home *** CD-ROM | disk | FTP | other *** search
- Comments: Gated by NETNEWS@AUVM.AMERICAN.EDU
- Path: sparky!uunet!gatech!paladin.american.edu!auvm!VCCSCENT.BITNET!SOMITCW
- Message-ID: <IBM-MAIN%92073011134648@RICEVM1.RICE.EDU>
- Newsgroups: bit.listserv.ibm-main
- Date: Thu, 30 Jul 1992 12:10:08 EST
- Sender: IBM Mainframe Discussion list <IBM-MAIN@RICEVM1.BITNET>
- From: SOMITCW@VCCSCENT.BITNET
- Subject: MVS Allocation Recovery
- Lines: 69
-
- On Wed, 29 Jul 1992 23:56:59 CDT, NPAPE@VM1.TUCC.TRINITY.EDU said:
- >How does your shop handle a production job which has gone into MVS
- >allocation recovery for a disk drive? Our applications programming
- >manager wants to be called in all situations before any action is taken.
-
- The best approach now is to have the operator cancel the job.
- Your applications programming manager can only watch the operator
- reply CANCEL. In the past the operator might have been able to
- recover. In the future there will hopefully again be recovery.
-
- A brief history of disk allocation recovery.
-
- Over 15 years ago, the system would list offline devices and
- pending offline devices, and give the operator a reply ID.
- The operator would issue unload commands, change mount use,
- sometimes vary disk on or offline, and reply the address of SYSRES
- or any other device to the reply. The system would process the
- unload commands and vary commands and then retry allocation.
- ( Note: Back then, we used pending offline status to block
- allocation until IBM took the very useful feature away. )
-
- About 15 years ago IBM made some unfriendly changes. They changed
- allocation recovery to only accept a reply of an offline device that
- was in the list originally displayed and hadn't been replied before
- and the job must be looking for public space or a private request
- must be satisfied with an offline request plus a few other quirks.
- Allocation recovery for disk was broke. IBM was requested to fix it
- and they did. Their response was to remove the words 'Allocation
- Recovery' from the console messages.
-
- ( This was about the same time that IBM decided that tape volumes
- should remain on the tape drives when jobs ended. They finally
- fixed the system back.
- P.P.S. If IBM is listening, thanks for fixing it back, but
- adding a JCL option to leave some tapes up and have others unload
- would have been better. )
-
- ( This was about the same time that they decided to drop support
- for CVOL catalogs in IEHPROGM. They finally fixed the system back.
- P.P.S. If IBM is listening, thanks for fixing IEHPROGM back,
- at the time we really needed it. )
-
- ( That was about the same time that they decided to disallow tape
- swaps to pending offline tape drives so you could no longer cause the
- bad from drive to be offline as UCB's swap in memory. Never fixed. )
-
- About 2 years ago IBM received several complaints because with
- VATLST finally fixed, even the once in a blue moon theoretical
- successful allocation recovery for public work space was impossible.
- IBM decided to allow customers a chance to change the mount use of
- a disk volume and then reply. We put their fix on last year here
- but it didn't work. We are working on getting maintenance done again.
- If it still doesn't work, I will call IBM and ask them to try again.
- Why can't IBM just drop their bank of offline disk requirement and
- allow the operator to reply RETRY, WAIT, or CANCEL? Of course they
- could go the other way and add HOLD or NOHOLD complications.
- ( P.P.S. If IBM is listening, thanks for fixing VATLST and thanks
- for attempting to allow disk allocation recovery again. It has
- been a long time. )
-
- While I'm complaining, IBM needs to fix the termination message
- NOT RECATLG. The message means that IBM saved overhead of recata-
- loging a MODded data set because the volume list didn't change.
- The NOT RECATLG messages scares and confuses the users.
- Better messages would be RECATALOG NOT NECESSARY or RECATALOG NOT
- NEEDED. Or at the worst, IBM could add the overhead and recatalog
- the data set and change the message to say RECATALOGED.
- If this is still a problem after a years worth of maintenance that
- we are applying has been completed, I will open a PMR with IBM.
-