Hi Yakov,

If we are to talk real - let's do it.

mLDP is effectively subset of PIM control plane enhanced to carry MPLS
labels over LDP transport. Unicast LDP control plane procedures have
as I said in my former mail in vast majority nothing in common with
mLDP and vice versa. And I am talking about real implementations and
not RFCs.

If you call using completely different protocol and stuffing it under
LDP umbrella as (LDP extension) that's clearly your choice.

But for the SPRING WG point what matters is that when one can use
segment routing to remove need of unicast LDP it does not mean that
segment routing also needs to artificially rename and assimilate PIM
to claim both unicast and multicast coverage.

In IP networks multicast has already number of practical solutions on
the table to control construction of its distribution trees that I do
not think "extensions" of segment routing are required in this space
at least in this point of time.

Perhaps it would be productive to check with others in SPRING WG both
operators and vendors on their opinion in this matter.

Yours,
Robert.


On Fri, Mar 28, 2014 at 2:47 AM, Yakov Rekhter <[email protected]> wrote:
> Robert,
>
>> Hi Robin,
>>
>> Please notice that mLDP and LDP have nothing in common as far as protocol
>> itself. So those are two separate protocols.
>
> To put some reality check on your claim, and especially on the part
> that "mLDP and LDP have nothing in common as far as protocol itself",
> let me point out the following from Section 1 of rfc6388 ("Label
> Distribution Protocol Extensions for Point-to-Multipoint and
> Multipoint-to-Multipoint Label Switched Paths"):
>
>                          This document describes extensions to LDP for
>    setting up point-to-multipoint (P2MP) and multipoint-to-multipoint
>    (MP2MP) LSPs.
>
> Yakov.
>
> P.S. Just to avoid any further confusion, rfc6388 is the specification
> for mLDP.
>
>>
>> The name may be confusing ;)
>
>>
>> Cheers,
>> R.

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

Reply via email to