Hi Robert,

In-line [Uma]:
--
Uma C.

From: [email protected] [mailto:[email protected]] On Behalf Of Robert Raszuk
Sent: Wednesday, January 25, 2017 11:40 AM
To: Uma Chunduri <[email protected]>
Cc: RTGWG <[email protected]>
Subject: Re: Routing in DC RTGWG interim - updated

/* Trimming the lists a bit */

Hi Uma,

Well I would actually start with common understanding and agreement as to 
what's wrong with RFC7938 based deployments in the first place ? Is there some 
document or draft which describes it well ?
[Uma]: Great question.  Am not aware of any and was asking this question to 
myself back when I was reviewing/going-through one of the proposals..
               But AFAIS an inclination to use TE in the fabric with a 
controller and hence you need a full topology view at your disposal. Also link 
(you have massive ECMP any ways in the CLOS) and node protection?? May be more….

Is BGP local auto discovery an issue ? Then let's solve it .. it is simple 
problem. Some already solved it with auto BGP peering to link local addresses.
[Uma]: Possible.
               I am not sure this is the central issue ..but for SPF based 
approaches to have the topology view this could be one part of it (and hence in 
bgp-spf,  usage of BGP-LS to advertise the TCP peering as a Link NLRI with new 
TLVs distinguishing itself from IGP adj representation).
             Again, am not fully sure how auto discovery would be critical if 
it is EBGP  single-hop peering sessions.


To perhaps restate Keyur's question ... Do we all agree with the requirements 
as presented by Tony P which are triggering RIFT ?
[Uma]: Have few questions .. but some other day.


[Inline image 1]

Cheers,
R.



On Wed, Jan 25, 2017 at 8:26 PM, Uma Chunduri 
<[email protected]<mailto:[email protected]>> wrote:
One quick suggestion for the solution presenters:

Perhaps it’s good to have a section in respective documents how the proposal 
can bring operational simplicity (if at all)  than existing deployments (say 
than RFC 7938 based..).
--
Uma C.


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

Reply via email to