I'd start by reviewing the allocation messages in the joblog for the "working" UNIT=WRK case. There's an IEF236I message that "starts" the allocation for the step, and that'll be followed by various other messages (IEF237I, IGD101I, among others) that indicate what device was allocated to each DD. There will also be disposition messages (for example, IEF285I if the data set was kept etc.) There might be a clue based on what device is being chosen. You could also look at the corresponding messages for the "failing" UNIT=WORK case, but given that it failed, you might not get as much information from that.
The "SMS is activated and esoterics go sideways" cases - at least the ones that I'm familiar with - usually involve the SMS ACS routines using the UNIT= value to make the request SMS-managed (in which case, the devices defined to the esoteric are ignored - in fact, the esoteric doesn't need to be defined.) Or a variation on that, where the ACS routines used to mark mark the request SMS-managed but didn't (maybe because somebody changed them) and the esoteric isn't defined so nobody knows what UNIT=XYZ is. Unfortunately, for these kinds of things, you have to dig into those routines to see what they're doing. I'd be a bit surprised if SMS was involved in the failing case, because I think IEF506I only applies to non-SMS-managed requests. But you might find something else, like SMS is managing the request in the working case. -Scott Ballentine ([email protected]) IBM z/OS Allocation Development On Thu, 18 Sep 2025 16:14:06 -0500, R Do <[email protected]> wrote: >"IEF506I - The system could not find enough storage volumes to satisfy the >request. An attempt will be made to allocate the data set to a private volume." > >My SMS exposure is limited albeit I can say that the SMS config that is in >play on this sandbox LPAR is barebones. > >There are sufficient volumes mounted STORAGE. The job is a BR14 asking for a >1 cyl dset allocation using a new esoteric called WORK. i.e. UNIT=WORK. Job >receives "IEF238D - Reply [DEVICE NAME] or 'CANCEL'". If I create an >identical esoteric called WRK w/ the exact same devices, it works. So >UNIT=WRK works albeit UNIT=WORK fails. i.e. WRK and WORK esoterics are >defined exactly the same. > >I have read other posts in the archive where SMS is activated and esoterics go >sideways however, there was not sufficient resolution information. > >Any ideas? > >---------------------------------------------------------------------- >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
