Re: [Mpls-interop] MPLS over MPLS-TP

["Drake, John E" <[e-mail]>]

 

>-----Original Message-----
>From: Sprecher, Nurit (NSN - IL/Hod HaSharon) 
>[mailto:[e-mail]] 
>Sent: Thursday, December 11, 2008 8:55 AM
>To: ext Ben Niven-Jenkins; [e-mail]
>Subject: Re: [Mpls-interop] MPLS over MPLS-TP
>
>The first option also allows you to have better trouble 
>shooting in case of a problem (thanks to the clear separation).

JD:  Regardless of whether you have MPLS TP/PW/MPLS *or* MPLS TP/MPLS,
you will have two label stacks.  The only difference is whether the last
label in the provider's stack is an PW label *or* an MPLS label.  This
was the point in Andy's first e-mail, which as I read it, said that
there was nothing magic about that label being a PW label.
 
>
>-----Original Message-----
>From: [e-mail]
>[mailto:[e-mail]] On Behalf Of ext Ben 
>Niven-Jenkins
>Sent: Thursday, December 11, 2008 00:41
>To: [e-mail]
>Subject: [Mpls-interop] MPLS over MPLS-TP
>
>Colleagues,
>
>What's the current view of how to support MPLS over MPLS-TP?
>
>Do we put a PW in the middle (like draft-bryant) or do we just 
>use the label stack and have MPLS directly over MPLS-TP?
>
>Advantage of the first is that it gives layer separation (e.g. 
>If the two layers are operated by different parties), 
>advantage of the second is it's probably a bit simpler.
>
>Do the drafts we have say anything about the MPLS over MPLS-TP case?
>Same
>for IP over MPLS-TP?
>
>Thanks
>Ben
>
>_______________________________________________
>Mpls-interop mailing list
>[e-mail]
>https://www.ietf.org/mailman/listinfo/mpls-interop
>_______________________________________________
>Mpls-interop mailing list
>[e-mail]
>https://www.ietf.org/mailman/listinfo/mpls-interop
>
_______________________________________________
Mpls-interop mailing list
[e-mail]
https://www.ietf.org/mailman/listinfo/mpls-interop