Re: [Roll] Deprecating DCO status

[Alvaro Retana <[e-mail]>]

Ok…if the registry is created by unaware-leaves, and the “no routing entry”
value is added by npdao, we should be ok.

Alvaro.

On October 28, 2020 at 12:22:09 PM, Pascal Thubert (pthubert) (
[e-mail]) wrote:

Hello Rahul

> Thanks, Pascal, Alvaro for the edits.
>
> @Pascal Thubert, Regarding "('E' flag set and 'A' flag not set)", this
may
> confuse the reader and also may not be relevant for this spec since it is
ok for
> an implementer to simply use 0/1 as a Status value without the
understanding
> of E/A flags in unaware-leaves. Can we skip that part since anyways we
are
> referencing unaware-leaves?

RFC 6550 specifies that values >= 128 are rejection. The 'E' flag is
backward compatible. The reason why I mention it is that DCO has a
normative ref to unuaware leaves, so it needs to implement the new RPL
Status, meaning that it needs to indicates the setting of the 'E' and 'A'
flags for each status it defines.

My question to you is whether DCO status code 1 was a rejection or not. If
it was a rejection, then the RPL Status value per RFC 6550 should have been
129; in the new format it is 1 again, but with the 'E' flag set. This si
what
https://tools.ietf.org/html/draft-ietf-roll-unaware-leaves-22#section-12.5
assumes.

If it is a warning, then a value of 1 is OK but still I'd mention the E and
A flags for completeness, both set to 0.
If so then the IANA declaration in unaware-leaves has to change.

I tend to think that I should remove

| 1 | No routing-entry for the | [EFFICIENT-NPDAO] |
| | indicated Target found |

>From the initialization of the IANA section and let you define it as a new
entry in the registry in DCO.

What do you all think?

Pascal