All, This whole thread has really confused me. Why the blazes do we think that SMS is preventing multi volume allocation when it looks to me that it is regular, normal, everyday allocation that is preventing multi volume allocation. Ignoring guaranteed space attribute, if you allocate a dataset with UNIT=(3390,3) using IEFBR14 all you will ever get is the primary space requested on a single volume. That's it, end of story whether the dataset is SMS managed or not. One to five empty extents on one volume. When you subsequently open and write to the dataset allocation has no knowledge of the original allocation of 3 volumes because you have done nothing to generate candidates in the catalog. You will write to it on the subsequent open and ABEND when you hit 16 extents on the first volume. That's how allocation works. If you want the dataset to overflow to 3 volumes then you must specify UNIT=(3390,3) in the JCL of the job step that actually writes to the dataset. This works whether it is SMS managed or not. Ron Sent via the Samsung Galaxy S®6 active, an AT&T 4G LTE smartphone-------- Original message --------From: Neil Duffee <nduf...@uottawa.ca> Date: 4/8/2016 05:06 (GMT+08:00) To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: [IBM-MAIN] How force dataset non-SMS? Caveat: daily digestion causes response delays...
Charles: I'm gonna go back over some territory to make sure I understand the *original* purpose. I *believe* you simply want a multi-volume dataset for SMF17 (Scratch/delete) work in the simplest manner possible. It can be SMS or no, can be any dsOrg, & doesn't require large space. (or any records, even) All good so far? Presuming a yes, I point out that PS files (not EXT) max@16 extents/volume. Given that, combine IEBDG (I dropped the suggesting author, sorry), [1] with Tom's SPACE=(TRK,(1,1)),UNIT=(SYSALLDA,5) or an SMS DataClas that has VolCount/DynVolCnt>1. You should get 16 tracks on each volume until x37 for volume count. I'd even go with LRECL>29k (0.5 track length) to get 16 records per volume. The creating job should take more time to start/stop than output the necessary bytes. [1] You could probably IDCAMS PRINT something simple as well if you use the large record size. --------> signature = 8 lines follows <-------- Neil Duffee, Joe Sysprog, uOttawa, Ottawa, Ont, Canada telephone:1 613 562 5800 x4585 fax:1 613 562 5161 mailto:NDuffee of uOttawa.ca http:/ /aix1.uOttawa.ca/ ~nduffee "How *do* you plan for something like that?" Guardian Bob, Reboot "For every action, there is an equal and opposite criticism." "Systems Programming: Guilty, until proven innocent" John Norgauer 2004 "Schrodinger's backup: The condition of any backup is unknown until a restore is attempted." John McKown 2015 -----Original Message----- From: Charles Mills [mailto:charl...@mcn.org] Sent: April 6, 2016 11:36 Subject: Re: How force dataset non-SMS? Still the same answer. I want a multi-volser dataset and SMS is getting in the way of that. Pure local test. No customers in the problem set. -----Original Message----- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf Of Paul Gilmartin Sent: Wednesday, April 06, 2016 10:15 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: How force dataset non-SMS? On 2016-04-06, at 00:54, Vernooij, CP (ITOPT1) - KLM wrote: > ... Why do you need a non-managed dataset? ... > I might imagine tesing. The OP might be operating as an and desiring to test code that shoulc work in any possible customer environment, either SMS or non-SMS. ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
