Hi Robert,
very interesting!
I'd try to give my understanding by extending your quote from RFC 5586:
"However, in other MPLS environments, this document places no restrictions on where the GAL may appear within the label stack or its use with PWs."
I interpret that in non-TP environments GAL may be used as suggested in the draft but in MPLS-TP - the control word and VCCV must be used.

Regards,
Greg

Hi Shahram,
I agree, that would make it unambiguous. Perhaps we might revisit a question that was somewhat lightly mentioned during the WG meeting in Anaheim  "Should use of CW in all PWs be mandatory?"

Regards,
Greg

Hi Greg,
 
Even accepting your interpretation, this still doesn’t solve the problem of PW OAM for MPLS-TP when CW is not present. My suggestion is to mandate CW if PW OAM is required.
 
Regards,
Shahram
 

From: pwe3-bounces <at> ietf.org [mailto:pwe3-bounces <at> ietf.org] On Behalf Of Greg Mirsky
Sent: Wednesday, June 02, 2010 5:05 PM
To: Robert Rennison
Cc: swallow <at> cisco.com; matthew.bocci <at> alcatel-lucent.co.uk; pwe3 <at> ietf.org; Luca Martini
Subject: Re: [PWE3] Possible Contradiction re use of GAL in pwe3-static-pw-status

Hi Robert,
very interesting!
I'd try to give my understanding by extending your quote from RFC 5586:
"However, in other MPLS environments, this document places no restrictions on where the GAL may appear within the label stack or its use with PWs."
I interpret that in non-TP environments GAL may be used as suggested in the draft but in MPLS-TP - the control word and VCCV must be used.

Regards,
Greg

On Wed, Jun 2, 2010 at 8:15 AM, Robert Rennison <Robert.Rennison <at> ecitele.com> wrote:

Correcting Luca’s email address, outlook  address auto-complete dug this up from the depths.
 
 
 

From: pwe3-bounces <at> ietf.org [mailto:pwe3-bounces <at> ietf.org] On Behalf Of Robert Rennison
Sent: Wednesday, June 02, 2010 9:57 AM
To: luca <at> level3.net; swallow <at> cisco.com; matthew.bocci <at> alcatel-lucent.co.uk
Cc: pwe3 <at> ietf.org
Subject: [PWE3] Possible Contradiction re use of GAL in pwe3-static-pw-status


Luca,George,Matthew,
The section 5.4.1 of the static-pw-status draft  appears to allow me to use the GAL on a PW which is not using the control word, that’s nice, since some PWs do not use the CW.
From draft-ietf-pwe3-static-pw-status-00
5.4.1. Label stack for a message destined to the next PE
  A PE that needs to forward a PW OAM status message to the adjacent PE
   across the PSN tunnel, MUST set the PW label TTL field to 1.
   Furthermore if the control word is not in use on the particular PW,
   the PE MUST also place the GAL reserved label [RFC5586 <http://tools.ietf.org/html/rfc5586> ], below the PW
   label also with the TTL field set to 1.

However from  RFC 5586  I see.
 
RFC 5586 4.2.  GAL Applicability and Usage

   In MPLS-TP, the GAL MUST be used with packets on a G-ACh on LSPs,
   Concatenated Segments of LSPs, and with Sections, and MUST NOT be
   used with PWs. It MUST always be at the bottom of the label stack
   (i.e., S bit set to 1).

This appears to preclude the use of the GAL on a PW.
 
So,  how can I run  pw-status on a PW and not violate RFC 5586.  I’m sure there’s some subtlety I am missing here so hoping for clarity.
 
 
Cheers
 
Rob Rennison
 
 

_______________________________________________
pwe3 mailing list
pwe3 <at> ietf.org
https://www.ietf.org/mailman/listinfo/pwe3
 

_______________________________________________
pwe3 mailing list
pwe3 <at> ietf.org
https://www.ietf.org/mailman/listinfo/pwe3

From: pwe3-bounces <at> ietf.org [mailto:pwe3-bounces <at> ietf.org] On Behalf Of Thomas D. Nadeau
Sent: Thursday, June 03, 2010 1:34 PM
To: Shahram Davari; Greg Mirsky; Robert Rennison
Cc: George Swallow; matthew.bocci <at> alcatel-lucent.co.uk; pwe3 <at> ietf.org; Luca Martini
Subject: Re: [PWE3] Possible Contradiction re use of GAL in pwe3-static-pw-status

    I agree %100. This is something that we discussed in the last PWE3 meeting with only minor objections. The consensus seemed to be that we should mandate the CW going forward, and also to retrofit old/existing RFCs that make it optional.

    --Tom


On 6/2/10 8:21 PM, "Shahram Davari" <davari <at> broadcom.com> wrote:

Hi Greg,
 
Even accepting your interpretation, this still doesn’t solve the problem of PW OAM for MPLS-TP when CW is not present. My suggestion is to mandate CW if PW OAM is required.
 
Regards,
Shahram
 

From: pwe3-bounces <at> ietf.org [mailto:pwe3-bounces <at> ietf.org] On Behalf Of Greg Mirsky
Sent: Wednesday, June 02, 2010 5:05 PM
To: Robert Rennison
Cc: swallow <at> cisco.com; matthew.bocci <at> alcatel-lucent.co.uk; pwe3 <at> ietf.org; Luca Martini
Subject: Re: [PWE3] Possible Contradiction re use of GAL in pwe3-static-pw-status

Hi Robert,
very interesting!
I'd try to give my understanding by extending your quote from RFC 5586:
"However, in other MPLS environments, this document places no restrictions on where the GAL may appear within the label stack or its use with PWs."
I interpret that in non-TP environments GAL may be used as suggested in the draft but in MPLS-TP - the control word and VCCV must be used.

Regards,
Greg

On Wed, Jun 2, 2010 at 8:15 AM, Robert Rennison <Robert.Rennison <at> ecitele.com> wrote:

Correcting Luca’s email address, outlook  address auto-complete dug this up from the depths.
 
 
 

From: pwe3-bounces <at> ietf.org [mailto:pwe3-bounces <at> ietf.org] On Behalf Of Robert Rennison
Sent: Wednesday, June 02, 2010 9:57 AM
To: luca <at> level3.net; swallow <at> cisco.com; matthew.bocci <at> alcatel-lucent.co.uk
Cc: pwe3 <at> ietf.org
Subject: [PWE3] Possible Contradiction re use of GAL in pwe3-static-pw-status


Luca,George,Matthew,
The section 5.4.1 of the static-pw-status draft  appears to allow me to use the GAL on a PW which is not using the control word, that’s nice, since some PWs do not use the CW.
From draft-ietf-pwe3-static-pw-status-00
5.4.1. Label stack for a message destined to the next PE
  A PE that needs to forward a PW OAM status message to the adjacent PE
   across the PSN tunnel, MUST set the PW label TTL field to 1.
   Furthermore if the control word is not in use on the particular PW,
   the PE MUST also place the GAL reserved label [RFC5586 <http://tools.ietf.org/html/rfc5586> ], below the PW
   label also with the TTL field set to 1.

However from  RFC 5586  I see.
 
RFC 5586 4.2.  GAL Applicability and Usage

   In MPLS-TP, the GAL MUST be used with packets on a G-ACh on LSPs,
   Concatenated Segments of LSPs, and with Sections, and MUST NOT be
   used with PWs. It MUST always be at the bottom of the label stack
   (i.e., S bit set to 1).

This appears to preclude the use of the GAL on a PW.
 
So,  how can I run  pw-status on a PW and not violate RFC 5586.  I’m sure there’s some subtlety I am missing here so hoping for clarity.
 
 
Cheers
 
Rob Rennison
 
 

_______________________________________________
pwe3 mailing list
pwe3 <at> ietf.org
https://www.ietf.org/mailman/listinfo/pwe3
 

_______________________________________________
pwe3 mailing list
pwe3 <at> ietf.org
https://www.ietf.org/mailman/listinfo/pwe3

Dear Luca,
I think that I can not find requirement in MPLS-TP that would be close to requirement to use CW in PW as you've indicated. I read RFC 5586 as requirement to use CW in PWs if DCN (CSN, MSN) required. I think that PW in MPLS-TP network might not use CW. Then the PW would not be available to MPLS-TP OAM. We can view this as implicit requirement, practical thing to do, but, in my view, non-CW PWs are within MPLS-TP.

Regards,
Greg