HP3000-L Archives

October 1995, Week 3

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:
Jeff Kell <[log in to unmask]>
Reply To:
Jeff Kell <[log in to unmask]>
Date:
Thu, 19 Oct 1995 19:47:46 EDT
Content-Type:
text/plain
Parts/Attachments:
text/plain (46 lines)
On Wed, 18 Oct 1995 13:04:31 GMT jeff brown said:
>What kind of experience is out there upgrading from 4.x to 5.x MPEix?  Did
>any of the new POSIX items create unexpected headaches?
 
This has been asked numerous times before.  You say "4.x" but if that means
"4.5" then you've already had most of the trauma :-)  Let's assume 4.0.
 
Disk usage increases substantially.  Performance may be impacted on small
memory systems.  There are some patches to get (notably PowerPatch 2, a.k.a.
Express 2).  No update in my entire life has been uneventful, but considering
the added functionality between 4.0 and 5.0, HP has done an outstanding job
despite my sometimes critical remarks here.  It is probably a good idea to
ask here about the update, or review the archives for material, and I won't
re-iterate them here (others probably will).  But I can say:
 
* Get the necessary software.  READ THE DOCUMENTATION AND THE NOTES.  Most
  all of the problems are noted in the documentation (like checking your
  DDS firmware revision if applicable).
 
* Verify your SLT (as documented), your available disc space, and for the
  very paranoid, VSTORE verify your FOS and SUBSYS tapes.
 
* Order the latest PowerPatch; and if you have access to an engineer as
  with PSS level support, get reactive patches for your products; most
  notably networking products (NMS, NST, FTP, etc) and MPE patches.  Never
  install a virgin push this late in the game since patches are available.
 
* Most existing things still work.  Most patches are for new things.
 
I know of three things that broke that worked fine before, and aren't
otherwise documented in the standard materials:
 
* COBOL files opened for I-O mode:  if you READ a record then WRITE it,
  it updates the record you read.  This doesn't fly on 5.0, you have to
  change the WRITE to a REWRITE.
 
* Fortran/iX runtime library changed.  We had one case of unresolved external
  in a program we don't have the source for (3rd party - SPSS).
 
* Compiled C/iX programs under 5.0 will likely not run on pre-5.0 systems.
 
There have been some other quirks, but most are fairly well documented or else
deal with new functionality.
 
[\] Jeff Kell <[log in to unmask]>

ATOM RSS1 RSS2