RE: ASON reqts

["Ash, Gerald R (Jerry), ALABS" <gash@att.com>]

Bert,

I think much of this discussion between ITU-T and IETF, and what you are proposing, has already occurred, some time ago:

dimitri> as of today we have two flavors "ASON/GMPLS" and "GMPLS"

steve> Can we please avoid creating a third variant of the protocol?

bart> I think the point was rather to go back to just a single
bart> GMPLS spec instead of two (let alone three) variants...

bert> If that is the case, and if we want to do that in IETF CCAMP 
bert> (not sure it is part of the current CCAMP charter),
bert> then it seems to me that we (CCAMP) should exercise these steps:
bert> 
bert> - send some liaison to ITU-T SG 15 to explain that we found that
bert>   there seems to be a GMPLS (for IETF) and a GMPLS-ASON standard
bert> - that we think that this is NOT goodness, and that we regret that
bert>   we did send the ASON people away a few years back and did not
bert>   closely follow their work in ITU
bert> - that we now believe that we should have worked better together,
bert> - that we would like to explore ways to try and merge the two
bert>   solution back into one common solution
bert> 
bert> And then await the response and hope ITU will agree and then see
bert> how the work can be organized.

Recall that we had extended discussions last January regarding the ASON-GMPLS issues, wherein many of the same points were made, and conclusions reached:

1. Brief History of ASON-GMPLS extensions:

In February 2002, ITU-T sent IETF/CCAMP a liaison statement identifying the 'gaps' between the ITU-T ASON requirements (sent 7 months earlier) and GMPLS protocols, which included:
a. call & connection separation, 
b. additional error codes/values, 
c. restart mechanisms,
d. support of crankback capability.

The liaison was presented to IETF/CCAMP at IETF-53/Minneapolis (March 2002), and requested assistance in closing the gaps and invited input from IETF.  IETF/CCAMP proposed to 'close the gaps' and CCAMP charter extensions to address same were suggested at IETF-54/Yokohama, but were not made.  This caused frustration within ITU-T in not getting attention to the needed ASON requirements in IETF/CCAMP.  

ITU decided to extend GMPLS protocols themselves, leading to the present interoperability concerns with ITU-GMPLS and GMPLS.  At the April/May 2002 ITU-T meeting, draft Recommendations were put forward to close the gaps, namely, draft Recommendations G.7713.2 (RSVP-TE ITU-GMPLS extensions) and G.7713.3 (CR-LDP ITU-GMPLS extensions).  

ITU-T sent a liaison in May 2002 to ask for comments on the draft Recommendations, and to request alignment with GMPLS, as follows: "Please consider including the proposed solutions provided in G.7713.2 and G.7713.3 to update the existing GMPLS signaling work in support of ASON requirements."  This request shows that the intent was to have extensions made by CCAMP to GMPLS protocols based on the ASON requirements and thus close the 'gaps'.

Steve Trowbridge noted that "After some private communication with the Area directors, we received some advice that one tool that might be used to finally get the IANA codepoint assignment complete would be to publish what we were doing in ITU-T as informational RFCs.  This is the stage we are at today."

In summary, ITU-T kept the IETF informed of its work, invited help to address the ITU-T ASON requirements, invited alignment of the base GMPLS protocols, but there was no response or interest from the IETF to do this.  The unfortunate outcome of this is to be standardizing 2 versions of GMPLS: ITU-GMPLS and GMPLS.

2. Several concerns were raised to this outcome, such as:

a. "I'm concerned with non-IETF groups like ITU-T deciding to develop their own incompatible extensions.  Groups like these should not be extending IETF protocols without consulting with the relevant IETF working groups.  Otherwise we end up with extensions that duplicate existing IETF functionality, don't coexist gracefully, and/or break interoperability."

b. "Which spec does a vendor implement and an operator use, given interoperability needs, etc.?  It would be analogous to the IETF specifying their version of G.709."

c. "The ITU has developed protocol extensions for GMPLS, something outside the charter of the ITU."

3. A resolution to the problem was proposed and got some agreement:

"It would be worthwhile to package the needed ASON extensions into a proposed GMPLS upgrade and presented to CCAMP as such: bring the requirements to CCAMP as a requirements draft, thrash them out, and get the necessary extensions adopted in CCAMP."

There were many who agreed with this approach, that is, to extend GMPLS (as originally proposed by the ITU-T) to meet the ITU-T ASON requirements.  Several of us joined the current effort with the intent that this is the direction we were taking.

IMO, you have proposed something that is quite consistent with all of the above.  However, much of what you propose has already transpired (perhaps 'unofficially') between the IETF and ITU-T.

Thanks,
Jerry