Re: TCP Checksum Interoperability

der Mouse <mouse@Rodents.Montreal.QC.CA> Sat, 06 April 2002 00:28 UTC

Date: Fri, 5 Apr 2002 19:28:33 -0500 (EST)
From: der Mouse <mouse@Rodents.Montreal.QC.CA>
Message-Id: <200204060028.TAA25017@Sparkle.Rodents.Montreal.QC.CA>
Mime-Version: 1.0
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: 8bit
To: tcp-impl@grc.nasa.gov
Subject: Re: TCP Checksum Interoperability
In-Reply-To: <E16tdNG-0000uh-00@the-village.bc.nu>
References: <E16tdNG-0000uh-00@the-village.bc.nu>
Sender: owner-tcp-impl@grc.nasa.gov
Precedence: bulk
Status: RO
Content-Length: 807
Lines: 17

>> A TCP riding on IPv6 may receive old segments an unbounded time
>> later without any other network element breaking a spec.
> This was true with IPv4.

No; the TTL is specified in seconds, with the additional note that it
must be decremented by at least one per hop even if the hop takes less
than a second.  Hops that take multiple seconds but decrement the TTL
by only one are out of spec for IPv4.

That's not to say that they're nonexistent, uncommon, or even that they
break stuff in practice.  Heck, I've seen ping show me RTTs upwards of
40 seconds, without even having to resort to avian carriers.  But they
_are_ spec violations.

/~\ The ASCII				der Mouse
\ / Ribbon Campaign
 X  Against HTML	       mouse@rodents.montreal.qc.ca
/ \ Email!	     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B