Hi Robert

On Tue, Nov 30, 2021 at 12:09 PM Robert Raszuk <[email protected]> wrote:

> Hi Greg,
>
> GIM2>> Now we have to reconcile states reported by RRs.
>>
>
> Not really. That "reconciliation" is native and automatic. No need to do
> anything. If you are hearing updates from two RRs you only consider it DOWN
> when both RRs withdraw the very path. Otherwise it is all UP.
>
> #3 - If my network to PE fails but RR-PE works fine PE will be considered
>>> alive. Network down detection is not the goal for discussing PUA/PULSE.
>>>
>> GIM2>> Thank you for the clarification. Though I wonder how such
>> separation benefits network operation.
>>
>
> Well that can happen if you really use separate data plane for your
> control channels from RRs. Honestly I have not seen much deployments which
> would use such model.
>

   Gyan> My verbiage used in describing this may have been confusing but
what Greg mentioned related to false positive and negative shed some light
on this topic.  So it’s basically the forwarding plane path PE to PE
between any pair of ingress or egress PEs from a design perspective should
never have an RR in the path as the RR hardware is not as robust for high
speed forwarding plane traffic as a PE or P.  Thus the RR from a
topological POV sits to the side in each POP so it’s outside of the direct
PE to PE path forwarding plane for customer traffic.  For this reason the
RR to PE peering path as it’s not the same as the PE to PE path it may
result in false positive or negative.  I am not sure that can be fixed as
that is inherent to any RR design.

>
> Many thx,
> R.
>
> --

<http://www.verizon.com/>

*Gyan Mishra*

*Network Solutions A**rchitect *

*Email [email protected] <[email protected]>*



*M 301 502-1347*
_______________________________________________
Lsr mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/lsr

Reply via email to