Re: Fwd: New Version Notification for draft-hinden-6man-hbh-processing-01.txt

[Fernando Gont <[e-mail]>]

Hi,

On Thu, 2021-06-10 at 16:15 -0700, Tom Herbert wrote:
> On Thu, Jun 10, 2021 at 4:06 PM Fernando Gont <
> [e-mail]> wrote:
> > On Thu, 2021-06-10 at 15:47 -0700, Tom Herbert wrote:
> > [....]
> > > 
> > 
> > If the feature cannot be relied upon, a fallback mechnism needs to
> > be
> > implemented. Now, as a protocol designer, why would I even bother
> > implementing functionality with an unreliable mechanism,
> > particularly
> > when I could simply rely on something else?
> > 
> > 
> > OTOH, if I know that you have a fallback mechanism to take care of
> > situations where packets with EHs are dropped, why should I allow
> > you
> > EH packets in my network in the first place?
> > 
> Because fallbacks offer degraded service not equivalent service. For
> instance your question could be restated as: given that IPv4 is the
> fallback for IPv6, why should you allow IPv6 packets in your network
> in the first place?

If IPv6 had brokenness associated with it's deseign that couldn't be
worked around, then yes, people wouldn't bother trying IPv6 before
IPv4.


-- 
Fernando Gont
Director of Information Security
EdgeUno, Inc.
PGP Fingerprint: DFBD 63E3 B248 AE79 C598 AF23 EBAE DA03 0644 1531