ADRDSSU- Copy generation to generation

devmisra · New User Joined: 29 Jan 2009 Posts: 39 Location: India

Hi,

I want to copy data from generation to generation.
Source GDG is AAAAAAAA.BBBB.MONTHLY.VENDOR
Dest GDG is AAAAAAAA.BBBB.MONTHLY.VENDOR.BK

i tried to copy generation to generation thru PGM=ADRDSSU
but it failed in prod as source dest is having 4 qualifier and dest gdg having 5 qualifier.

here is the jcl

devmisra · New User Joined: 29 Jan 2009 Posts: 39 Location: India

Seems like i wrongly copied the error mask.

Please Ignore the previous error message,

Correct error msg is,

expat · Posted: Wed Jul 28, 2010 1:33 pm

Unfortunately the renaming capabilities of DFdss are not brilliant, in fact, they are fairly basic at best.

You can only change the HLQ using wildcards with Dfdss, so you will need to manually specify the output dataset names.

devmisra · New User Joined: 29 Jan 2009 Posts: 39 Location: India

Thank you Expat.

But i have to copy around 200 generations. Is there any other way to achieve this? The jcl will run in production.

expat · Posted: Wed Jul 28, 2010 1:52 pm

Only with a workaround, like CSI to list the generations and then build the rename cards.

PeterHolland · Posted: Wed Jul 28, 2010 1:57 pm

What happens if you add REPLACEU - after RENUNC ?

expat · Posted: Wed Jul 28, 2010 2:02 pm

Unfortunately the same result. I did try that before posting

PeterHolland · Posted: Wed Jul 28, 2010 2:06 pm

Expat, im so sorry about that

Then i guess we are exhausted with probable solutions.

expat · Posted: Wed Jul 28, 2010 2:08 pm

Aha .................. going through my archive JCL library a possible solution raises its little head as I did recall some changes to the rename functionality from the olde world releases to the newer releases. In fact, probably a long long time ago

With the rename in DFdss you can not add or drop another LQ, but, you can rename one of the LQs,

Try something like this

PeterHolland · Posted: Wed Jul 28, 2010 2:27 pm

I found something like this (expat do we have the same libraries?) :

devmisra · New User Joined: 29 Jan 2009 Posts: 39 Location: India

Thanks a lot. I already tested above codes.Its working.But in both the cases we use same no of qualifiers in source and dest gdgs.

Well, can we do like this?

1. Copy generation to generation using ADRDSSU. ( both the gdgs have 4 qualifiers).
2. Rename the dest generations (4 qualifier gdg) to other gdg which is having 5 qualifiers.

PeterHolland · Posted: Wed Jul 28, 2010 3:51 pm

Expat, stated the following :

expat · Posted: Wed Jul 28, 2010 3:58 pm

Why the desire for a fifth qualifier, surely a backup is a backup in anyones language.

If it is merely a backup question then why not let HSM (or whatever) deal with it automatically.

devmisra · New User Joined: 29 Jan 2009 Posts: 39 Location: India

Yes, you both are correct and also am adding this "Client is always a Client." :-)

The thing is Client has changed the Proc for 5 qualifer gdg and we have to go for this. No way out for me other than copying one by one.

dick scherrer · Posted: Wed Jul 28, 2010 7:35 pm

Hello,

If the client insists, then the client can pay for the time/effort required

nevilh · Active User Joined: 01 Sep 2006 Posts: 262

We had a similar requirement and in the end we ran the copy with the PARM='TYPRUN=NORUN' and saved the output to a disk file . We then used the saved file as input to a rexx that generated a rename job. If this is a one off task it is easy to do it by hand, if it is a regular task it is a pain .

devmisra · New User Joined: 29 Jan 2009 Posts: 39 Location: India

and its very unfortunate here... :-) Only effot and no pay.. :-(

Pete Wilson · Posted: Mon Aug 16, 2010 9:50 pm

Note to IBM, how about and ADDQUAL parameter, we have been waiting patiently for a LONG time and other vendors have provided this for decades...grrrr

FDRDSF it can add qualifiers if you have it:

RESTORE example:
A dataset named "LEDGER.TRANSACT" is to be restored as
"LEDGER.JULY.TRANSACT(+1)”.

//RESTORE EXEC PGM=FDRDSF,REGION=32M
//SYSPRINT DD SYSOUT=*
//SYSUDUMP DD SYSOUT=*
//TAPE1 DD UNIT=TAPE,DSN=BACKUP,DISP=OLD,
// VOL=SER=12155
//SYSIN DD *
RESTORE TYPE=DSF,RECAT
SELECT DSN=LEDGER.TRANSACT,NEWINDEX=.+JULY(+1)

This an example of COPY:

COPY a set of SMS-managed datasets, creating SMS-managed copies. The original storage and
management classes of the input datasets will be passed to the SMS ACS routines, which may
accept or override them. Datasets with a first-level index beginning with "APPL" will be copied. The
NEWI= parameter is used to insert "TEST" as a new second-level index in all the output dataset
names.
//COPYSMS EXEC PGM=FDRCOPY,REGION=32M
//SYSPRINT DD SYSOUT=*
//SYSUDUMP DD SYSOUT=*
//SYSIN DD *
COPY TYPE=DSF
SELECT CATDSN=APPL*.**,NEWI=.+TEST

New Index rules:

NEWINDEX=
NEWI=
Copy or move the selected datasets using a new name formed by adding or replacing
one or more index levels in the original name; replacement index levels do not have to
be the same length as the original indexes they replace. In the simplest case, each index
level specified in NEWI is used in place of the corresponding index in the original name.
Any remaining index levels at the end of the name are copied unchanged. This can
easily be used to change the first indexes of the name.
For example, if the input dataset is A.B.C.D,
NEWI=D results in D.B.C.D (first index replaced)
NEWI=DD.E results in DD.E.C.D (first 2 indexes replaced)
If a period is specified without any preceding characters, one original index level is
copied from the input dataset name to the output. This allows you to easily modify
indexes in the middle of the name.
For example, if the input dataset is A.B.C.D,
NEWI=..E results in A.B.E.D (third index replaced)
NEWI=FF...G results in FF.B.C.G (first and fourth indexes replaced)
If + is specified before a new index level, that new index is inserted into the output
dataset name at that point. If ++ precedes the new index, it will be added to the end of
the name. If - is specified, the next input index level will be dropped from (not copied to)
the output name.
For example, if the input dataset is A.B.C.D,
NEWI=+F results in F.A.B.C.D (new first index added)
NEWI=..+F results in A.B.F.C.D (new third index added)
NEWI=++F results in A.B.C.D.F (new last index added)
NEWI=..- results in A.B.D (third index dropped)
NEWI=Q.-.+E results in Q.C.E.D (combination)
Note that, except for the ++ option, every period in the NEWI= mask corresponds to a
period (one index level) in the original (input) dataset name. The resulting new name will
be checked to insure it meets IBM standards.
If the NEWI= value ends in a GDG relative generation number, e.g.,
NEWI=..NEWMAST(–2), that relative number will be added to the end of the newname,
and a LOCATE done to get the proper absolute generation number.
NEWI= is a convenient way to rename every input dataset, while using some index
levels from the original name and replacing other indexes or adding new indexes