Re: [Sidrops] I-D Action: draft-ietf-sidrops-8210bis-11.txt

[Job Snijders <[e-mail]>]

Dear Warren,

On Tue, Oct 03, 2023 at 02:59:04AM +0300, Warren Kumari wrote:
> > I disagree with this proposed proceeding.
> >
> > 1) If I recall correctly, two RTR indepenednt implementers suggested the
> > flags field to be zeroized. Yet, the authors of
> > draft-ietf-sidrops-8210bis went another direction. (??!!)
> 
> Ah, I had understood that this option had been agreed to by the WG. If
> this is not in fact the case, then I suspect it will need to go back
> through the WG to have consensus called (hopefully only on this
> point).

Two independent implementers suggested:

'forcing all 16 flag bits to zeri is probably the best course'
https://mailarchive.ietf.org/arch/msg/sidrops/H7dB6GQ5Flly-Dtim5N_7dAeU34/

'It is probably enough to remove / force-to-zero the AFI flags field in
the ASPA PDU.'
https://mailarchive.ietf.org/arch/msg/sidrops/MZhj8HZXj0Tu3rKtCyahF8eFxAA/

I think it is appropriate for this document to come back to the working
group, and before moving forward again, that there is a actual
demonstration of implementations by virtue of reports about those
implementation efforts. I'll do my part in trying to provide running
code. In the previous WGLC for 8210bis zero implementation reports were
shared with this working group. :-(

At this point in time it is not clear to me what relationship exists -
if any - between RTR implementation projects adding support for ASPA,
and the authors of the 8210bis document, but it's not hard to imagine
that a disconnect between the two sides of the coin would make efforts
in this context harder than needed, as we are seeing feedback being
ignored. To overcome such a potential disconnect, I think it would be
reasonable to ask Claudio Jeker and Martin Hoffman to edit the
rfc8210bis document before putting it up for WGLC again.

Kind regards,

Job