Hello, Pete,
Thanks a lot for your feedback! In-line....
On 9/9/20 16:39, Pete Resnick via Datatracker wrote:
[....]
Major issues: None
draft-ietf-v6ops-cpe-slaac-renum
Minor issues:
The shepherd writeup says:
The document so far has been approved by the V6OPS working group
(successful working group last call). The document does not specify
new protocol, but rather changes to the default parameters in
existing protocols.
However, the document is Informational, as confirmed by the shepherd writeup.
If this is actually updating default parameters in protocols, that sounds like
it should either be a Standards Track document or more likely a BCP. As 2026
says:
The BCP subseries of the RFC series is designed to be a way to
standardize practices and the results of community deliberations. [...]
...[G]ood user
service requires that the operators and administrators of the
Internet follow some common guidelines for policies and operations.
While these guidelines are generally different in scope and style
from protocol standards, their establishment needs a similar process
for consensus building.
That sounds like what this is doing, especially with all of the 2119 language
in here. Maybe this is Informational because 7084 (and 6204 before it) were
Informational, but perhaps 7084 (and other such document) should be BCP as
well. Indeed, it sounds like all of these SLAAC operational documents could be
in one BCP together.
FWIW, the reason for which this document is informational is because the
document it's formally updating (RFC7084) is also informational. -- Me,
I'd probably agree with you that both RFC7084 and this document should
be BCPs, rather than Informational. I'd like to hear from our AD
regarding how to proceed here.
FWIW, I'm fine with changing the track to BCP, although I'd also note
that there's plenty of existing practice of documents of this type
published as Informational.
Either way, Informational seems wrong.
Nits/editorial comments:
Throughout the document, it says, "This document RECOMMENDS..." or "This
document also RECOMMENDS" or "Additionally, this document RECOMMENDS". RFC 2119
does not use "RECOMMENDS". You can say "CE Routers SHOULD..." or "A Router
Lifetime of ND_PREFERRED_LIMIT is RECOMMENDED" or if you must "It is
RECOMMENDED that..." (blech, I hate the passive form), since SHOULD and
RECOMMENDED are equivalent in 2119, but using the "This document RECOMMENDS..."
form is weird and isn't in 2119.
Fair enough. I'll apply the suggested edit unless I hear otherwise from
others.
In 3.3, it says:
o Upon changes to the advertised prefixes, and after bootstrapping,
the CE router advertising prefix information via SLAAC SHOULD
proceed as follows:
But then each of the things under there has a SHOULD or a MUST. The SHOULD here
is confusing. Instead, the sentence could simply be:
o Upon changes to the advertised prefixes, and after bootstrapping,
the CE router advertising prefix information via SLAAC proceeds as
follows:
Similarly:
This document RECOMMENDS that if a CE Router provides LAN-side DHCPv6
(address assignment or prefix delegation), the following behavior be
implemented:
Just make the sentence:
If a CE Router that provides LAN-side DHCPv6 (address assignment or
prefix delegation), then:
FWIW, the motivation for the "SHOULD" in Section 3.3 is that it
generally implies that the device records prefixes on non-volatile
storage. But there are valid reasons for which a device might be unable
to (e.g., economical, if you wish).
Then, the "MUSTs" elsewhere essentially try to signal how crucial
implementation of each specific behavior is.
Thoughts?
Thanks!
Regards,
--
Fernando Gont
SI6 Networks
e-mail: fgont@xxxxxxxxxxxxxxx
PGP Fingerprint: 6666 31C6 D484 63B2 8FB1 E3C4 AE25 0D55 1D4E 7492
--
last-call mailing list
last-call@xxxxxxxx
https://www.ietf.org/mailman/listinfo/last-call
