At 5:01 PM -0400 10/2/02, Rob Austein wrote:
>I made the mistake of allowing my arm to be twisted into reviewing
>draft-ietf-ipngwg-addr-arch-v3-10.txt last week, and was sad to find
>what appears to be an ambiguity in some of text that deals with
>subnet-scope multicast.  Given that this document was already before
>the IESG at the time I found this, I've been discussing this with our
>AD, who brought in our WG chairs once he and I agreed that there might
>be a problem here, but we felt that the discussion of what to do about
>this really should take place out in the open on the WG mailing list.

As part of the AD/chair discussion, I responded to Thomas's report of
the issue as follows:

>To: Thomas Narten <[EMAIL PROTECTED]>
>From: Steve Deering <[EMAIL PROTECTED]>
>Subject: Re: IPv6 subnet-local addresses and draft-ietf-ipngwg-addr-arch-v3-10.txt
>Cc: Bob Hinden <[EMAIL PROTECTED]>, Margaret Wasserman <[EMAIL PROTECTED]>, Rob 
>Austein <[EMAIL PROTECTED]>, Erik Nordmark <[EMAIL PROTECTED]>
>
>At 10:43 AM -0400 10/2/02, Thomas Narten wrote:
>>One last (??) issue on this document has popped up. The issue as I
>>understand it is that with the addition of subnet-local multicast
>>scope, the document leaves out details about how routers are supposed
>>to know what the actual subnet boundaries are.
>
>Thomas,
>
>Every router (whether IPv4 or IPv6) knows what subnets its own interfaces
>belong to (or, more accurately, what subnet numbers are assigned to
>the links to which it has interfaces).  That is the most basic
>configuration info provided to a router -- it is provided with that info
>in order to do unicast routing and forwarding.  To enforce subnet-local
>scope it is necessary simply to inhibit the forwarding of subnet-local-
>destined packets between interfaces that do not belong to the same subnet.
>I would have thought that would be obvious.  For those for whom it is not
>obvious, there is additional detail on the default configuration of
>scope zone boundaries in the scoped address architecture spec, along
>with lots of other info required to implement address scoping.
>
>More comments in-line below...
>
>>Rob Austein <[EMAIL PROTECTED]> writes:
>>
>> > ...  Excerpting from the draft:
>>
>> >              2  link-local scope
>> >              3  subnet-local scope
>> >              4  admin-local scope
>>
>> >         ...
>>
>> >              link-local and site-local multicast scopes span the same
>> >              topological regions as the corresponding unicast scopes.
>>
>> >              subnet-local scope is given a different and larger value
>> >              than link-local to enable possible support for subnets
>> >              that span multiple links.
>>
>> >              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.
>
>Note that the determination of the span of subnet-scope is an example of
>"automatic derivation from...other, non-multicast related configuration",
>as mentioned in the description of admin-local.
>
>>Here is a suggestion:
>>
>>1) change the wording of the subnet-local definition to say something
>>   like:
>>
>>              subnet-local scope is given a different and larger value
>>              than link-local to enable possible support for subnets
>>              that span multiple links. By default, routers assume
>>              that subnet scope and link-local scope are equivalent.
>
>I think that such a change is unwarranted if it will mean even
>more delay in the approval and publication of the spec.  If you can
>handle it as a Note to the RFC Editor or something like that, then
>fine.  However, I have a few problems with your added sentence
>above: 
>
>    - it's odd to stick that little implementation note there in the
>      middle of the scope descriptions
>
>    - it should refer to nodes, not just routers
>
>    - your statement would not necessarily be true for routers that do
>      support multi-link subnets -- for the them, the default might be
>      *not* to assume that subnet-local and link-local scope are
>      equivalent.
>
>Here's an alternative to your sentence which bypasses those problems:
>
>         In the normal case of a subnet confined to a single link,
>         subnet-scope is equivalent to link-scope.
>
>>2) to the admin-local scope, tweak the wording to say something like:
>>
>>              admin-local and all larger scopes must be
>>              administratively configured, i.e., they are not
>>              automatically derived from physical connectivity or
>>              other, non-multicast-related configuration.
>>
>>Make sense?
>
>I don't object to that changed wording, but neither do I see the
>necessity of it.
>
>Steve


In a response to that message, Rob asked me if I had forgotten about
unnumbered point-to-point links.  I answered as follows:

>Yes, I did forget about them, but I think it's obvious how to handle them:
>they are not part of a subnet that exists on any other link, so subnet-
>scope multicasts would not be forwarded to or from an unnumbered link.

Steve

--------------------------------------------------------------------
IETF IPng Working Group Mailing List
IPng Home Page:                      http://playground.sun.com/ipng
FTP archive:                      ftp://playground.sun.com/pub/ipng
Direct all administrative requests to [EMAIL PROTECTED]
--------------------------------------------------------------------

Reply via email to