillion CBX data
lines... ;)
-- 
				Valdis Kletnieks
				Computer Systems Senior Engineer
				Virginia Tech

Attachment: pgpu4wNedjZ1U.pgp
Description: PGP signature