headers
David Meyer | 7 Jan 2004 21:18

mboned: draft-ietf-mboned-ipv4-mcast-unusable-01.txt


	Folks,

	Pekka had some fairly detailed comments on this one. I'm
	looking for some discussion on whether we need to
	reconsider the approach taken in the draft (noting that
	his was the only comment and the WG last call will end on
	01.09.2004).

	Thanks,

	Dave
_______________________________________________________________
user interface: http://darkwing.uoregon.edu/~llynch/mboned.html
web archive:  http://darkwing.uoregon.edu/~llynch/mboned/
Permalink | Reply | 0 comments |
headers
Bill Nickless | 7 Jan 2004 22:17
Favicon

Re: mboned: I-D ACTION:draft-ietf-mboned-ipv4-mcast-unusable-01.txt]

Good afternoon Pekka,

I thought about your comments over the holiday break, and here's what I 
came up with:

Operationally, I think there are at least three places where access control 
lists should (by best practice) be installed: in the hosts/customer 
routers, at the edge of the AS (for PIM), and between ASes (for 
MSDP).  Using the "belt and suspenders" paranoid approach, I would suggest 
that the elements in each of these three access controls would be cumulative.

In other words:

The lines in Section 6 that refer to Section 7 should be installed on hosts 
(probably Ethernet kernel drivers) and on host-facing routers.

The lines in Section 6 that refer to Sections 7 and 8 should be installed 
on PIM routers that face other administrative domains.

The lines in Section 6 that refer to Sections 7, 8, and 9 should be 
installed on MSDP-speaking routers that peer with other administrative domains.

At first glance, I don't think this document is the place to include the 
above explanation; instead, I think that's better put in a BCP.

I was working on this document first because it was holding up some other 
RFCs, but expect to go through the IPv4 BCP shortly as well to include 
things like the above.

Does this make sense?  Should I go ahead and make the semi-editorial
(Continue reading)

Permalink | Reply | 6 comments |
headers
David Meyer | 7 Jan 2004 22:23

mboned: draft-ietf-mboned-ipv4-mcast-unusable-01.txt

	Folks,

	First, sorry if this turns out to be a duplicate (my last
	posting seems to have disappeared). In any event, I had
	scheduled the WG last call on this document to end on
	01.09.2004. However, there was one set of comments from
	Pekka regarding, among other things, the structure of the
	document (see http://darkwing.uoregon.edu/~llynch/mboned/msg02187.html).
	What I'm looking for is some discussion on where we want
	to go with this one. We do have one draft that is hung on
	a normative reference to this (draft-ietf-mboned-msdp-deploy-04.txt), 
	so I would like to resolve this one.

	Comments to the list please.

	Thanks,

	Dave

	
_______________________________________________________________
user interface: http://darkwing.uoregon.edu/~llynch/mboned.html
web archive:  http://darkwing.uoregon.edu/~llynch/mboned/
Permalink | Reply | 0 comments |
headers
David Meyer | 7 Jan 2004 23:10

mboned: Please comment (in the event you didn't see this)

	One note: I have asked the secretariat to update the
	mailing list information. 

	Dave

----- Forwarded message from IESG Secretary <iesg-secretary <at> ietf.org> -----

>> From: IESG Secretary <iesg-secretary <at> ietf.org>
>> To: iesg <at> ietf.org, <iab <at> iab.org>, <dmm <at> 1-4-5.net>, <sob <at> harvard.edu>
>> Subject: Internal WG Review: Recharter of MBONE Deployment (mboned)
>> Date: Fri, 2 Jan 2004 16:10:45 -0500 (EST)
>> 
>> 
>> A new charter for the MBONE Deployment working group (MBONED)
>> in the Operations and Management Area of the IETF is being considered.
>> The draft charter is provided below for your review and comment.
>> 
>> Review time is one week.
>> 
>> The IETF Secretariat
>> 
>> MBONE Deployment (mboned)
>> --------------------------------------------------
>> Note: The charter for MBONED is included below.
>> However, to assist you in your review, the changes from the previous
>> version as presented to the Secretariat by Bert Wijnen are as follows:
>> 1. The Goals and Milestones have been updated.
>> 2. Three additional activities have been added to the list. they are:
>> 
>>    - Update RFC 3171/BCP 51 based on experience.
(Continue reading)

Permalink | Reply | 14 comments |
headers
Toerless Eckert | 8 Jan 2004 09:41
Picon
Favicon

Re: mboned: Please comment (in the event you didn't see this)

While MSDP MIB and rfc3171 are IPv4 only, i think the IMDOC effort
must of course also include IPv6, but given that rfc1112 does only
address IPv4, it would be good to add something like "for both IPv4
and IPv6" to the end of the roadmap RFC goal statement.

On Wed, Jan 07, 2004 at 02:10:09PM -0800, David Meyer wrote:
> 	One note: I have asked the secretariat to update the
> 	mailing list information. 
> 
> 	Dave
> 
> ----- Forwarded message from IESG Secretary <iesg-secretary <at> ietf.org> -----
> 
> >> From: IESG Secretary <iesg-secretary <at> ietf.org>
> >> To: iesg <at> ietf.org, <iab <at> iab.org>, <dmm <at> 1-4-5.net>, <sob <at> harvard.edu>
> >> Subject: Internal WG Review: Recharter of MBONE Deployment (mboned)
> >> Date: Fri, 2 Jan 2004 16:10:45 -0500 (EST)
> >> 
> >> 
> >> A new charter for the MBONE Deployment working group (MBONED)
> >> in the Operations and Management Area of the IETF is being considered.
> >> The draft charter is provided below for your review and comment.
> >> 
> >> Review time is one week.
> >> 
> >> The IETF Secretariat
> >> 
> >> MBONE Deployment (mboned)
> >> --------------------------------------------------
> >> Note: The charter for MBONED is included below.
(Continue reading)

Permalink | Reply | 11 comments |
headers
David Meyer | 8 Jan 2004 16:31

Re: mboned: Please comment (in the event you didn't see this)

	Toreless,

On Thu, Jan 08, 2004 at 12:41:43AM -0800, Toerless Eckert wrote:
>> While MSDP MIB and rfc3171 are IPv4 only,

	(i).	RFC 3307 (Allocation Guidelines for IPv6
		Multicast Addresses), is the analog to 3171 for
		IPv6. However, as you know it's a completely
		different problem.  

>> i think the IMDOC effort must of course also include IPv6, but
>> given that rfc1112 does address IPv4, it would be good to add
>> something like "for both IPv4 and IPv6" to the end of the
>> roadmap RFC goal statement. 

	(ii).	The goal/milestone corresponding to what we were
		calling, for lack of a better name, IMDOC, is

		- Develop a roadmap informational RFC that
		  describes the current IETF multicast
		  architecture, including references to the
		  relevent IETF documents and guidance for
		  implementers and network operators along the
		  line of RFC 1122/1123. 

		Scott helped me craft this goal/milestone
		language, and the reference to 1123/1123 didn't
		imply IPv4 or IPv6. Its just a reference. Scott,
		do we need to rev this to call out v6 (since
		neither is mentioned, I assumed it would be clear
(Continue reading)

Permalink | Reply | 10 comments |
headers
Toerless Eckert | 8 Jan 2004 16:44
Picon
Favicon

Re: mboned: Please comment (in the event you didn't see this)

On Thu, Jan 08, 2004 at 07:31:58AM -0800, David Meyer wrote:
> On Thu, Jan 08, 2004 at 12:41:43AM -0800, Toerless Eckert wrote:
> >> While MSDP MIB and rfc3171 are IPv4 only,
> 
> 	(i).	RFC 3307 (Allocation Guidelines for IPv6
> 		Multicast Addresses), is the analog to 3171 for
> 		IPv6. However, as you know it's a completely
> 		different problem.  

Right, eg: the two new work items MSDP MIB and rfc3171 nly
apply to IPv4 and the proposed WG work item text is ok and non-confusing.

> 	(ii).	The goal/milestone corresponding to what we were
> 		calling, for lack of a better name, IMDOC, is
> 
> 		- Develop a roadmap informational RFC that
> 		  describes the current IETF multicast
> 		  architecture, including references to the
> 		  relevent IETF documents and guidance for
> 		  implementers and network operators along the
> 		  line of RFC 1122/1123. 
> 
> 		Scott helped me craft this goal/milestone
> 		language, and the reference to 1123/1123 didn't
> 		imply IPv4 or IPv6. Its just a reference. Scott,
> 		do we need to rev this to call out v6 (since
> 		neither is mentioned, I assumed it would be clear
> 		that it was both)?

Well, _i_ was assuming that IMDOC is to cover both IPv4 and IPv6. If
(Continue reading)

Permalink | Reply | 1 comment |
headers
David Meyer | 8 Jan 2004 16:47
Picon
Favicon

Re: mboned: Please comment (in the event you didn't see this)

On Thu, Jan 08, 2004 at 07:44:14AM -0800, Toerless Eckert wrote:
>> On Thu, Jan 08, 2004 at 07:31:58AM -0800, David Meyer wrote:
>> > On Thu, Jan 08, 2004 at 12:41:43AM -0800, Toerless Eckert wrote:
>> > >> While MSDP MIB and rfc3171 are IPv4 only,
>> > 
>> > 	(i).	RFC 3307 (Allocation Guidelines for IPv6
>> > 		Multicast Addresses), is the analog to 3171 for
>> > 		IPv6. However, as you know it's a completely
>> > 		different problem.  
>> 
>> Right, eg: the two new work items MSDP MIB and rfc3171 nly
>> apply to IPv4 and the proposed WG work item text is ok and non-confusing.
>> 
>> > 	(ii).	The goal/milestone corresponding to what we were
>> > 		calling, for lack of a better name, IMDOC, is
>> > 
>> > 		- Develop a roadmap informational RFC that
>> > 		  describes the current IETF multicast
>> > 		  architecture, including references to the
>> > 		  relevent IETF documents and guidance for
>> > 		  implementers and network operators along the
>> > 		  line of RFC 1122/1123. 
>> > 
>> > 		Scott helped me craft this goal/milestone
>> > 		language, and the reference to 1123/1123 didn't
>> > 		imply IPv4 or IPv6. Its just a reference. Scott,
>> > 		do we need to rev this to call out v6 (since
>> > 		neither is mentioned, I assumed it would be clear
>> > 		that it was both)?
>>
(Continue reading)

Permalink | Reply | 0 comments |
headers
Scott Bradner | 8 Jan 2004 16:49
Picon
Favicon

Re: mboned: Please comment (in the event you didn't see this)

                Scott helped me craft this goal/milestone
                language, and the reference to 1123/1123 didn't
                imply IPv4 or IPv6. Its just a reference. Scott,
                do we need to rev this to call out v6 (since
                neither is mentioned, I assumed it would be clear
                that it was both)?

I'd suggest calling it out - assumptions of v6-ness are often wrong :-)

Scott
_______________________________________________________________
user interface: http://darkwing.uoregon.edu/~llynch/mboned.html
web archive:  http://darkwing.uoregon.edu/~llynch/mboned/
Permalink | Reply | 7 comments |
headers
Scott Bradner | 8 Jan 2004 17:14
Picon
Favicon

Re: mboned: Please comment (in the event you didn't see this)

- Develop a roadmap informational RFC that describes the current
  IPv4 and IPv6 IETF multicast architectures, including
  references to the relevent IETF documents and guidance for
  implementers and network operators. 

that looks good to me

Scott
_______________________________________________________________
user interface: http://darkwing.uoregon.edu/~llynch/mboned.html
web archive:  http://darkwing.uoregon.edu/~llynch/mboned/
Permalink | Reply | 1 comment |

Gmane