PMP Mail Archive: RE: PMP> Justification for prtAlertTime and prtAlertTimeGrou

RE: PMP> Justification for prtAlertTime and prtAlertTimeGrou

Wagner, William (WWagner@digprod.com)
Thu, 20 Nov 1997 17:26:40 -0500

I agree with Bob. Having the event times refer back to MIB-II allows
determination of the clock time (although there would be some minor
calculation necessary in the application). I assumed Jay was jesting
with his comment.

The problem of two NIC's had been discussed in this context. The formal
solution (System refers to the controller not the NIC) is practically
unpalatable and probably any consistent implementation would be quite
usable in that, in most cases, the difference would not be significant.
Of course, if the NICs can be independently reset and that reset
includes sysUpTime, then realistically, the NIC's should be
resynchronized.

W. A. Wagner (Bill Wagner)
OSICOM/DPI

> -----Original Message-----
> From: Bob Pentecost [SMTP:bpenteco@boi.hp.com]
> Sent: Thursday, November 20, 1997 5:02 PM
> To: 'Jay Martin'
> Cc: pmp@pwg.org
> Subject: RE: PMP> Justification for prtAlertTime and
> prtAlertTimeGrou
>
> I'm not understanding something here. Jay says
> "(Is there any reason a client app would want
> to some way coodinate the value of MIB-II's sysUpTime with a
> prtAlertTime value?? Seems unlikely...)"
>
> What good is it to know the alert time if you don't relate it to the
> current sysUpTime?
>
> Bob
>
>
>
> ----------
> From: Jay Martin[SMTP:jkm@underscore.com]
> Sent: Thursday, November 20, 1997 12:01 PM
> To: Bob Pentecost
> Cc: pmp@pwg.org
> Subject: Re: PMP> Justification for prtAlertTime and prtAlertTimeGrou
>
> Bob,
>
> Regarding the second topic of your previous reply:
>
> > While we're on the topic of prtAlertTime, I had a question come to
> me about this object. As I had mentioned when we were changing this
> object to mandatory, the sysUpTime is kept in our Network Interface
> Card (the issue at the time was due to problems getting the time from
> the NIC, but I don't care to revisit that aspect). What's the printer
> to do when it has two such cards installed? One might argue that
> there's one power supply so the times start when the power is turned
> on, but the time could also be kept in an external NIC that
> communicates with the printer over the parallel port. Also, I'm not
> sure, but the clock drift between two NIC timers could be significant.
> >
> > Any thoughts?
>
> Good question. I would think that the manufacturer would simply
> choose one of the NICs as the "master" and use its values. As
> long as the same NIC values are used, then it probably wouldn't
> matter that much. (Is there any reason a client app would want
> to some way coodinate the value of MIB-II's sysUpTime with a
> prtAlertTime value?? Seems unlikely...)
>
> Of course, we don't build printers, so this opinion may not be
> shared with those who do.
>
> ...jay
>
> ----------------------------------------------------------------------
> -- JK Martin | Email: jkm@underscore.com --
> -- Underscore, Inc. | Voice: (603) 889-7000 --
> -- 41C Sagamore Park Road | Fax: (603) 889-2699 --
> -- Hudson, NH 03051-4915 | Web: http://www.underscore.com --
> ----------------------------------------------------------------------
>