Re: [Softwires] Review of draft-ietf-softwire-multicast-prefix-option-11

[<[e-mail]>]

Dear Roni,

Thank you for the review. 

Please see inline.

Cheers,
Med

> -----Message d'origine-----
> De : Roni Even [mailto:[e-mail]]
> Envoyé : lundi 9 janvier 2017 07:52
> À : [e-mail]
> Cc : [e-mail]; [e-mail]; draft-ietf-softwire-multicast-
> [e-mail]
> Objet : Review of draft-ietf-softwire-multicast-prefix-option-11
> 
> Reviewer: Roni Even
> Review result: Almost Ready
> 
> I am the assigned Gen-ART reviewer for this draft. For background on
> Gen-ART, please see the FAQ at
> <http://wiki.tools.ietf.org/area/gen/trac/wiki/GenArtfaq>.
> Please resolve these comments along with any other Last Call comments
> you may receive.
> Document:  draft-ietf-softwire-multicast-prefix-option-11
> Reviewer: Roni Even
> Review Date:2017-1-9
> IETF LC End Date: 2017–1-12
> IESG Telechat date:
> 
> Summary: This draft is almost  ready for publication as a standard
> track RFC.
> 
> 
> Major issues:
> 
> Minor issues:
> 
> 1.	In section 4 first paragraph say “DHCP servers supporting
> OPTION_V6_PREFIX64 should be configured with U_PREFIX64 and at least
> one multicast PREFIX64 (i.e., ASM_PREFIX64 and/or SSM_PREFIX64).” From
> the rest of the section I understand that for SSM deployments both
> U_PREFIX64 and SSM_PREFIX64 MUST be configured.

[Med] Yes. If you prefer, I can change the text to make this more clear:

OLD:
  DHCP servers supporting OPTION_V6_PREFIX64 should be configured with
   U_PREFIX64 and at least one multicast PREFIX64 (i.e., ASM_PREFIX64
   and/or SSM_PREFIX64).

NEW:
   DHCP servers supporting OPTION_V6_PREFIX64 must be configured with
   ASM_PREFIX64 or SSM_PREFIX64, and may be configured with both.
   U_PREFIX64 must also be configured when SSM_PREFIX64 is provided.
   U_PREFIX64 may be configured when only ASM_PREFIX64 is provided.

> What is the reason for “should” in the first paragraph? Are there
> cases where ASM_PREFIX64 or ASM_PREFIX64 and SSM_PREFIX64 are
> specified and there is no need to specify U_PREFIX64, maybe these
> cases should be described.
> 

[Med] The presence of the unicast address is mandatory for the SSM case because it is required to form an IPv6 address from an IPv4 address to subscribe to a multicast content from a particular source. For the ASM case, the configuration of the U_PREFIX64 is not mandatory in the following cases: (1) a local mapping algorithm is enabled by the function that grafts the IPv4 multicast host side with an IPv6 multicast tree or (2) in deployments that make use of the WKP (64:ff9b::/96, RFC6052).

I can add this NEW text:

   Note that U_PREFIX64 is not mandatory for the ASM case if, for
   example, a local address mapping algorithm is supported or the Well-
   Know Prefix (64:ff9b::/96) is used.

> 
> Nits/editorial comments:
> 1.	RFC2119 keywords in the document are sometime capitalized and
> sometime not. I think it will be good to have consistency and if they
> do not intend to have RFC2119 semantics some other words should be
> used
> 

[Med] I guess you are referring to Section 4. We are not using normative language on purpose because of previous comments we received from some DHC experts (T. Lemon). The use of normative text for the server behavior would mean that we are updating RFC 3315, which we do not want to do. This is why we are defining this section as configuration guidelines.