<SNIPPAGE> I don't like doing the above, but management basically wants a rule of "any job which will execute more than ... minutes must be scheduled by Production Control via CA-7." Of course, a rule without enforcement is not a rule. It is a request. <SNIPPAGE>
This can certainly be coded up. However, wall clock time can vary based on many factors. 1) Human response to a mount request. 2) Virtual tape and/or HSM recalls 3) comparative load on the machine 4) allocation recovery loops if mountable devices get over committed. . . . IMO, a lousy way to run a railroad, but it's their railroad. My suggestion would be to "Force' $ESTIME to the value desired, and terminate after a fudge factor e.g. $HASP308 pops up after 1 hr. check to see if 1hr, 15 minutes have elapsed and then cancel. If < 1 hr, 15 min go back to sleep until the next time. When the fudge factor is finally exceeded, boom! I predict this will last until the 1st user gets burned and cry's to their management. A fitting topic for April 1st. -----Original Message----- From: IBM Mainframe Discussion List [mailto:[email protected]] On Behalf Of Thompson, Steve Sent: Wednesday, April 01, 2009 1:01 PM To: [email protected] Subject: Re: JES2 exit? -----Original Message----- From: IBM Mainframe Discussion List [mailto:[email protected]] On Behalf Of John McKown Sent: Wednesday, April 01, 2009 12:35 PM To: [email protected] Subject: JES2 exit? <SNIPPAGE> I don't like doing the above, but management basically wants a rule of "any job which will execute more than ... minutes must be scheduled by Production Control via CA-7." Of course, a rule without enforcement is not a rule. It is a request. <SNIPPAGE> So any and all jobs that execute 1 second beyond get cancelled, right? I can think of two system issues that can cause heartburn. Should the system have an enque hang, or allocation (as in it being single threaded for ALL address spaces) get stuck for some reason, all batch jobs subject to this rule could get "executed." Now for the busy situation that causes this problem: What happens when the tape system you have has all its tape heads tied up when a large data set is needed has been migrated off to tape? Do all the jobs that are waiting for restores get charged with the wall time it takes to get their files restored? Just a few items that management, who may know nothing about how real processing takes place, may need a small bit of education about. Regards, Steve Thompson -- Opinions expressed by this poster may not be those of poster's employer. -- ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
