Re: journals -- MIDRANGE-L

"Some IBM guy" is pretty vague.  His approach is very system programmer-y.

One better way: write the label information to a journaled table with an
"added" timestamp and have the printing program update the record with a
"printed" timestamp.  This accomplished the same result, offers
visibility, and provides an easier interface for recovery.

I have plenty of experience working with journaling, including working with
extracted journal data, and I prefer to use other techniques to handle
asynchronous tasks.  I've used straight called programs, triggers, data
queues, logical files with EOFDLY, and logical files with program-controlled
waits at EOF.  These are not cases of "technique-shopping": there are
specific technical and application reasons for using one technique instead
of another.  For production applications, I don't need to dig through
journal receivers; one major reason is that journal receivers can be
detached and deleted.

However, a lot depends upon the characteristics and architecture of the
system driving the label printing program.  You can't begin to consider any
solution until you fully understand the problem.

-reeve


On 5/5/06, Katy Carter <kathy.carter@xxxxxxxxxxxxxxxxxxxxxx> wrote:


Hello everyone,



I need opinions.  I have been an RPG programmer for 20 years.   I still
see
things I've never seen before but this one doesn't make any since to me.



I have been programming for a manufacturing company about 1 year now.
The
company had some IBM guy to come in and develop the barcode label and
place
the record into production (MAPICS).    Before the label is printed it is
first journaled into a journal with the journal receiver etc....,  then
received from the journal and sent to a program that prints the bar code
label.



Now, I was always taught that journaling was only for backup and recovery
purposes,  which I know what some people will say,  that he designed it
this
way incase there was a disaster and the records could be received but,
wasn't there a better way to achieve this.   It appears to me that this
way
would be cumbersome,  eat up the DSAD of the machine and he is journaling
the record first  and receiving it back before he prints the label.
Shouldn't he have at least printed the label first before he journaled the
record.   I just want your all's opinion about this design.  Maybe there
is
something I don't see here but the journal sequence goes haywire all the
time and we have to reconstruct it.     EVERYBODY, ANYBODY GIVE ME YOUR
THOUGHTS ON THIS?????





Kathy Carter

Application Developer



--
This is the Midrange Systems Technical Discussion (MIDRANGE-L) mailing
list
To post a message email: MIDRANGE-L@xxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: http://lists.midrange.com/mailman/listinfo/midrange-l
or email: MIDRANGE-L-request@xxxxxxxxxxxx
Before posting, please take a moment to review the archives
at http://archive.midrange.com/midrange-l.