Re: comments on draft-ietf-iplpdn-frmib-dte-04.txt

[Rick Cornetti <[e-mail]>]

> 
>    >Is missing one value, in my reckoning.  ANSI T1.617a-1994 revises
>    >Annex D in an incompatible way.  (Thanks, ANSI!  The "element
>    >identifier" of "Link integrity verification information element" is
>    >changed from 0x19 to 0x03.)
> 
>    What idiot did that? Did they do anything useful with the protocol (tell
>    you the Bc/Be/throughput values, perhaps) that would justify implementing
>    it?
> 
> No idea which idiot.  The 20 "idiots" of working group T1S1.2 are
> listed on page vii...
> 
> 
> There is no change in the coding of this IE.  My only guess is that
> someone else in another ANSI working group allocated EI 0x19 in the
> ANSI codeset (shift 5 in US), so that they had to resolve the
> conflict.  There's no conflict in T1.607-1990, the only codeset 5 EI
> there is 0x1d for "operator system access IE".  However, I don't have
> ANSI T1.608, which also is noted to allocate EI's in codeset 5.
> 
> I don't know why they stuck with codeset 5 here.  So long as they were
> changing it, they could have used the ITU-T EI of 0x93 in codeset 0.
> 
> However, there is a meaningful change in the ANSI 1994 LMI.  They
> added a "delete" bit in the "PVC status" IE.  However, they DIDN'T
> change the EI for that, it's still 0x07 (codeset 5).  Maybe that's why
> they changed the other EI?!  Strange way to flag the change!!
> 
> 
> (acronyms: EI :== element identifier, IE :== information element).
> 
> 
> Far as I know, most FR switches are still provisioned for Interim LMI.
> All these other ones are just a nuisance to implementors, adding no
> value over Interim LMI.  Still, some fool will setup a network
> somewhere with the ANSI 1994 one, then it's back to the grindstone...
> 
> Remember, it could be worse, we could be writing the 50 flavors of
> full ISDN call control (DSS1), with codesets 5, 6, and 7, different on
> every switch vendor in every country in every operating authority!
> 


I noted this discrepancy too. Here is the response from the frame relay list:


----- Begin Included Message -----

From [e-mail] Mon Jan  9 12:43:55 1995
Date: Mon, 9 Jan 95 09:25:00 PST
From: [e-mail] (Art Biesiada)
Subject: FR: RE: Annex D LIV IE -- which is correct
To: [e-mail]
Sender: [e-mail]
Reply-To: [e-mail]
X-Info1: submissions to [e-mail]
X-Info2: [Un]Subscribe requests to [e-mail]
X-Info3: example- unsubscribe frame-relay [e-mail]
X-Info4: archives (soon to be announced)
Content-Length: 697

I have a copy of ANSI T1-617a-1994, where it states on page 6:
"In D.3.2, change Link integrity verification information element identifier
from "00011001" to "00000011" ...

So the correct ID is 0x03.

Art Biesiada
Ascom Timeplex



> Author: NAME: [e-mail] <SYS30::INTERNET::[e-mail]>
> 
> I have detected a discrepancy in T1.617 Annex D in the
> Link Integrity Verification IE. 
> 
> In T1.617 "Draft for T1 Default Ballot" (T1S1/91-352), 
> the LIV IE ID (octet 1) is shown as 0x03. In the published 
> ANSI T1.617-1991, Annex D, the LIV IE ID is shown as 0x19.
> 
> Which is correct?
> 
> Thanks in advance.
> 
> Rick Cornetti
> Wandel & Goltermann, Inc.
> RTP, NC, USA, 27709



----- End Included Message -----


So it appears that there was a "typo" in ANSI T1.617-1991, Annex D and that
ANSI fixed it in the 1994 version.


Regards,
Rick Cornetti
Network Systems
Wandel & Goltermann
RTP, NC, USA