Re: Turning routers into hosts (Re: Death by extension header (was:RE: New Version Notification for draft-li-6man-hbh-fwd-hdr-00.txt))

On Mon, Jul 13, 2020 at 09:59:03PM -0400, Warren Kumari wrote:
> ... are you sure you don't mean "Turning Routers into Smoking Holes in the
> Ground"
> 
> Regards,
>   An operator of a device that tried to punt packets to the slow path.

Where were you when we deployed in ca. 1989 version 1.0 of DVMRP across the
Internet which was using a loose source route to a remote tunnel endpoint
to avoid encapsulation overhead. And all the routers along that internet
path happily punted all those loose source route packets. I think it took
a few months before operators recognized the problem though...

If smoking out routers with slow path is still a thing,
i guess there must be some redeeming gratification in it.  ;-))

Cheers
    Toerless

> > Regards,
> >    An author of RFC3234
> 
> 
> > >
> > > because in-flight processing of packets' EHs, payloads etc, rather
> > > than just dumb, fast and simple forwarding them, is host processing of
> > > packets.
> > >
> > > How to tell it's host processing?
> > >
> > > - processing of packets requires going past the IPv6 fixed header
> > >
> > > - in-flight processing would fail if everything but the fixed header
> > > was encrypted
> > >
> > > - processing beyond the fixed (forwarding) header is occuring at a
> > > node that *doesn't* hold the packet's destination address:
> > >
> > >
> > > RFC8200:
> > >
> > > router       a node that forwards IPv6 packets not explicitly
> > >                 addressed to itself.  (See Note below.)
> > >
> > >    host         any node that is not a router.  (See Note below.)
> > >
> > > (Note is about a node's interfaces being forwarding or not)
> > >
> > >
> > > In this discussion and similar, people are really discussing paths
> > > between a packet's original source and a final destination constructed
> > > of a set of packet host processing hops.
> > >
> > >
> > > Regards,
> > > Mark.
> > >
> > >
> > >
> > >> Cheers
> > >>     Toerless
> > >>
> > >> --------------------------------------------------------------------
> > >> IETF IPv6 working group mailing list
> > >> ipv6@ietf.org
> > >> Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6
> > >> --------------------------------------------------------------------
> > >
> > > --------------------------------------------------------------------
> > > IETF IPv6 working group mailing list
> > > ipv6@ietf.org
> > > Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6
> > > --------------------------------------------------------------------
> > >
> >
> > --------------------------------------------------------------------
> > IETF IPv6 working group mailing list
> > ipv6@ietf.org
> > Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6
> > --------------------------------------------------------------------
> >
> 
> 
> -- 
> I don't think the execution is relevant when it was obviously a bad idea in
> the first place.
> This is like putting rabid weasels in your pants, and later expressing
> regret at having chosen those particular rabid weasels and that pair of
> pants.
>    ---maf

-- 
---
tte@cs.fau.de