SYS1.MODGEN(IARRAX) RAXLVMEMLIMS DS CL1 Source of Address Space memory limit. * When RaxLvMemLimS=RAXLVUSI and * RAX_SMFLIMRas is 0, IEFUSI set the * MEMLIMIT. When RaxLvMemLimS=RAXLVUSI and * RAX_SMFLIMRas is non-zero, RAX_SMFLIMRas * contains info about the setting of * RaxLvMemlimit. @0DC RAXLVSMF EQU 1 MEMLIMIT set by SMF either in SMFPRMxx * or by use of SMF default value=0 @PHC RAXLVJCL EQU 2 MEMLIMIT set by the JCL @PHC RAXLVREG0 EQU 3 MEMLIMIT Unlimited based on * REGION=0 specification @PHC RAXLVUSI EQU 4 MEMLIMIT set by IEFUSI @PHC RAXLVOMVS EQU 5 MEMLIMIT set by UNIX OMVS segment @PHC RAXLVSETR EQU 6 MEMLIMIT set by UNIX setrlimit @PHC RAXLVSPW EQU 7 MEMLIMIT set by UNIX spawn @PHC RAXLVSETO EQU 8 MEMLIMIT set by UNIX * SETOMVS command @PHC RAXLVAUTH EQU 9 MEMLIMIT set by authorized * application modification @PHC RAXLVURG EQU 10 Special case of MEMLIMIT * getting set in IEFSMFIE * (IEFUSI set REGION size) @PHC
Jim Mulder z/OS Diagnosis, Design, Development, Test IBM Corp. Poughkeepsie NY "IBM Mainframe Discussion List" <IBM-MAIN@LISTSERV.UA.EDU> wrote on 11/19/2018 11:12:18 AM: > From: "Sean Gleann" <sean.gle...@gmail.com> > To: IBM-MAIN@LISTSERV.UA.EDU > Date: 11/19/2018 04:28 PM > Subject: Re: Region size for OMVS tasks > Sent by: "IBM Mainframe Discussion List" <IBM-MAIN@LISTSERV.UA.EDU> > > Interesting stuff, here... > Before making any changes, analysis of the type-30 data showed > SMF30RGN=56623104 (54M) and SMF30MES - the source of MEMLIT - as '1', which > apparently means that MEMLIMIT is set by SMF. Also, SMF30MEM was coming > back as 1280M - the same as the MEMLIMIT in SMFPRMxx. So far, so good. > So I set the ASSIZEMAX for my own profile to 2G and logged off-n-on. No > change in the type-30 values > Then I set the associated MEMLIMIT value to 2G & logged off and on again. > This time, SMF30MEM is coming back as 2048M. Great! I can modify the > MEMLIMIT value and perhaps solve my problem. I'll have to re-engage one of > the developers for that, could take a bit of time. > But a curiosity has cropped up along the way... > Now, my SMF30MES is coming back as '5' - which is undocumented, according > to the System Management Facilities manual I'm using (SA22-7630-26 - which > I believe is the latest, though I stand to be corrected). Apparently, the > values for SMF30MES can be one of 1, 2, 3, 4 or 10 (decimal), so where the > '5' is coming from is a mystery. I'm not sure it's important and I shan't > chase it up with IBM, but thought it worth mentioning. > > Sean ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
