Posted: Wed Oct 06, 2004 9:09 am Post subject: GSAM datasets for datasets allocated in TAPE
Hi,
I have a program which uses GSAM datasets. These GSAM datasets are used for output purpose only. So what I did is, in 2nd step I allocated the GSAM datasets and in the 3rd step I used them.The job is working perfectly fine. (I allocated the Datasets in DASD).
Due to some memory constraints at our shop, we decided to allocate the datasets in TAPE rather than DASD. When we did this, Job is abended saying Dataset not found. (The dataset corresponding to GSAM - DDFB0001).
Do we need to take any special considerations while allocating a dataset under TAPE when used as GSAM dataset?
Here is the DBD definition of my GSAM
Code:
TITLE 'ASSEMBLE OF DBDNAME=DDFB0001 '
DBD NAME=DDFB0001, C
ACCESS=(GSAM,BSAM), C
PASSWD=NO, C
VERSION= DATE 07/09/03 TIME 16.31
DSG001 DATASET DD1=DDFB0001, C
DD2=DDFB0001, C
SIZE=(0), C
RECFM=FB
DBDGEN
FINISH
END
Joined: 26 Nov 2002 Posts: 12376 Topics: 75 Location: San Jose
Posted: Wed Oct 06, 2004 9:31 am Post subject:
Siva,
You cannot pre-allocate a tape file using IEFBR14. IEFBR14 does not write a EOF marker. so your step020 allocation is a mere waste. However you can allocate using IEBGENER, but you don't know which volume it is allocated to.
So the best way is to write to tapes in a BMP process is that you allocate it in the BMP step itself
Joined: 29 Jun 2004 Posts: 106 Topics: 41 Location: Bangalore, INDIA
Posted: Fri Nov 12, 2004 1:18 am Post subject:
Hi Kolusu,
Due to interest in learning IMS-DB I want to know regarding
Quote:
I was unable to use BMPPROC
What could be the reasons for unable in using the BMPPROC. Please can you clarify my doubt. _________________ Best Regards,
----------------
Rammohan Pabba
Software Engineer
Joined: 26 Nov 2002 Posts: 12376 Topics: 75 Location: San Jose
Posted: Fri Nov 12, 2004 6:35 am Post subject:
Quote:
What could be the reasons for unable in using the BMPPROC. Please can you clarify my doubt.
ramu_mohan21,
I am not sure as to why siva is unable to use a proc. My guess would be he might be running the program in the JCL itself without using a PROC. or he does not want to change the proc.
Joined: 19 Dec 2004 Posts: 9 Topics: 2 Location: Chennai, India
Posted: Sun Dec 19, 2004 7:21 pm Post subject:
Quote:
So the best way is to write to tapes in a BMP process is that you allocate it in the BMP step itself
As far as I know, allocating a dataset in the BMP step is a bad option. If the job fails for some reason, and we need to OS restart the job, then it creates problems. It would be better to use IEBGENER rather than allocating in the BMP itself. _________________ Thanks,
Rek
Joined: 26 Nov 2002 Posts: 12376 Topics: 75 Location: San Jose
Posted: Mon Dec 20, 2004 6:24 am Post subject:
Rek,
Quote:
As far as I know, allocating a dataset in the BMP step is a bad option. If the job fails for some reason, and we need to OS restart the job, then it creates problems.
You would not have any problem , if proper symbolics are passed for DISP parameter during a restart.
Quote:
If the job fails for some reason, and we need to OS restart the job, then it creates problems.
I would like to hear one of the problems you faced.
Joined: 03 Jan 2003 Posts: 550 Topics: 23 Location: Michigan, USA
Posted: Mon Dec 20, 2004 10:44 am Post subject:
BMPPROC is not a standard JCL procedure as supplied with IMS - The delivered procedure is IMSBATCH. Each shop then may modify this or create other versions as needed.
It is possible to allocate GSAM datasets in the JCL proc if the job scheduler can handle the disposition correctly (and generation if required).
Joined: 19 Dec 2004 Posts: 9 Topics: 2 Location: Chennai, India
Posted: Tue Dec 21, 2004 5:25 pm Post subject:
Kolusu,
Our Environment uses CA-11. It scratches the datasets when the disposition of that dataset in the job is (NEW,,). We had an instance, where CA11 deleted the dataset when we restarted from the BMP step which allocated the dataset, whereas it should not. _________________ Thanks,
Rek
Joined: 26 Nov 2002 Posts: 12376 Topics: 75 Location: San Jose
Posted: Wed Dec 22, 2004 5:21 am Post subject:
Quote:
Our Environment uses CA-11. It scratches the datasets when the disposition of that dataset in the job is (NEW,,).
rek,
You should never restart a BMP step with CA-7 restart. You need it to restart as a regular restart step with a restart card coded on the job card. Inorder to do that you need to add this line in your BMP step
Joined: 03 Jan 2003 Posts: 550 Topics: 23 Location: Michigan, USA
Posted: Wed Dec 22, 2004 9:48 am Post subject:
Kolusu,
We have an exit in CA-11 to avoid the scratch of the output files. We are able to modify the JCL (in the 254 library) with the correct restart command and can start from the abending step.
You are right though that without thsi exit, we would have problems.
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum
FAQ
Search
Quick Manuals
Register
Profile
Log in to check your private messages
Log in
