Re: scheduled job to run every n minutes?

[Tom Liotta <qsrvbas@xxxxxxxxxxxx>]

Jonathan Mason wrote:

>   The program can be
> shut down cleanly by setting a data area with the value "*SHUTDOWN" which is
> checked before the FTP script is processed.

One _possible_ alternative is simply to have the program check the 
"End status" of the job. Use the Retrieve Job Information (QUSRJOBI) 
API with format JOBI0600 and check position 71 (offset 70) of the 
returned structure; a '1' means ENDJOB was requested either against 
the job itself or because the system or subsystem is being ended.

This takes more thought because a *dtaara value can be checked much 
later, but "End status" must be checked within the *CNTRLD time 
limit. (Both alternatives work essentially the same for ENDJOB *IMMED.)

OTOH, there's no need to maintain an additional object and there's 
no need to take an additional special action if the system/subsystem 
must be brought down for some emergency. Simply end the subsystem, 
rather than remember to set the *dtaara, wait for the job to 
respond, and then end the subsystem -- and then remember to UNset 
the *dtaara!

Tom Liotta


> -----Original Message-----
>
> I want to thank all of you for the suggestions. My preference would be to 
> get the Advanced Scheduler but my boss does not want to spend the money on 
> it (yet). So it looks like the suggestion for the CL program that is 
> launched at the beginning of the day by the scheduler and then suspends 
> itself will be what I will try to do.
>
> I do have one concern. How can I prevent the job log from this long running 
> job from growing to an enormous size? I need to see the output from the FTP 
> command each time in case of a failure. How should I handle this?