The MPLS WG Archive

[Seisho Yasukawa <yasukawa.seisho@lab.ntt.co.jp>]

Hi MPLS List,

As promised, this is the first of several emails to resolve the remaining 
open issues in 
http://www.ietf.org/internet-drafts/draft-ietf-mpls-p2mp-sig-requirement-00.txt

Section 4.9 describes the variation of LSP parameters (e.g. priority, 
bandwidth, etc.) between different branches of a P2MP LSP. The text 
currently says

    Any solution MUST NOT allow for variance of these parameters. That
    is,
    - no attributes set and signaled by the ingress of a P2MP LSP may be
      varied by downstream LSRs
    - there MUST be homogeneous QoS from the root to all leaves.

There has been some discussion that it might be desirable to make some 
variations under the control of the ingress LSR. Also, it has been pointed 
out that make-before-break is commonly used to vary these parameters.

We would like to propose to clarify this by changing section 4.9 to read as 
follows.

4.9 Variation of LSP Parameters

    Various parameters to an LSP (such as priority, bandwidth, etc.) are
    signaled along each branch of the LSP.

    Any solution MUST NOT allow for variance of these parameters within a
    single P2MP LSP. That is:
    - No attributes set and signaled by the ingress of a P2MP LSP may be
      varied by downstream LSRs.
    - There MUST be homogeneous QoS from the root to all leaves of a
      single P2MP LSP.

    Variation of parameters may be allowed so long as it applies to the
    whole LSP from ingress to all egresses.


If you have comments opposed to this text it would be helpful if you could 
supply realistic functional requirements that explain how you might deploy 
anything different.

Thanks,
Seisho 


_______________________________________________
mpls mailing list
mpls@lists.ietf.org
https://www1.ietf.org/mailman/listinfo/mpls