RE: not allowing operator from end a specific job

midrange-l-request@xxxxxxxxxxxx wrote:

>   5. RE: not allowing operator from end a specific job (Evan Harris)
>
>the subsystem will wait for a specified time limit (which I have a feeling 
>might just be *NOMAX,,,) if you use end option *controlled (from memory 
>this is the default) but ending it *immed will normally cause the jobs 
>running in it to end in fairly short order. 

Agreed. However, the jobs still must end first, AFAIK. It won't matter if 
*CNTRLD or *IMMED; if a job refuses to end for some unknown reason, the 
subsystem won't "end".


>Having said this I have seen jobs that did not end in response to an ENDSBS 
>command but ended quite happily when the job itself was manually ended by a 
>specific ENDJOB request. I have not taken the time to work through the 
>various reasons this might occur.

Nor have I. The condition occurs so rarely and there are no indications of the 
cause, so it's practically impossible to replicate.


>I was kind of speculating that removing *JOBCTL would then remove the 
>ability to use ENDSBS anyway (no manuals to hand and too late to go hunting)

Heh, yeah, there's the catch. I'd forgotten that ENDSBS requires *JOBCTL 
anyway, so that kind of makes it moot.

Tom Liotta