Mysterious (Sort?) Problem

[Mike Godsey]

I have had this happen before. Increase the process/report limit. What
happens is it creates a fixed size file that completely fills and then PH
resets the file and continues to write records beginning again at the start
of the file. By increasing the limit a larger file is created and the
overflow does not happen. If this a fixed file ( not created in the job)
then purge it or MPEX ALTFILE the file to increase it's size.
Mike Godsey

By increasing the
----- Original Message -----
From: "Ron Burnett" <ron@cryptic.rch.unimelb.EDU.AU>
To: <powerh-l@sphere.swau.edu>
Sent: Thursday, July 20, 2000 6:34 AM
Subject: Mysterious (Sort?) Problem


> Hello Everyone,
>
> I'm stumped by this.
>
> I've got a program that has run with no errors for years ..... we're now
> up to PowerHouse 8.19C on MPE/iX 5.5 PP4.
>
> Basically, the program reads a (Image/SQL) data set containing
> the record number/name and a few other details of people who have
> an alias, and contains a pointer to another record where all their
> data is stored, including the name by which they are perhaps better
> known.  (There is nothing sinister about this--it is a medical history
system
> for a paediatric hospital.)
>
> Anyway, the program sorts the composite record and prints a
line-per-patient
> report to a disc file which is then FCOPY-ed to tape and sent off to be
> printed on microfiche, along with some other stuff (separate tape,
separate
> program) which is used, in case the computer system is temporarily
> unavailable.
>
> This month, with only just under 19,000 records to sort, the program
> reports the correct number of records read, sorted and the expected number
> of pages printed, but the output file has only 88 records, beginning in
the
> middle of the pentultimate page with Zxxx surnames, a page header, then
> some more records and end of report.
>
> The program explicitly purges the old file, if it exists (if finfo
> ('fname','exists') then
> purge fname endif).  Yet the STDLIST says, at the end of the report phase,
> 'fname' exists, OK to purge   ..... and being a job, goes right ahead and
> finishes up without an error.
>
> So how can the 'fname' exist?  Do we have a sort error?  Might it help if
> I put in some code to override the default sort workfile specs (how do I
do
> this?).
>
> I've broken the program into parts, made sure none of the sort items are
> numeric, combined the (four separate) sort keys into a single field with
> a define, and the results are always the same ...... I only get the bottom
> 80 or so records.
>
> Any clues?  Is this a known bug?
>
> Many thanks for any help.
>
> Ron Burnett
> Women's & Children's Health
> Melbourne, Australia
> ron@lis.rch.unimelb.edu.au
>
>
> = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
=
> Subscribe: "subscribe powerh-l" in message body to
majordomo@lists.swau.edu
> Unsubscribe: "unsubscribe powerh-l" in message to majordomo@lists.swau.edu
> This list is closed, thus to post to the list, you must be a subscriber.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
Subscribe: "subscribe powerh-l" in message body to majordomo@lists.swau.edu
Unsubscribe: "unsubscribe powerh-l" in message to majordomo@lists.swau.edu
This list is closed, thus to post to the list, you must be a subscriber.