Paul,
you seemingly misunderstood or misread my question, so I'll quote
two lines:
! Among hosts and routers that it crosses, the hosts (client and
! server) are normally the ones with lower MTU.
so if the claim above is true, the intermediary routers will never
fragment packets.
I am not talking about theory and specs, but from a *practical* perspective.
and since quoting Stevens is well-seen, I'll cite TCPIPv2, pp. 275:
"On 'world.std.com', 9.5% of the packets received were fragments.
'world' has more NFS activity, which is a common source of IP
fragmentation."
I simply understand from the numbers that 9.5 is a high value and
is only justified by the use of NFS, which is UDP based.
Why will an ISP be setting up a stupid router with a small MTU?
(well, sure if packets originated in a corp net where a stupid router
is used, then *there'll be fragments*, but I m more concerned about
packets sent from home computers, which are harder to track in case
of attack).
best regards,
mouss
> -----Original Message-----
> From: [EMAIL PROTECTED] wrote:
> mouss wrote:
> >
> > I still can't see why a TCP packet need to be fragmented by the
> IP layer.
> >
> > Among hosts and routers that it crosses, the hosts (client and
> server) are
> > normally the ones with lower MTU. But then, they talk TCP and can adjust
> > their payload size. and before hat, first packets are small, so when is
> > fragmentation of TCP packets happening?
>
> [....]
>
> See Stevens' TCPIPv1 pp. 148-157, 236-238, 340-344 for more detailed
> info.
>
> -paul
>
-
[To unsubscribe, send mail to [EMAIL PROTECTED] with
"unsubscribe firewalls" in the body of the message.]
