Path: utzoo!bnr-vpa!bnr-fos!bigsur!bnrgate!is2!marmen
From: marmen@is2.bnr.ca (Rob Marmen 1532773)
Newsgroups: comp.sys.apollo
Subject: Intermittent DN4500 Problem
Keywords: dn4500 parity crash intermittent
Message-ID: <119@bnrgate.bnr.ca>
Date: 19 Oct 89 19:19:28 GMT
Sender: news@bnrgate.bnr.ca
Lines: 33


	We are experiencing an intermittent problem on some of our
dn4500s. These machines are equiped with 16 meg of memory, a 348 
meg disk drive and an f series monitor.

	The machines will mysteriously reboot themselves. They will
display the rainbow, run diagnostics and bring up the o/s. The 
error log will sometimes show a memory parity error, but sometimes
no error is logged, just the startup. We have trapped a crash in 
progress and the error was "memory parity error". However, the 
workstation had several different and verified sets of memory
boards. We suspect that the message is a fake.

	When the machine crashes with users logged in, the users had
been running a large application with large data files. The most
common operation before the crash was a save of the data to disk.
Needless to say, the users are not happy.

	The problem is not network or application related. Some 
machines are extremely reliable, while a machine next to it
will crash once a day.

	Local Apollo support has been doing a fantastic job trying to
debug the problem. However, I would like to know if anyone else
has seen the problem? We're rapidly running out of ideas.
Anything at this point will be welcome.

Thanking everyone in advance...    rob...
  
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
| Robert Marmen             marmen@bnr.ca  OR             |
| Bell Northern Research    marmen%bnr.ca@cunyvm.cuny.edu |
| (613) 763-8244         My opinions are my own, not BNRs |