Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 30 Jun 1998 18:16:46 -0700 (PDT)
From:      Marc Slemko <marcs@znep.com>
To:        Yee Man Chan <ymc@eecs.umich.edu>
Cc:        freebsd-hackers@FreeBSD.ORG, questions@FreeBSD.ORG
Subject:   Re: something is wrong with freebsd (fwd)
Message-ID:  <Pine.GSO.3.96.980630181026.8604A-100000@redfish>
In-Reply-To: <Pine.GSO.3.96.980630143958.2508C-100000@shivam.eecs.umich.edu>

next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, 30 Jun 1998, Yee Man Chan wrote:

> Hi, 
> 
> 	My supervisor sends me the result he collected using tcpdump when 
> running my client-server program post before. Does anyone have an idea of
> what's going on?
> 
> 
> Thanks.
> Yee Man
> 
> P.S. If anyone can't get the source code I posted before. Please don't
> hesitate to request it directly from me.
> 
> Msg from my supervisor:
> 
> Here're the tcpdump of the client and server on freebsd only, sun only,
> or one on each.  You can see that freebsd consistently splits up pkts
> of size 101 to 207 bytes into 2 pkts, but sending pkts of size 208 to 
> maxseg bytes in one pkt.

Known problem; wasn't introduced by FreeBSD code, but hasn't been fixed
either.  Should be in the archives somewhere.  Search for a 
"why 100 byte TCP segments" thread.

There is a half page on it in Stevens vol 3(?) or something, not at
home now can't check.

It has to do with how it shoves things into mbufs and mbuf clusters.


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-questions" in the body of the message



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.GSO.3.96.980630181026.8604A-100000>