Re: ForCES Protocol Implementation Issues

" SUBSCRIBE FORCES B. J. Kang " <ttt710516@GMAIL.COM> Wed, 19 September 2007 06:34 UTC

Message-Id: <WED.19.SEP.2007.023419.0400.>
Date: Wed, 19 Sep 2007 02:34:19 -0400
From: "SUBSCRIBE FORCES B. J. Kang" <ttt710516@GMAIL.COM>
Subject: Re: ForCES Protocol Implementation Issues
Mime-Version: 1.0
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain; charset="big5"

On Mon, 17 Sep 2007 23:20:13 -0400, Joel M. Halpern 
<joel@stevecrocker.com> wrote:

>Jamal and I have been discussing the FE Status variable issue.
>The following summarizes my concerns, and if we are prepared to allow
>the specification of the legitimate value to set as a comment, my
>proposed changes to the status enumeration.
>
>
>The primary issue is that there is no way to cleanly represent the
>fact that the CE is only allowed to set the variable to a subset of
>the enumeration.  (Even SNMP does not handle that formally.)
>
>As a secondary matter, there is the question of what value the CE
>sets the variable to when it wants to turn the FE back on.  It can
>not set it to either OperUp or OperDown.  So we need at least one
>more value to represent "the CE has told me to become operational,
>but I have not yet managed it."  Which is the value the CE would set
>the variable to.
>
>As a minor fillip, since I consider administrative something of
>larger scope than CE control, I wanted a distinct value for the CE to
>use for "I, the CE, am ordering you to stop."
>

Thanks for all your help

>Yours,
>Joel