Author: steve-myers
Posted: Tue Mar 14, 2017 7:12 pm (GMT 5.5)
I'm pretty sure Mr. Porcelan's description is correct. I think I have seen this behavior myself. But is there any formal statement this is the correct behavior? To my mind it violates the concept of DISP=SHR. I always thought the running DISP=SHR jobs should block the DISP=OLD job, but the queued DISP=OLD job should not block new DISP=SHR jobs.
I grant there is also a fair argument that the queued DISP=OLD job should block new DISP=SHR jobs as Mr. Poreclan and I both observed, but is this actually documented? I recall wondering about this something like 20 years ago, but I don't think I ever researched it. I can say the antique JCL Reference manual I just looked at seems to be silent about this.
Posted: Tue Mar 14, 2017 7:12 pm (GMT 5.5)
I'm pretty sure Mr. Porcelan's description is correct. I think I have seen this behavior myself. But is there any formal statement this is the correct behavior? To my mind it violates the concept of DISP=SHR. I always thought the running DISP=SHR jobs should block the DISP=OLD job, but the queued DISP=OLD job should not block new DISP=SHR jobs.
I grant there is also a fair argument that the queued DISP=OLD job should block new DISP=SHR jobs as Mr. Poreclan and I both observed, but is this actually documented? I recall wondering about this something like 20 years ago, but I don't think I ever researched it. I can say the antique JCL Reference manual I just looked at seems to be silent about this.