Gentlemen, This is all fine a good, but why check when a job is submitted? Why not check before and eliminate the problem (and potential delay in batch)?
Mitch McCluhan, Modernization Analyst IBM GBS/AMS/BAM Consultant -----Original Message----- From: Robert A. Rosenberg <[email protected]> To: IBM-MAIN <[email protected]> Sent: Mon, Dec 8, 2014 9:51 pm Subject: Re: Ancient IEFUSI At 09:32 +0000 on 12/08/2014, Rob Scott wrote about Re: Ancient IEFUSI: >In the past I have used JES2 exit 6 to fail jobs that specify the >TIME keyword - there is a supplied and supported routine "XINTKEY" >that you can $CALL to extract specific JCL parameters. > >Other methods include using JES2 exit 2/3 or IEFUJV to >unconditionally strip any TIME keyword from JCL statements therefore >allowing the jobclass defaults to be enforced. > These methods show how to detect the use of a TIME parameter in the JCL. Based on the way that IEFUSI and IEFUTL are used (as described in the quote below) what you can do is set the CEPUCOM flag if you see a TIME Parm in IEFUJV. Since you can detect and remove TIME= there you can just detect it (leaving it there) and set the flag for later checking. > >Rob Scott >Lead Developer >Rocket Software >77 Fourth Avenue . Suite 100 . Waltham . MA 02451-1468 . USA >Tel: +1.781.684.2305 >Email: [email protected] >Web: www.rocketsoftware.com > > >-----Original Message----- >From: IBM Mainframe Discussion List >[mailto:[email protected]] On Behalf Of Andrew Metcalfe >Sent: 07 December 2014 15:07 >To: [email protected] >Subject: Re: Ancient IEFUSI > >One of the reasons our IEFUSI examines the SCTX is to determine >whether the step time (SCTXSTL) differs from the JES2 JOBCLASS >setting. If so the assertion is that the user has coded TIME= on the >EXEC card. IEFUSI sets a bit in the CEPUCOM area which is >interrogated by IEFUTL when deciding whether to grant CPU time >extension. If the user coded TIME= on the EXEC card he gets the >bullet rather than an extension. > >How might I best determine (in IEFUTL) whether the user coded TIME=, >so that I can maintain the same behaviour. > >Thanks > >Andrew ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
