HP3000-L Archives

May 1998, Week 2

HP3000-L@RAVEN.UTC.EDU

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
"Schelin, Howard (Miami)" <[log in to unmask]>
Reply To:
Schelin, Howard (Miami)
Date:
Mon, 11 May 1998 16:48:50 -0400
Content-Type:
text/plain
Parts/Attachments:
text/plain (32 lines)
In need of insight

Background
For years we have moved patches across out 3000 machines by
1-      Installing patches on our development machine
2-      Creating a SLT tape. And a store tape with files from all the HP
accounts eg. Sys Telesup etc.
3-      Updating production machines with the SLT and restoring files
(minus files that are unique to each machine eg. nmconfig , log
files,logdcc,command ....
4-      HP endorsed this method and it has worked without incident until
now
5-      HP told us the above method would be valid in updating from 5.0
to 5.5

Problem
Last week we updated a production machine to 5.5 using the above method.
We were looking good; system came up; background jobs were running and
then our batch took off. Two executables aborted with a Data Memory
Protection Trap. Both these executables run to completion on the
development machine that is running 5.5 (where the SLT and accompanying
files come from). We backdated the production machine to 5.0 and the
executables ran to completion.  To Summarize; a executable runs to
completion on a 987 running 5.5 and this same executable has a DMPT on a
996 running 5.5. Are there some machine differences or is the method of
updating no longer valid with 5.5. Is Virtual Space Management different
on a 987 vs a 996

Thanks
Howard Schelin
[log in to unmask]     phone 305-376-3026      fax 305 376 3388

ATOM RSS1 RSS2