MVSFORUMS.com

lal · Beginner Joined: 21 Oct 2003 Posts: 70 Topics: 25

Hi Friends,
I have a requirement where-in I need to copy the dataset which is saved in the specific volumes(40 different volumes) during our daily cycle for the last 40 cycles. .
Note:- There are only 40 versions of dataset in 40 volumes.
For Ex:-
PROD.MK.DLY is a dataset which will be created on a daily basis and stored in a specfic volume. I need to copy this dataset for the last 40 cycles (ie which is stored in 40 different volumes) into sequential file.

I can specify 40 different VOL=SER names in SORTIN and can achieve the same but instead of specifying 40 different SORTIN's can I have some generic type like VOL=SER=* or something.

Thanx in advance..

Regards,
Lal

Phantom · Posted: Mon Dec 27, 2004 1:12 am Post subject:

Lal,

I don't think you need to specify all the 40 volumes in ur JCL. By default, JCL will pickup the data from all the volumes it resides in. Remove the VOL=SER parameter from your JCL and try submitting the job.

PS: I have done it many times and it worked for me. So, it shouldn't be a problem.

Thanks,
Phantom

lal · Beginner Joined: 21 Oct 2003 Posts: 70 Topics: 25

Hi Phantom,
Thanx for the response but still your solution doesn't works. I had checked it previously without giving the VOL=SER it just picks the current dataset not the earlier ones and just to make sure I tried it once again and got the same results. So just picking the current one not the earlier ones.

Regards,
Lal

taltyman · Posted: Mon Dec 27, 2004 11:31 am Post subject:

Why not use GDG instead of what appears to be uncataloged datasets?

Phantom · Posted: Tue Dec 28, 2004 12:18 am Post subject:

taltyman,

Was my suggestion wrong ? When it works for me, why is Lal saying that it doesn't work for him ? Many times, I tried to extract data for a particular account no. from a huge tape dataset which resides in more than 6 volumes. I usually don't provide the VOL=SER parameter. When the job is running I could see the volumes being referred one by one. After the job completes I can see all the 6 volume nos being displayed in the JES messages in my spool.

Am I missing something or is it Intallation dependant or as you said is this a problem with Uncataloged datasets ?

Thanks,
Phantom

taltyman · Posted: Tue Dec 28, 2004 8:26 am Post subject:

Phantom, I think your solution is for a single dataset cataloged across mutliple volumes, that's very common for potentialy large datasets. In that case the catalog provides the volsers as needed for your job. His problem "sounds" like he has 40 unique datasets on 40 different volsers. The only way I know of to have 40 datasets with the same name on 40 different volsers is for at least 39 of them to be uncataloged. Lal needs to provide the details.

Phantom · Posted: Thu Dec 30, 2004 1:53 am Post subject:

Taltyman,

Thanks for the clarification. Sorry for not getting back sooner.

Thanks a lot,
Phantom