From: McGuinness Todd (todd.mcguinness@nagra.com)
Date: Mon Apr 15 2002 - 08:37:07 EDT
Thanks to David J. DeWolfe whom has experienced this issue previously and so
elequantly described his experience and some tests below:
Cheers,
tm
David J. DeWolfe writes:
> Todd;
>
> Here's an internal email I sent out quite some time ago regarding this
issue:
>
> >As of yesterday, Thursday 08/25/2000, the following error started
> >appearing on Umga2's console whenever Umga2 was powered-on or init'd at
> >the SRM console prompt (P00>>> init):
> >
>
>---------------------------------------------------------------------------
> >Testing the System
> >Testing the Memory
> >Testing the Disks (read only)
> >
> >*** Hard Error - Error #8 -
> >Diagnostic Name ID Device Pass Test Hard/Soft
1-JAN-2066
> >exer_kid 0000025d 0 0 1 0
12:00:01
> >Buffer counts differ - buf1:0, buf2:512, location:2a00
> >*** End of Error ***
>
>---------------------------------------------------------------------------
> >
> >This first appeared when I powered Umga2 on after powering if off to
> >remove the 18G internal disk that I had borrowed from Paxson in order to
> >install the DE602-AA (NIC) drivers via the NHD (New Hardware Delivery)
> >mechanism. Umga2 successfully booted up after the error and all appeared
> >to be fine.
> >
> >This morning I opened SQR # C000825-1277 and spoke with Al in California.
> >His search for the error message revealed that it was related to the CCL
> >(Command Console Lun) associated with the HSZ70 controller(s) and the
disk
> >test actually attempting to read from the "device". We verified this by
> >detaching the SCSI cable from Umga2's kzpba the other end of which is
> >attached to the HSZ70's in 6U. I init'd with the cable detached and the
> >error did not occur. As soon as I reattached the cable (and init'd) the
> >error reappeared.
> >
> >Why we have not seen this behavior on any of the other systems attached
to
> >HSZ70's is unknown, though it could be something specific to the Umga2
> >firmware release (v5.7-8). Why we hadn't seen it previously on Umga2 is
> >unknown. Whether the removal of the 18G disk from Umga2's internal shelf
> >somehow triggered this is unknown as well. The analyst indicated that
this
> >would likely be fixed in a subsequent firmware release though he never
> >referred to the problem as a "known" bug.
> >
> >During debugging, the analyst had me do the following at the console
prompt:
> >
> > P00>>> sys_exer
> >
> >which exercises devices (in the background) by reading/writing from/to
> >memory and reading from devices (including disks, floppies and cd's).
> >
> > P00>>> show_status
> >
> >shows the status of the above tests. The output looks like:
> >
> >P00>>>show_status
> >ID Program Device Pass Hard/Soft Bytes Written Bytes
Read
> >-------- ------------ ------------ ------ --------- -------------
> >-------------
> >00000001 idle system 0 0 0 0
0
> >000002d6 memtest memory 3 0 0 2147483648
2147483648
> >000002da memtest memory 3 0 0 2155872256
2155872256
> >000002e4 memtest memory 3 0 0 2147483648
2147483648
> >000002eb exer_kid dka0.0.0.7.1 0 0 0 0
3865600
> >000002ec exer_kid dka100.1.0.7 0 0 0 0
3865600
> >000002ed exer_kid dkd0.0.0.9.0 114 0 0 0
1175040
> >000002ee exer_kid dkd100.1.0.9 0 0 0 0
1174528
> >000002ef exer_kid dkd300.3.0.9 0 0 0 0
1174528
> >000002f6 exer_kid dqa0.0.0.105 0 0 0 0
1410048
> >0000030d exer_kid dva0.0.0.0.0 0 0 0 0
320512
> >
> >and to stop the tests you do:
> >
> > P00>>> kill_diag
> >
> >Al indicated that these are good tests to run after a system crash since
> >you can run them at the console level.
>
>
>
>
> At 01:22 AM 4/13/2002, you wrote:
>
> >Hello all,
> >
> >I have been incredibly hammered the last 2 days sorry if my name is
getting
> >too irritating in your inbox.
> >
> >I am having the following error in the SRM of an ES40:
> >
> >*** Hard Error - Error #8 -
> >Diagnostic Name ID Device Pass Test Hard/Soft
> >1-JAN-2000
> >exer_kid 0000015b dkc200.2.0.1.0 0 0 1 0
> >12:00:01
> >Buffer counts differ - buf1:0, buf2:512, location:2a00
> >
> >*** End of Error ***
> >
> >The dkc200 is a HSZ80 recently installed by COMPAQ - and I am really
> >surprised that the would leave it in this state, but I am here on a
> >Saturday after a long nite to deal with more issues. C'est la vie.
> >
> >Does anyone have any information on how I can correct the buffer counts?
or
> >how I can do anything here to fix this?
> >
> >cheers,
> >
> >tm
>
>
> David
> mailto:sxdjd@ts.sois.alaska.edu
>
Todd M. McGuinness
NagraVision SA
0041217320623
This archive was generated by hypermail 2.1.7 : Sat Apr 12 2008 - 10:48:38 EDT