Re: PerfHist-TC-MIB

[Rajesh Abbi <[e-mail]>]

Hi Kaj,

I agree with Ken about the need for a Day-Counter that is synchronized
with the time-of-day clock.  Also, there is presently no MIB object to
configure the start-of-day time (which is by default 00:00).

Also, it should be recommended that the 15-minute interval be aligned
to a 1/4 hour boundary wrt the time-of-day.

I am not sure I understand the concept of 'xyzValidIntervals' and
'xyzInvalidIntervals'.

I believe each interval MUST be marked Valid/Invalid.  Just specifying
total number of valid/invalid intervals will not do.  If an interface
has been UP and running over 24-hours but was placed out-of-service for
5-minutes in the last hour, what happens ?  The way I read the
description,
the 'xyzValidIntervals' will be reset to 0.

The present definition of 'xyzInvalidIntervals' identifies an 'invalid'
interval as one in which "no valid data is available".  What about
intervals
in which 'some' valid data is available ?

If the 'xyzValidIntervals' can potentially consist of 'valid' &
'invalid'
intervals, then it is named inappropriately.

I still believe that this problem should be addressed as follows:
o Have a 'xxxNumIntervals' for a table which indicates total number of
  intervals available.
o Have a 'xxxDataSuspect' flag in each interval to indicate if there was
a 
  problem with the data collection in that interval.  If 'no' data is
collected
  in an interval, it could return 'noSuchObject' error.

Rajesh
========================================================================
Rajesh Abbi
Alcatel Network Systems
2912 Wake Forest Road
Raleigh, NC 27609-7860
Phone: [phone]
Fax  : [phone]
Email: [e-mail]
========================================================================


Ken Chapman wrote:
> 
> Kaj,
> I have an issue with it.  I agree with the three TC, but there is one
> missing:  PerfDayCount.
> It is quite different from PerfTotalCount; it starts at a specific time
> each day an is the aggregate of the 96 15-minute interval in the subsequent
> 24-hour interval.
> This is in support of the requirements specified in ANSI T1.231-1993,
> clause 9.1.2.1 (Current day register and Recent day registers).
> (I assume that the purpose of the MIB work is to support industry
> standards!)
>         Ken
> 
> At 06:55 PM 5/26/98 -0400, Kaj Tesink wrote:
> >At 03:55 PM 5/26/98 -0400, Ken Chapman wrote:
> >>Kaj,
> >>What's the status of <draft-ietf-atommib-perfhistTC-01.txt>?
> >>Is it dead?
> >>      Ken
> >
> >has been quality reviewed which resulted in minor comments.
> >i'm accommodating that now and will post in a few days
> >
> >kaj
> >
> >
> >>
> >
> >_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/
> >_/                                                                      _/
> >_/  Kaj Tesink                                                          _/
> >_/  Bellcore                                                            _/
> >_/  Business Internet Professional Services  vmail [phone]       _/
> >_/  331 Newman Springs Rd.                   email [e-mail]  _/
> >_/  Red Bank, NJ 07701                     faxmail [phone]       _/
> >_/                                                                      _/
> >_/  * NOTE AREA CODE CHANGE (908) --> (732) EFFECTIVE JUNE 1, 1997 *    _/
> >_/                                                                      _/
> >_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/
> >
> >
> >