SetLL & ReadE problem
itccbrno
we have problem with join files. (Not DDM).
The topic is to process the group of records with the same partial key.
SetLL opcode is used to position the pointer to desired position in file and
the ReadE opcode to process the group of records.
When the desired group exists, the routine works fine.
When the desired group doesn't exists, so SetLL opcode points the file
before the next group, subsequent ReadE opreration returns the records from
next group even the KLIST values requested nonexisting group! So result
indicator NR should be on, but it is not until the not wanted group ends!
Please help.
--
Robert
jorge bessa
It seems that it is the same problem we had in our office ... I think that
you need a PTF (I don't know which one) ... we had this problem with join
file and find out that the problem was a PTF not installed on V4R3.
Hope this helps you .....
itccbrno wrote in message <3711BBE4...@mbox.vol.cz>...
Paul Nicolay
From the IBM APAR database...
Abstract
LIC-DB READE RETURNS THE WRONG RECORD
Error Description
If a READE is done to a JLF with a key that doesn't exist, the
EOF indicator is not set to *ON and the wrong record is read.
This problem only happens to RPGLE in R430 (not in OPM).
Problem Summary
**********************************************************
* PROBLEM: (MA19007) Licensed Program = 5769999 *
* Data Integrity *
**********************************************************
* USERS AFFECTED: All OS/400 users. *
*********************************************************
* RECOMMENDATION: Apply LIC PTF MF20516. *
**********************************************************
If a READE is done to a JLF with a key that doesn't exist the
EOF indicator is not set to *ON and the wrong record is read.
This problem only happens to RPGLE in R430 (not in OPM).
Problem Conclusion
The compare key values were not properly set. The code has
been changed to set the compare values properly.
Temporary Fix
*********
* HIPER *
*********
Comments
Circumvention
None.
PTFs Available
R430 MF20516
Hope this helps...
Kind regards,
Paul
---------------
itccbrno wrote in message <3711BBE4...@mbox.vol.cz>...
>Programs on our development AS400 V4R2 runs OK but on customer side on V4R3
>we have problem with join files. (Not DDM).
>
>The topic is to process the group of records with the same partial key.
>SetLL opcode is used to position the pointer to desired position in file
and
>the ReadE opcode to process the group of records.
>
>When the desired group exists, the routine works fine.
>
>When the desired group doesn't exists, so SetLL opcode points the file
>before the next group, subsequent ReadE opreration returns the records from
>next group even the KLIST values requested nonexisting group! So result
>indicator NR should be on, but it is not until the not wanted group ends!
>
>Please help.
>--
>Robert
The contents of this message express only the sender's opinion.
This message does not necessarily reflect the policy or views of
my employer, Merck & Co., Inc. All responsibility for the statements
made in this Usenet posting resides solely and completely with the
sender.
Charles R. Pence
>
> Programs on our development AS400 V4R2 runs OK but on customer side on V4R3
> we have problem with join files. (Not DDM).
>
> The topic is to process the group of records with the same partial key.
> SetLL opcode is used to position the pointer to desired position in file and
> the ReadE opcode to process the group of records.
>
> When the desired group exists, the routine works fine.
>
> When the desired group doesn't exists, so SetLL opcode points the file
> before the next group, subsequent ReadE opreration returns the records from
> next group even the KLIST values requested nonexisting group! So result
> indicator NR should be on, but it is not until the not wanted group ends!
>
> Please help.
IIRC you need only install the HIPer PTFs for V4R3M0; probably also available
via the DB Fixpack for that release -- SF99103.
Regards, Chuck
All comments provided "as is" with no warranties of any kind whatsoever.
ohscjp
do the reade loop if the indicator is on...
Charles R. Pence a écrit dans le message <3711F447...@vnet.ibm.com>...