[Int-area] tunnel fragmentation

"Templin, Fred L" <Fred.L.Templin@boeing.com> Wed, 14 October 2015 16:47 UTC

From: "Templin, Fred L" <Fred.L.Templin@boeing.com>
To: "int-area@ietf.org" <int-area@ietf.org>
Thread-Topic: tunnel fragmentation
Thread-Index: AdEGn5KQwEeijcUfQtOfPclQIDKZIg==
Date: Wed, 14 Oct 2015 16:44:21 +0000
Message-ID: <2134F8430051B64F815C691A62D9831832F20730@XCH-BLV-504.nw.nos.boeing.com>
Accept-Language: en-US
Content-Language: en-US
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/int-area/0Jy5flM9QIHk-tETHqhJO3R2n_c>
Subject: [Int-area] tunnel fragmentation
Precedence: list

In intarea list discussions in the August timeframe, it was established that there
are circumstances when the tunnel ingress may need to fragment and have the
egress reassemble. For tunnels over IPv4, IPv4 fragmentation may not be a good
alternative in all cases due to the issues identified in RFC4963. For tunnels over
IPv6, IPv6 fragmentation might not work over paths that unconditionally drop
IPv6 fragments. These issues could be addressed using tunnel fragmentation,
where the encapsulation layer itself includes a fragment header.

One example of this is GRE encapsulation. If the GRE header were extended
to include a fragment header, then the GRE encapsulation layer itself could
perform fragmentation so that IP layer fragmentation is not necessary. A
fragmentation header extension for GRE is proposed in:

https://datatracker.ietf.org/doc/draft-templin-intarea-grefrag/

A second example is Generic UDP Encapsulation, as discussed in:

https://datatracker.ietf.org/doc/draft-herbert-gue-fragmentation/

What do people think about this?

Thanks - Fred
fred.l.templin@boeing.com

[Int-area] tunnel fragmentation Templin, Fred L