Re: [Int-area] IPv6 fragmentation for IPv4

Hi Tom,

> -----Original Message-----
> From: Tom Herbert [mailto:tom@herbertland.com]
> Sent: Tuesday, May 23, 2017 1:25 PM
> To: Templin, Fred L <Fred.L.Templin@boeing.com>
> Cc: Joe Touch <touch@isi.edu>; int-area@ietf.org
> Subject: Re: [Int-area] IPv6 fragmentation for IPv4
> 
> On Tue, May 23, 2017 at 1:13 PM, Templin, Fred L
> <Fred.L.Templin@boeing.com> wrote:
> > Joe,
> >
> >> -----Original Message-----
> >> From: Int-area [mailto:int-area-bounces@ietf.org] On Behalf Of Templin, Fred L
> >> Sent: Tuesday, May 23, 2017 11:49 AM
> >> To: Joe Touch <touch@isi.edu>; int-area@ietf.org
> >> Subject: Re: [Int-area] IPv6 fragmentation for IPv4
> >>
> >> Hi Joe,
> >>
> >> > -----Original Message-----
> >> > From: Joe Touch [mailto:touch@isi.edu]
> >> > Sent: Tuesday, May 23, 2017 11:01 AM
> >> > To: Templin, Fred L <Fred.L.Templin@boeing.com>; int-area@ietf.org
> >> > Subject: Re: IPv6 fragmentation for IPv4
> >> >
> >> > Hi, Fred (et al.),
> >> >
> >> > On 5/23/2017 9:17 AM, Templin, Fred L wrote:
> >> > > Joe, I wanted to run an idea by you. We all know that IPv4 fragmentation has
> >> > > problems because of the 16-bit ID field. So, why not insert an IPv6 Fragment
> >> > > Header between the IPv4 header and the upper layer protocol data, then
> >> > > use IPv6-style fragmentation instead of IPv4 fragmentation?
> >> >
> >> > IPv4 fragmentation has several impediments:
> >> >     - small ID field
> >> >     - lack of a reassembly checksum
> >> >     - lack of a fixed-location flow ID
> >> >
> >> > Using IPv6-Frag as the next header solves only the first of these. The
> >> > last is significant - putting a new header would defeat IPv4 flow ECMP
> >> > even for the first fragment.
> >>
> >> ECMP gateways could be updated to look at the ULP headers
> >> following the IPv6 Frag header in the first fragment.
> >>
> >> > IPv6 includes a flow field that serves this
> >> > purpose.
> >>
> >> How does it work for plain-old IPv4 fragmentation? I would think
> >> that ECMP gateways would look at the IP ID and try to associate
> >> the fragments so they all get equal ECMP treatment, i.e., the
> >> same as for vanilla IPv4.
> >
> > Here's another think - since the IPv6 Frag Header already has a
> > 32-bit IP ID that we are using for fragmentation, and since we
> > are asking the IPv4 header to set DF=1, the 16-bit IP ID field in
> > the IPv4 header is available for use as a flow field - right?
> >
> Fred,
> 
> I think this would over kill. Assuming fragmentation remains the rare
> case, getting the ECMP hash over the addresses should be sufficient.
> ECMP is a performance optimization, once you're fragmenting that's
> already giving up a lot.

Yes, I don't disagree. But writing a hash of the 5-tuple in the IPv4
ID field of each fragment (and setting DF=1) shouldn't be so hard.

Thanks - Fred

> Tom
> 
> > Thanks - Fred
> >
> >> Thanks - Fred
> >> fred.l.templin@boeing.com
> >>
> >> > Joe
> >>
> >> _______________________________________________
> >> Int-area mailing list
> >> Int-area@ietf.org
> >> https://www.ietf.org/mailman/listinfo/int-area
> >
> >
> > _______________________________________________
> > Int-area mailing list
> > Int-area@ietf.org
> > https://www.ietf.org/mailman/listinfo/int-area