tilting at windmills discussion 2

Dorian Kim dorian at cic.net
Wed Jun 19 22:01:14 UTC 1996


On Wed, 19 Jun 1996, Vadim Antonov wrote:

> >   512  544  576 1024 1536 2048 2560 3072 3584 4096 4608
> >  .004 .002 .189 .000 .069 .053 .000 .000 .000 .000 .000
> >
> >Collected at a private interconnect between CICNet, OARnet and OSU. over a
> >period of a week or so.
>
> This is not surprising if you think of it.  Short packets are
> pure ACKs, telnet keystrokes and SYNs.  What is worrying is the
> peak at 576.  It means TCP unable to grasp that path MTU is
> really 1.5K.

Agreed. There's alot of bad/old implementations out there, and that's
something that should cause some worry..

-dorian






More information about the NANOG mailing list