Hi Wes,

(FWIW, I agree with Kathleen's DISCUSS on this but one question
below before I post my own ballot...)

On 02/05/16 20:31, George, Wes wrote:
> 
> On 5/2/16, 1:04 PM, "Kathleen Moriarty" <[email protected]>
> wrote:
> 
> 
>>
>> ----------------------------------------------------------------------
>> DISCUSS:
>> ----------------------------------------------------------------------
>>
>> 1.  Why is this document preceding the BGP spec?  Shouldn't this be part
>> of the BGPSec protocol document?  If BGPSec isn't getting deployed
>> because of the AS path migration problem and this gets us a little
>> further, but not quite as secure, maybe that's a trade off we need to
>> accept.  But this document coming through first is a little concerning
>> even though the protocol spec is a normative reference.
> 
> WG] This is largely a timing problem. The protocol draft was written
> first, but after this issue was identified, AS-migration was written, and
> both drafts were revised and reviewed in parallel.
> The WG discussed integrating AS-migration into the protocol spec, and
> decided to keep it separate. At the time, this was due to several
> dependencies and assumptions:
> First, the WG consensus between SIDR, GROW, and RTGWG was that BGPSec
> definitely needed to allow for AS-Migration, but before BGPSec could be
> expected to support AS-Migration, those tools needed to be documented as a
> formal part of the BGP protocol (in RTGWG), rather than just being de
> facto standard on account of their being implemented by all major router
> vendors and widely used by ISPs. That resulted in RFC 7705 (aka
> idr-as-migration), also written largely in parallel with sidr-as-migration
> (in fact they started as one document and were split).
> Second, the BGPSec protocol document was thought to be much closer to
> completion and publishing than either of these documents, and was expected
> to be completed well prior to sidr-as-migration such that it didn't make
> much sense to delay the protocol doc to integrate them.
> In the meantime, the protocol document underwent additional reviews that
> uncovered issues requiring more significant revision and delayed its
> progress such that the as-migration document is now ready first. The
> protocol document has had edits to ensure proper integration between the
> two documents where appropriate (adding clarity via references and
> language to ensure no conflict in what is being directed in each standard).
> Last, part of the rationale for keeping them separate is that while the WG
> strongly recommends that both parts be implemented, BGPSec does not
> strictly require the additional support for AS-migration (I.e.
> Implementing the functionality defined in BGPSec protocol document is a
> MUST, AS-migration is a SHOULD), and thus having the documents be separate
> makes a clear distinction between basic protocol function and add-on
> features.

This is only a timing problem if bgpsec doesn't change in some
incompatible manner. If such a change happens then this is more
than a timing issue.

What'd be bad about just holding this in the WG until bgpsec is
ready? Since there's a normative reference anyway getting the
IESG to approve both at once seems like it'd be better.

Cheers,
S.

> 
>>
>>
>> 2.  The introduction makes this sound rather innocuous, but the security
>> considerations section is more explicit that this is a work around BGPSec
>> and isn't quite as secure.  I'd like to see some text explaining this
>> better in the introduction, more similar to what's in the first paragraph
>> of the security considerations section.
> 
> WG] The intent was not to imply that BGPSec with AS-Migration support is
> fundamentally less secure than BGPSec without it, so if that's what we're
> conveying between the two sections, I'd appreciate additional guidance on
> what specifically is giving you that impression. It's not really a
> work-around BGPSec as much as it's observing that the BGPSec protocol as
> written would break this common method for AS-Migration, and this document
> addresses that gap by ensuring that it can be done within the security
> framework provided by BGPSec.
> 
> 
> Thanks,
> 
> Wes George
> 
> 
> Anything below this line has been added by my company’s mail server, I
> have no control over it.
> -----------
> 
> 
> 
> 
> ________________________________
> 
> This E-mail and any of its attachments may contain Time Warner Cable 
> proprietary information, which is privileged, confidential, or subject to 
> copyright belonging to Time Warner Cable. This E-mail is intended solely for 
> the use of the individual or entity to which it is addressed. If you are not 
> the intended recipient of this E-mail, you are hereby notified that any 
> dissemination, distribution, copying, or action taken in relation to the 
> contents of and attachments to this E-mail is strictly prohibited and may be 
> unlawful. If you have received this E-mail in error, please notify the sender 
> immediately and permanently delete the original and any copy of this E-mail 
> and any printout.
> 

Attachment: smime.p7s
Description: S/MIME Cryptographic Signature

_______________________________________________
sidr mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/sidr

Reply via email to