Re: Modem Statistics Group

"Mark S. Lewis" <mlewis@telebit.com> Wed, 08 December 1993 19:09 UTC

Received: from ietf.nri.reston.va.us by IETF.CNRI.Reston.VA.US id aa10863; 8 Dec 93 14:09 EST
Received: from CNRI.RESTON.VA.US by IETF.CNRI.Reston.VA.US id aa10859; 8 Dec 93 14:09 EST
Received: from apache.telebit.com by CNRI.Reston.VA.US id aa15683; 8 Dec 93 14:09 EST
Received: from america.Sunnyvale.Telebit.CO (america-bb.sunnyvale.telebit.com) by apache.telebit.com (4.1/SMI-4.1/Telebit-Apache-Brent-930718) id AA19858 to ietf-archive@cnri.reston.va.us; Wed, 8 Dec 93 11:01:27 PST
Received: from yoyo.telebit.com by america.Sunnyvale.Telebit.COM (4.0/america.telebit.com-DBC-930718) id AA12981 to modemmgt@apache.Sunnyvale.Telebit.COM; Wed, 8 Dec 93 11:01:21 PST
Date: Wed, 08 Dec 1993 11:01:21 -0800
Sender: ietf-archive-request@IETF.CNRI.Reston.VA.US
From: "Mark S. Lewis" <mlewis@telebit.com>
Message-Id: <9312081901.AA12981@america.Sunnyvale.Telebit.COM>
Received: by yoyo.telebit.com (4.1/SMI-4.1) id AA12610; Wed, 8 Dec 93 11:01:22 PST
To: crozman@usr.com
Cc: modemmgt@telebit.com, rroyston@usr.com
In-Reply-To: <9311077552.AA755283336@robogate.usr.com> (crozman@usr.com)
Subject: Re: Modem Statistics Group
Reply-To: Mark.S.Lewis@telebit.com
Mime-Version: 1.0
Content-Type: text/plain; charset="US-ASCII"

Thanks for the comments.  Sounds like we drop:

   mdmStatWatchdogTimeouts
   mdmStatInactivityTimeouts

These are in question:

    mdmStatsDTERingNoAnswers
    mdmStatsMdmRingNoAnswers (not in recent versions of draft)

I suggest without strong justification and consensus that we need
these, they should be deferred until the extensions mib.  (People that
want these should make their case.)

Lastly, I like Rick's suggestion about the keeping the stats by link
rate.  We need to add description to them that explains they are
counted at their initial rate not some subsequently negotiated rate.

There will be an entire instance of the modem mib for each modem.
This has not been reflected in the draft yet, but this is the plan.
Everything in the current draft will be pushed down one level into a
table indexed by modem.  These changes need to be made to the draft.

... Mark

==========--------------       Mark S. Lewis      ----------==========
Mark.S.Lewis@Telebit.com       Telebit Corp.      Voice (408) 745-3232

>>>>> On Tue, 07 Dec 93 08:55:36 CDT, "Rozman, Chris" <crozman@usr.com> said:


>      I agree with Rick's assessment of the first two of the three objects in 
>      point #1, but I would question eliminating mdmStatsDTERingNoAnswers.  It 
>      often does require special hardware, but it is very useful to a manager of 
>      a modem pool.  Also, what about mdmStatsMdmRingNoAnswers?  Has this ever 
>      been in the MIB?  If so, is it still?  The difference between broken DTEs 
>      and broken Modems is important!
>      
>      As for point #2, I may agree, but I have some questions.  Is it your intent 
>      to doubly index this table by "which modem" and "link rate", with a third 
>      column for a count of the number of calls that are INITIATED at that rate?  
>      (of course link rates can change during the call)  I suppose this would be 
>      a sparse table???  If I understand correctly, this doesn't sound bad.
>      
>      -Chris

> ______________________________ Reply Separator _________________________________
> Subject: Modem Statistics Group
> Author:  ,"Royston, Rick" <rroyston@usr.com> at Internet
> Date:    12/6/93 5:12 PM


>   
>   In reviewing the 10/21/93 MIB I have the following comments about the 
>   objects.
>   
>   1.  I would like to delete the following objects as of limited usefulness
>   
>     mdmStatWatchdogTimeouts (result of software bug)
>     mdmStatInactivityTimeouts (too hard to define inactivity)
>     mdmStatsDTERingNoAnswer (too associated only with Analog modems.  Often 
>   requires special hardware not present
>   
>   
>   2.    I would like to offer a replacement table for the utilization 
>   counters mdmStat[2400orLess|2400To14000|GreaterTHan14000]Connections
>   should be replaced with a two column table.  THe first column would be 
>   the link rate, and the second column would be a counter that indicated 
>   the number of calls established at each speed.  There would be no 
>   distinction between incoming and outgoing calls.
>   
>   
>   Any comments??
>   
>   
>   Rick