FWIW, not a good justification but perhaps some light on the topic: a JFCB has a fixed-length 44-byte dataset name area JFCBDSNM. Obviously, that will not accommodate most zFS filenames. When a DD statement references a zFS file, the associated JFCBDSNM contains '...PATH=.SPECIFIED...'. (If it were up to me I might have put the first or the last 44 characters of the filename there and set bits to indicate (a.) zFS and (b.) truncation when relevant. IBM did not ask me. Once again.)
So the message writer was not so much intentionally obfuscating the path name as simply (dumbly?) displaying the "dataset name" from the JFCB. Charles -----Original Message----- From: IBM Mainframe Discussion List [mailto:[email protected]] On Behalf Of Paul Gilmartin Sent: Thursday, February 12, 2015 10:10 AM To: [email protected] Subject: IKJ56228I vs. IEFA107I BPXWDYN() tells me: ***** in -sh -iLm ***** User@MVS:500$ rexx "say bpxwdyn( 'alloc path(''/foo/bar'') msg(2)' )" IKJ56228I PATH /foo/bar NOT IN CATALOG OR CATALOG CAN NOT BE ACCESSED -32745 ... a reasonably informative diagnostic (but I find the reference to "CATALOG" misleading -- I doubt that a catalog search was ever attempted. IIRC, I went to PMR on this long ago. WAD.) But in JCL for the same path: 46 //SMPNTS DD PATHOPTS=ORDONLY, // PATH='/foo/bar/.' ... IEFA107I C07SETUP NTSTEST SMPNTS - DATA SET ...PATH=.SPECIFIED... NOT FOUND IEF272I C07SETUP NTSTEST - STEP WAS NOT EXECUTED. Why is the pathname obfuscated in the message? Don't both BPXWDYN and Initiator invoke Allocation ang get similar replies? Why doesn't Initiator just show the pathname? ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
