RE: Record Lock on Read from IFS file?

Chuck,

I think I understand you better now. The default view of WRKJOB
OPTION(*PGMSTK) does not show the detail you are referring to, but
F21=Include expands the detail considerably. I overlooked that option,
mostly due to my own ignorance and inability to interpret what is shown
there.

I agree with your assertion that the circumvention is not a real answer. I
do intend to explore this further, time allowing, but getting this long
promised program functional for the customer was a priority.

Many thanks. I will share anything new I learn from my additional testing.

______________________
Scott Sanders
scott@xxxxxxxxxxxxxxxxxx

-----Original Message-----
From: rpg400-l-bounces@xxxxxxxxxxxx [mailto:rpg400-l-
bounces@xxxxxxxxxxxx] On Behalf Of CRPence
Sent: Tuesday, May 15, 2012 3:00 PM
To: rpg400-l@xxxxxxxxxxxx
Subject: Re: Record Lock on Read from IFS file?

On 15 May 2012 11:42, Scott Sanders wrote:

CRPence on Tuesday, May 15, 2012 10:13 AM wrote:

On 14 May 2012 15:30, Scott Sanders wrote:

Job status sits at TIMW, for as long as I am willing to wait.

That job status reflects a "coded wait", meaning that some code
deliberately has requested to pause the thread, awaiting a timed
event; e.g. sleep() or DlyJob had been issue. So rather than
reviewing the job "status", review the program "stack"; WRKJOB
OPTION(*PGMSTK) to see what code has requested a timed-wait.

That parallels my experience, as well. But I have stepped through
every line of code in these programs in debug, and have not
encountered any deliberate wait.

That is a strange approach, because the program stack could easily
inform. An as-described long-wait should easily be captured in a
WRKJOB OPTION(*PGMSTK). FWiW there was no intention on my part to
imply that the coded wait was in code to which there would be any
source\debug access; i.e. not implied to be /user/ code.

However, I also seem to recall seeing that status appear on a job
retrieving or sending a large file via FTP to a remote system.

I suspect not. More likely DEQW, SELW, or some other non TIMW wait.
But taking an actual WRKJOB OUTPUT(*PRINT) instead of from memory is
helpful to enable referring back.

This lends weight, at least in my mind, to the thought of splitting
the FTP and file processing functions into separate jobs. Not just
separate programs in the same stack.

Of course that, a circumvention, just ignores what is the ailment;
i.e. akin to treating a symptom, but not the addressing the disease or
its preventive(s).

<<SNIP>>

Regards, Chuck
--
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.