Hi Pushpasis,
Please review attached file that depicts high level difference as it applies to
dual plane use case. Currently, there is no solution available to support
optimized loosed path per plane with no fallback to alternative plane if the
plane is partitioned.
Thanks
Arkadiy
From: Pushpas
Hi Shraddha,
On Tue, Mar 14, 2017 at 2:19 PM, Shraddha Hegde
wrote:
> Hi Stephano/Pushpasis,
>
> Anycast segments provide loose separation of routing planes. If there is a
> failure, traffic running over anycast segment is allowed
> To failover to different plane. The requirement, this draft tri
> From: John G. Scudder [mailto:j...@juniper.net] > Sent: Monday, March 13,
> 2017 9:54 PM
>
> Thanks, Stefano.
>
> Bruno, at your convenience can you confirm that you're satisfied with the
> resolution?
-11 addresses my comments.
Thank you John, Stefano.
> Looks
> OK to me even though
Hi Stephano/Pushpasis,
Anycast segments provide loose separation of routing planes. If there is a
failure, traffic running over anycast segment is allowed
To failover to different plane. The requirement, this draft tries to address is
the strict routing plane separation. Certain application traf
Hi Pushpasis,
I agree. The problem/use-case is already described in RFC7855, the required
protocol extensions are already documented in ospf, isis and bgp drafts, we
already have multiple implementations, and deployments have been done.
s.
> On Mar 14, 2017, at 8:20 AM, Pushpasis Sarkar
> w
Hi Authors,
First I must admit that I have not read the entire draft in details...
But from the abstract it seems that for the problem that this draft is
trying to address, a similar problem is already addressed in the Segment
Routing Problem Statement and Use-Case document (RFC 7855, section 3.3