Re: [PWE3] draft-medved-pwe3-of-config-00.txt posted, review please

Hi all again,

One more comment (forgot to include it in the original email) added at the end of the list in blue.

Regards,

     Sasha

________________________________
From: Alexander Vainshtein
Sent: Wednesday, July 04, 2012 7:05 AM
To: Andrew G. Malis; David Meyer; Jan Medved; Andrew McLachlan
Cc: pwe3@ietf.org; Mishael Wexler; Rotem Cohen; Gideon Agmon; Andrew Sergeev; Idan Kaspit; Vladimir Kleiner
Subject: RE: [PWE3] draft-medved-pwe3-of-config-00.txt posted, review please

Andy and all,

I've looked up the draft in question, and I have some comments/questions in addition to Andy's:

  1.  The PWE3 WG has defined multiple MIBs that can be used for PW setup and teardown by a      management application (MPLS-TP or not). IMO the draft should indicate that the method proposed there is an alternative not just to PW signaling via T-LDP (as Andy has said) but also to static PW setup using these MIBs.
  2.  The wording of the draft seems to suggest a difference between the head-end PE and the tail-end one, e.g., as in he following fragment: "Although the Virtual OF switch is the same a both the head-end and tail-end nodes, the group table is only used at the head-end node, where a Fast Failover group is set up for each pair of ports that correspond to the primary-backup Transport LSP pair.  At the tail-end node, only flows are set up." I do not understand how does this (and similar) fragments deal with the fact of PWs being inherently bidirectional.
  3.  The PWs run in MPLS (or MPLS-TP) tunnels in order to improve scalability of P routers, because the same tunnel between a given pair of PEs can be shared by multiple PW instances. It is not clear from the text how such sharing can be accommodated in the proposed OF switch model.
  4.   Reference to RFC 6423 suggests that OAM discussion in the draft is PW OAM and OAM for Tunnel LSPs. Does this imply that the protection actions are triggered by the PW-level OAM, and, what is supposed to happen if a pair of T-LSPs is shared by multiple PW instances?
  5.  The text defining the OF switch and its matching rules is not clear. In particular, it seems to imply that PW labels are checked from being received from a specific T-LSP (as a port of the OF switch). IMHO this a clear contradiction to the MPLS (and MPLS-TP) data plane as defined in RFC 3031, and to the fact that PW labels are allocated from the per-platform label space (as per RFC 4447).
  6.  In order to assign PW labels, the management application needs the information about the pool of unused MPLS labels on each of the PE devices. The methods for obtaining such information using the OpenFlow tehnique are not mentioned in the draft.

My 2c,

     Sasha

________________________________________
From: pwe3-bounces@ietf.org [pwe3-bounces@ietf.org] on behalf of Andrew G. Malis [amalis@gmail.com]
Sent: Tuesday, July 03, 2012 8:14 PM
To: David Meyer; Jan Medved; Andrew McLachlan
Cc: pwe3@ietf.org
Subject: Re: [PWE3] draft-medved-pwe3-of-config-00.txt posted, review please

Dave, Jan, Andrew,

An interesting draft. I have a few comments, with my WG chair hat off:

1. An editorial comment, there's no "-" or internal uppercase "W" in
"Pseudowire", which is the proper spelling (I saw at least three
different spellings in the draft).

2. Section 1.3 states: "PWs are
   configured as bi-directional paths over the MPLS-TP network, usually
   by an external management platform.  At present no open standards
   exist to provision these PWs, and therefore there is a reliance on
   vendor specific provisioning platforms."

Certainly, there's nothing that prevents the use of T-LDP for PW use
in MPLS-TP networks, this is already documented in section 5.2 of RFC
6373.

Speaking of which, you should also have a table in your draft similar
to the table in section 5.2 that shows how your draft meets the
MPLS-TP control plane requirements listed in section 2 of RFC 6373.

You may also wish to read draft-cao-pwe3-mpls-tp-pw-over-bidir-lsp,
which specifies useful functionality in this area.

In any case, you should be clear that this is an alternative control
plane to T-LDP.

3. Section 4: In addition to RFC 6423, you may wish to include a
reference to draft-ietf-pwe3-vccv-for-gal, which defines the VCCV mode
for use with MPLS-TP. You may also wish to reference RFC 5860.

Cheers,
Andy

On Tue, Jul 3, 2012 at 1:11 PM, David Meyer <dmm@1-4-5.net> wrote:
> Folks,
>
> "MPLS-TP PseudoWire Configuration using OpenFlow 1.3" has just been
> posted. Looking forward to your review.
>
> --dmm (for Jan, dward, and Andrew)
> _______________________________________________
> pwe3 mailing list
> pwe3@ietf.org
> https://www.ietf.org/mailman/listinfo/pwe3
_______________________________________________
pwe3 mailing list
pwe3@ietf.org
https://www.ietf.org/mailman/listinfo/pwe3

This e-mail message is intended for the recipient only and contains information which is CONFIDENTIAL and which may be proprietary to ECI Telecom. If you have received this transmission in error, please inform us by e-mail, phone or fax, and then delete the original and all copies thereof.