> the SYMBOLICRELATEd name must contain at least one substitutable symbol
Oh geez! Why? (Rhetorical question.) Well, that disqualifies it for my purposes. I could live with the ENQ restriction. It's a STEPLIB and everyone is DISP=SHR. Charles -----Original Message----- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf Of Paul Gilmartin Sent: Monday, August 29, 2016 5:05 PM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: Elementary dataset alias question On Mon, 29 Aug 2016 15:41:00 -0700, Lizette Koehler wrote: > >You can also look at SYMBOLICRELATE > DEFINE ALIAS (NAME(SYS1.FOOBAR) - > SYMBOLICRELATE('JOEBLOW.FOOBAR')) > "SYMBOLICRELATE" can be abbreviated "SYM". I avoid using abbreviations in a didactic context. I even say "ALLOCATE DATASET()", not "ALLOC DA()". >However, when a data set ALIAS of NON VSAM is created using DEFINE >ALIAS SYMBOLICRELATE and when a DELETE is done against the REAL data >set, the data set alias created using SYMBOLICRELATE is not deleted. >This is because the entry is not associated with any real data sets and it is >only a catalog entry. > The "however" suggests that you consider this a cautionary statement. Rather, I consider it useful. I can briefly delete the SYMBOLICRELATEd data set and rebuild it with no need to create the alias. However, there's a restriction not stated in the manual: the SYMBOLICRELATEd name must contain at least one substitutable symbol. Perhaps I need to submit an RCF. And another caution concerning ENQueues: At job initiation, the initiator attempts to ENQ only on the ALIAS name, waiting until contention is resolved. At step initiation the initiator attempts to ENQ on the RELATED name. Any failure due to contention is immediately fatal; no waiting for resolution because of the risk of deadlocks. ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
