Re: [bess] Alvaro Retana's No Objection on draft-ietf-bess-evpn-df-election-framework-07: (with COMMENT)

["Rabadan, Jorge (Nokia - US/Mountain View)" <[e-mail]>]

Hi Alvaro,

Thank you very much. Please see in-line.
Thx
Jorge

From: Alvaro Retana <[e-mail]>
Date: Wednesday, January 16, 2019 at 11:49 PM
To: The IESG <[e-mail]>, "Rabadan, Jorge (Nokia - US/Mountain View)" <[e-mail]>
Cc: "[e-mail]" <[e-mail]>, "[e-mail]" <[e-mail]>, "[e-mail]" <[e-mail]>, Stephane Litkowski <[e-mail]>
Subject: Re: Alvaro Retana's No Objection on draft-ietf-bess-evpn-df-election-framework-07: (with COMMENT)

On January 15, 2019 at 7:49:53 AM, Rabadan, Jorge (Nokia - US/Mountain View) ([e-mail]<mailto:[e-mail]>) wrote:

Jorge:

Hi!


...
(b) Form that text, what I get is that a new algorithm can define the meaning
of the bits. Is that correct? If so, (1) s/a specific DF Alg MAY determine
the use/a specific DF Alg MUST/SHOULD determine the use ...and... (2) for DF
Algs 0 and 1, what happens if any of the other bits are set?
[JORGE] the use of the reserved bits for any future DF Alg is optional, so, a MAY should be fine? I changed the text to:

The use of the bits is optional, yes.  But because each DF Alg may use the bits in a different way, I think that each DF Arg MUST specify how the bits are used.  If there’s no use for the bits, then a future DF Alg should still specify how they are used: ignored when received.
"- In general, a specific DF Alg MAY determine the use of the
reserved bits in the Extended Community, which may be used in a
different way for a different DF Alg. In particular, for DF Algs
0 and 1, the reserved bits are not set by the advertising PE and
SHOULD be ignored by the receiving PE."

Maybe if you changed the “In particular…” text to apply to every DF Alg (unless specifically specified in the future) then I think we could be ok with a MAY above.

[JORGE] ok, I think I understood now what you meant. I changed it to:

“- In general, a specific DF Alg SHOULD determine the use of the reserved bits in the Extended Community, which may be used in a different way for a different DF Alg. In particular, for DF Algs 0 and 1, the reserved bits are not set by the advertising PE and SHOULD be ignored by the receiving PE.”

BTW, why only SHOULD and not MUST?

[JORGE] why a PE SHOULD ignore the received rsv bits? Well, there may be other specs in the future that use DF Algs 0 or 1 and change the rsv bits (in fact there is one already). So I guess we want to leave a window open to that possibility.


(5) §3.2: "if even a single advertisement for the type-4 route is not received
with the locally configured DF Alg and capability, the Default DF Election
algorithm (modulus) algorithm MUST be used" Given that the PEs advertise only
their preferred algorithm/capability, it is possible that it also supports
other algorithm/capability combinations, which may have been advertised. I
assume that it is up to the implementation if they want to update their
advertisement. Do you want to say anything about this? Are there potential
downsides to an implementation changing their preferred combination?
[JORGE] If the advertised ext community is not consistent across all the PEs in the Ethernet Segment, the DF Alg falls back to the default DF Alg. This is added in this section and also in the security section. Besides these two bullets have been modified for clarity and completeness. Let us know if you are not ok with it.
"- Otherwise if even a single advertisement for the type-4 route is received without the locally configured DF Alg and capability, the Default DF Election algorithm (modulus) algorithm MUST be used as in [RFC7432]. This procedure handles the case where participating PEs in the ES disagree about the DF algorithm and capability to apply.
- The absence of the DF Election Extended Community or the presence of multiple DF Election Extended Communities (in the same route) MUST be interpreted by a receiving PE as an indication of the Default DF Election algorithm on the sending PE, that is, DF Alg 0 and no DF Election capabilities."

Yes, that text is fine with me.

The comment I tried to make above is different.  Let me try to give an example — let’s assume that 3 algorithms have been defined: the default and 2 more (A1 and A2).  Let’s assume that all PEs support all the algorithms, and that all, except 1 prefer A1 — the other router (R2) prefers A2.  In this scenario neither A1 nor A2 will be used…  My question above was along the lines of: if R2 prefers A2, then A1 and then the default…then it can advertise A1 once it sees that all other PEs prefer that…resulting in something better than the default.

[JORGE] in the past we discussed something like that but decided to keep it simple to avoid having route type 4 updates coming back and forth and make the DF Election more complicated. So at the moment implementations always fall back to default when there is no unanimity.


(6) The HRW1999 reference must be Normative.
[JORGE] please check out the discussion with Adrian and Satya related to this, where Adrian recommended to move it to informative references.
https://www.ietf.org/mail-archive/web/bess/current/msg03740.html

I don’t think that was Adrian’s intent when he said: "HRW1999 is provided as a normative reference, and from the text I can see why.”

In any case, I think the reference has to be Normative because HRW "must be read to...implement the technology in the new RFC”.

https://www.ietf.org/blog/iesg-statement-normative-and-informative-references/

[JORGE] we can change it as long as it does not create any issues. By the way, sorry, I provided the wrong Adrian’s email. I should have sent this: https://mailarchive.ietf.org/arch/msg/bess/T47tlmuswbj1VFDh5QkA-p8ZMHc
“In general (and I think your draft is an example of this) it is possible to describe/rewrite the pieces of normative text without infringing copyright. That usually reduces the reference to Informative and provides enough information in the RFC for implementation. Your draft is an example of this because you have described the algorithms in your text with enough detail to allow an implementation: the reference is really only there to provide context and proof of the algorithms. (And anyway, having found a freely accessible copy of the reference in your draft, we are probably home and dry.)”

Let us know if you still want us to change it to normative, please.



Thanks!

Alvaro.