Re: RRN and Database Files

Hi Bryce

The file the programs running over isn't 2.147 billions records large,
that's the maximum size that I keep seeing bandied around. We do have
some billed history detail files with 2.02 billion records in, but
they're not part of this problem. The question on maximum file size was
asked because if a file could have more records than that then the size
of the RRN field in the INFDS wouldn't be suitable and I'd have to
rethink my approach.

For the connection pools I've been told that the pool has a number of
available connections and each job grabs the first one available and
uses that. They don't share them concurrently, but because the program
doesn't set on *INLR the program data still exists when the program is
called again.

An example of the issue is:

Job_1 uses Connection_1 and Job_2 uses Connection_2 to call program to
look at same data. At the end of the process the file pointer is at
transaction 10 for account 1. Both jobs drop their connections.

Job_2 pages down and picks up Connection_1 to load second page of data.
At the end of the process the file pointer is at transaction 20 and the
connection gets dropped.

Job_1 now pages down and picks up Connection_1, but this now loads the
third page of data.

The issue is because the program stays open so I need to be able to
position the file back to its previous state (and reset any work/save
variables) the last time the program was called from the particular job.

Cheers

Jonathan


Bryce Martin <BMartin@xxxxxxxxxxxx> wrote :

Why do you have different users using the same connection? Why
wouldn't you implement connection pools and then make it so that each
job runs its own instance of the programs so that you don't have this
another thing.... 2.147 billion records? No wonder you are having
collision. And performance issues. Any DB file with that many
records is going to take forever to processes. I think that instead
should take a look at reworking your java code so that you can have
multiple users. Its a transaction file you are searching over so its
obviously just an inquiry. There is no reason why this can be run as
independent instances.


Thanks
Bryce Martin
Programmer/Analyst I
Ext. 4777



"Jonathan Mason " <jonathan.mason@xxxxxxxxxxxxxxxx>
Sent by: rpg400-l-bounces@xxxxxxxxxxxx
01/09/2009 07:50 AM
Please respond to
RPG programming on the IBM i / System i <rpg400-l@xxxxxxxxxxxx>


To
<rpg400-l@xxxxxxxxxxxx>
cc

Subject
RRN and Database Files






of modifying your RPG code you We have an RPG400 program that reads
then returns consolidated data back to it's calling Java client. Each
through a JDE transaction file and time the program returns data to
the Java client it remains open for the next call and accepts a
needs to restart from.

Because of performance issues the call from Java was modified recently
so that instead of retrieving all transactions for a customer it now
retrieves blocks of 50 or so transactions.

parameter that indicates where in the program it This is now causing a
further problem where multiple users are accessing the same customers
used by both jobs in turn.

I need to modify the RPG program so that it positions the file back to
the last record read by the Java client regardless of whether the
program instance has been used by any other client.

transaction list and the same Websphere connection is I'm looking at
using the RRN and passing this back to Java when I return the data and
has been a change in client. I rarely use RRN processing, so this has
raised some questions:

then chaining with this when the program restarts and there 1) If the
logical file is defined as keyed in the program can I chain to it
to the logical using the key values picked up from the physical?

2) I can get the RRN from the INFDS, but looking through the manuals,
using RRN or would I have to read the physical by RRN and then chain
Google and the archives I keep seeing that the maximum size of a 4
binary (integer) field is 2,147,483,647 (i.e. 2.147 billion records).
byte In RPG400 I have to use the INFDS to get the RRN for the record,
2.147 billion still the maximum number of records I can have in a file
or can I have more?

3) If the file can hold more than 2.147 billion records can I get the
RRN for these "additional" records within the RPG400 program or is the
larger RRN only available through the RECNO keyword in RPG IV?

Thanks in advance

Jonathan




_______________________________________________________
This message was sent using NOCC v1.14 webmail software
_______________________________________________________




--
This is the RPG programming on the IBM i / System i (RPG400-L) mailing
list
To post a message email: RPG400-L@xxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: http://lists.midrange.com/mailman/listinfo/rpg400-l
or email: RPG400-L-request@xxxxxxxxxxxx
Before posting, please take a moment to review the archives
at http://archive.midrange.com/rpg400-l.



but is --- This message (including any attachments) is intended only
for the use of the individual or entity to which it is addressed and
may contain information that is non-public, proprietary, privileged,
confidential, and exempt from disclosure under applicable law. If you
are not the intended recipient, you are hereby notified that any use,
dissemination, distribution, or copying of this communication is
strictly prohibited. If you have received this communication in error,
--
please notify us and destroy this message immediately. --- This is the
To post a message email: RPG400-L@xxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: http://lists.midrange.com/mailman/listinfo/rpg400-l
or email: RPG400-L-request@xxxxxxxxxxxx
Before posting, please take a moment to review the archives
at http://archive.midrange.com/rpg400-l.

Jonathan Mason
iSeries Consultant
www.astradyne-uk.com


_______________________________________________________
This message was sent using NOCC v1.14 webmail software
_______________________________________________________