Initiator HSM HRECALL/HDELETE
The fact that the MVS Initiator and HSM don't collaborate much is an old story which dates back to the early 1980s IIRC. Members RECALL and CLEANUP in FILE183 on the CBT tape (available here: http://gsf-soft.com/Freeware/ ) were developed 20 years ago to avoid HSM to perform useless RESTORE actions and speed-up useful ones. I hope the recently-updated Initiator knows that IEFBR14 is the 4-byte IBM version, not a special utility program written locally and also called IEFBR14, something I saw at-least once before. The MVS JCL was significantly enhanced in the late 1980s and early 1990s and I wish a number of JCL-related HSM issues were solved at the time. Why did it take 25 years to skip HRESTORE when PGM=IEFBR14 and DISP=(MOD,DELETE) ? How many millions of HSM ML2 tapes had to be mounted for nothing ? -- Gilbert Saint-Flour GSF Software http://gsf-soft.com/ -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Initiator HSM HRECALL/HDELETE
On Fri, 21 Aug 2009 16:04:46 +0200, Gilbert Saint-Flour wrote: > >I hope the recently-updated Initiator knows that IEFBR14 is the 4-byte IBM >version, not a special utility program written locally and also called >IEFBR14, something I saw at-least once before. > Discussed here several weeks ago, when an IBM employee replied, no, it doesn't know. >The MVS JCL was significantly enhanced in the late 1980s and early 1990s and I >wish a number of JCL-related HSM issues were solved at the time. Why did it >take 25 years to skip HRESTORE when PGM=IEFBR14 and DISP=(MOD,DELETE) ? >How many millions of HSM ML2 tapes had to be mounted for nothing ? > RESTORE? RECALL? Whatever. And I tried an IEFBR14 with DISP=SHR,UNIT=(,,DEFER) (z/OS 1.7). The data set was recalled regardless. Regrettable; not recalling until OPEN would be good notional agreement with DEFER. I'm curious: in days of yore, did DEFER actually defer mounting demountable DASD? -- gil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Initiator HSM HRECALL/HDELETE
Paul Gilmartin wrote: On Fri, 21 Aug 2009 16:04:46 +0200, Gilbert Saint-Flour wrote: I hope the recently-updated Initiator knows that IEFBR14 is the 4-byte IBM version, not a special utility program written locally and also called IEFBR14, something I saw at-least once before. Discussed here several weeks ago, when an IBM employee replied, no, it doesn't know. The MVS JCL was significantly enhanced in the late 1980s and early 1990s and I wish a number of JCL-related HSM issues were solved at the time. Why did it take 25 years to skip HRESTORE when PGM=IEFBR14 and DISP=(MOD,DELETE) ? How many millions of HSM ML2 tapes had to be mounted for nothing ? RESTORE? RECALL? Whatever. And I tried an IEFBR14 with DISP=SHR,UNIT=(,,DEFER) (z/OS 1.7). The data set was recalled regardless. Regrettable; not recalling until OPEN would be good notional agreement with DEFER. I'm curious: in days of yore, did DEFER actually defer mounting demountable DASD? -- gil I don't think anyone ever answered this. DEFER did work for mountable DASD. Mountable DASD allocation wasn't very smart though. It always selected the first available drive (not permanently mounted and not allocated). At slow times, this could lead to a pair of disks being repeatedly mounted on the same drive. ASP may have modified this, I don't know, we never ran it. I made a mod to store the time at unalloc, and at alloc if the volume wasn't already mounted it would search for an available drive with the oldest time stamp. -- Richard -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Initiator HSM HRECALL/HDELETE
Richard Peurifoy wrote: Mountable DASD allocation wasn't very smart though. It always selected the first available drive (not permanently mounted and not allocated). At slow times, this could lead to a pair of disks being repeatedly mounted on the same drive. ASP may have modified this, I don't know, we never ran it. It's a no win situation. The SVS system had a sysgen option, that may even have been the default, to rotate allocations. By then we didn't have mountable disks anymore, so I don't know whether it was tape only. And I had the opposite problem - on dedicated test time, unless I used a MOUNT command, my debug jobs would call for the tape to be remounted on the next drive over. Gerhard Postpischil Bradford, VT -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Initiator HSM HRECALL/HDELETE
On Mon, 31 Aug 2009 19:38:15 -0400, Gerhard Postpischil wrote: > >It's a no win situation. The SVS system had a sysgen option, >that may even have been the default, to rotate allocations. By >then we didn't have mountable disks anymore, so I don't know >whether it was tape only. And I had the opposite problem - on >dedicated test time, unless I used a MOUNT command, my debug >jobs would call for the tape to be remounted on the next drive over. > I've experienced that, even between steps of a single job. RETAIN didn't seem to help. It was best if all eligible drives except one were offline. -- gil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
