Re: [Roll] draft-ietf-6man-multicast-scopes updates RFC 4007 (Was Re: [roll] #132: draft-ietf-roll-trickle-mcast-04 - Clarify scope value of 3 - subnet-local)

[Ralph Droms <[e-mail]>]

On Oct 21, 2013, at 3:00 PM 10/21/13, 神明達哉 <[e-mail]> wrote:

> At Thu, 17 Oct 2013 09:10:50 -0700,
> Ralph Droms <[e-mail]> wrote:
> 
>> Noting this conflict, I propose adding a bit of text to
>> draft-ietf-6man-multicast-scopes to update RFC 4007 for consistency
>> with RFC 4291:
> 
> I think making these consistent is a good idea.
> 
>> OLD:
>> 
>>   o  The boundaries of zones of a scope other than interface-local,
>>      link-local, and global must be defined and configured by network
>>      administrators.
>> 
>> NEW:
>> 
>>   o  Admin-Local scope is the smallest scope that must be
>>      administratively configured, i.e., not automatically derived
>>      from physical connectivity or other, non-multicast-related
>>      configuration.
> 
> I'm okay with the proposed text, but on a closer look I've noticed
> a couple of subtle points you may want to consider:
> 
> - I see a slight difference between the OLD (RFC 4007) and NEW (RFC
>  4291) even after fixing the obvious inconsistency, in that the NEW
>  version does not necessarily say how the scopes larger than
>  admin-local scope is configured; technically it only states
>  interface-local and link-local (and "Realm-Local" if assigned)
>  would not be automatically configured.  In fact, the global scope
>  shouldn't be "administratively" configured, which is crystal clear
>  from the OLD version, but not necessarily in the NEW version.
> 
> - should we worry about possible future updates to the addressing
>  architecture and scope architecture documents?  If we keep these
>  documents separately and have a copy of the text in both, we may need
>  to update both copies when we need to make a change on this point as
>  more "currently reserved" scopes are assigned.

Both good points.  As long as we have to make an update to RFC 4007, let's make sure the update is correct.

> 
> These are probably too minor and maybe we can simply ignore the
> subtlety.  If that's our consensus I'm okay with that.  But if we want
> to address these points, maybe:
> 
> OLD:
> 
>   o  The boundaries of zones of a scope other than interface-local,
>      link-local, and global must be defined and configured by network
>      administrators.
> 
> NEW:
> 
>   o  The boundaries of zones of an interface-local, link-local, or
>      global scope automatically derived from physical connectivity.
>      For zones of other types of scopes, the IPv6 addressing
>      architecture specification defines how their boundaries must be
>      determined, whether automatically derived or administratively
>      configured.

Perhaps we want to go a step farther and take the zone boundary text out of RFC 4007 altogether?

OLD:

  o  The boundaries of zones of a scope other than interface-local,
     link-local, and global must be defined and configured by network
     administrators

NEW:

  o  The boundaries of zones of a scope are defined by the IPv6
     addressing architecture.

- Ralph


> 
> --
> JINMEI, Tatuya