Re: [mpls] applicability of draft-wijnands-mpls-mldp-in-band-signaling-02

[Yakov Rekhter <yakov at juniper.net>]

Ice,

> > At the last IETF I asked the authors to explicitly spell out the
> > applicability of this draft to various "flavors" of PIM.
> >
> > The latest version of the draft (-02) still does not address this.
> > To address this I propose to insert the following in the Introduction
> > section of the draft:
> >
> >    The procedures specified in this draft are sufficient to support
> >    IP multicast with PIM-SM in SSM mode. The procedures specified
> 
> That is too restrictive.
> 
> >    in this draft are insufficient to support IP multicast with
> >    PIM-SM in ASM mode for the reason mentioned in section 2.3 of
> 
> This prevents receiving a (S,G) from the P2MP LSP and send it down the  
> (*,G) PIM tree. This is the behavior we want to support in combination  
> with out-of-band source discovery via MSDP. The only thing we want to  
> restrict is transport (*,G) traffic down the P2MP LSP, that is covered  
> by section 2.3.

Then the draft should explicitly spell out "the behavior we want
to support", *and* the mechanisms/procedures to support such behavior,
including the interaction with the out-of-band source discovery via MSDP.

Yakov.