Subject: | |
From: | |
Reply To: | |
Date: | Fri, 21 Mar 1997 17:57:47 CST |
Content-Type: | text/plain |
Parts/Attachments: |
|
|
We recently moved from 5.0 to 5.5 on three of our HP3K's. First to
move was a 937/RX used primarily for application development. Install
was a 'bear' we but haven't experienced any problems with the O/S yet.
The installation was a breeze on the other two - A 996/400 and a
980/200. However, we have had to restore the 5.0 O/S due to system
aborts. The 996/400 aborted 4 times on the first production day
following the install. It was three days before the 980/200 crashed.
I'm not directly involved in systems management, so I'll apologize in
advance for the lack of detail... I'm told that the number of the
first two aborts on the 996/400 was #-996. This is supposed to be
something that HPRC has never seen before and no patch is available.
The last two abort numbers were #619. This, I'm told, was a known bug
but the patch had not yet been written. The #619 is the result of
executing output redirection to a shared file.
e.g. FILE OUTPUT = APPLLOG,OLD;DEV=DISC;SHR;ACC=APPEND
ECHO Your data > *OUTPUT
A proposed work around was to add a ;SAVE option to the file equate.
The ';SAVE' on an existing permanent file seemed a bit strange to me
but HPRC says that it works. Anyway, it would have taken extensive
research to determine the impact of the work around on all of our
applications so moving back to 5.0 looked like the better option.
Question... We now have our development box on 5.5 and our production
machines on 5.0.. HPRC does not support backward compatibility for
any code compiled on 5.5 and then moved to 5.0 but, they tell me I
shouldn't experience any problems doing so based on the 5.0 patch
level that we are at. Should we move it back to 5.0 as well or take
our chances.?? Any and all feedback is appreciated..
Thanks in advance, lea [log in to unmask]
|
|
|