Re: draft-ietf-behave-multicast
Tim Chown <tjc <at> ecs.soton.ac.uk>
2007-10-02 08:09:22 GMT
On Mon, Oct 01, 2007 at 01:08:22PM -0400, Ron Bonica wrote:
> Folks,
>
> draft-ietf-behave-multicast is up for IESG review this Thursday. I know
> that a few of you have already reviewed and commented on this document,
> so there shouldn't be many problems.
>
> However, if anyone sees any reason why this document should not
> progress, please email the list ASAP.
Just a handful of relatively minor comments:
1) The draft mentions MLD and MLDv2, yet states that IPv6 is out of scope
(though there are interesting ways to have a proxy proxy between IPv4
and IPv6 groups as per Stig's draft :)
2) The draft appears to focus on IGMP handling by NAT (presumably typically
SOHO devices), so maybe the title of the document should reflect that.
The text states that NATed networks and PIM handling are out of scope so
maybe we should call the document something more specific?
3) The text refers to 'inside segment(s)', does that mean that the
text is in fact covering a case with a NAT device serving multiple
internal subnets? Or is the text referring to forwarding between
internal interfaces purely layer 2 forwarding/bridging?
4) Perhaps clarify more clearly where Reports should be expected to flow
inside->out, and Queries from outside->in.
5) REQ5 mentions 'a protocol'... a bit vague - maybe just say the issue
(Continue reading)