|
View previous topic :: View next topic
|
| Author |
Message |
balaji81_k
Active User
Joined: 29 Jun 2005
Posts: 155
|
|
|
|
Hi ,
We are in process of automation for the output files created in monthly job for one of our real time application . So we are writing a new job to take the backup of output files as flat file whether the input is GDG / NON GDG .
We have to use naming convention in the output file for the Input files .
1. For Non GDG input file , the output file will be the same name followed by date (run date )(one of qualifier) that can be picked from job scheduler. We have no problem on it .
2. For GDG input file , we need to have output file as GDG base name followed the Qualifiers like below
for example
My GDG file is A.B.C.G0091V00
my output file A.B.C.V91.D(MMYY)
no problem about the last qualifier but for mentioning the version we stuck with that .
For that we are in process to do the below
1. Input file is of (a.b.c(+0)) and have temp dataset to have only latest version on it.
2. Copy step with symbolic parameters having Version number .
Can any one please help to provide guidelines to implement the above 2 steps . It will be really help for us. We are also trying from our end.
Thanks
Balaji K |
|
| Back to top |
|
 |
expat
Global Moderator
Joined: 14 Mar 2007
Posts: 8797
Location: Welsh Wales
|
|
|
|
I suggest that you use DFdss PGM=ADRDSSU to perform the backups and that way you will only need to change the name of the output backup dataset.
But why oh why do people insist on using dates in dataset names  |
|
| Back to top |
|
|
balaji81_k
Active User
Joined: 29 Jun 2005
Posts: 155
|
|
|
|
Hi,
Utility "ADDRSPU" is not enabled in our application .
Thanks
Balaji K |
|
| Back to top |
|
|
balaji81_k
Active User
Joined: 29 Jun 2005
Posts: 155
|
|
|
|
| Sorry its type ADRDSSU and Date we need to do as per our requirements. |
|
| Back to top |
|
|
Akatsukami
Global Moderator
Joined: 03 Oct 2009
Posts: 1788
Location: Bloomington, IL
|
|
|
|
| expat wrote: |
| But why oh why do people insist on using dates in dataset names |
Because they can't handle the concept of the data set containing a header record that contains, among many other useful things, the processing date 
Balaji, you have not told us what options are available to you. Writing such a tool in Rexx is quite trivial (although depending on volume may be impractical); it would be harder in COBOL or PL/I, although quite doable. However, I suspect that one of the "requirements" is that it must be implemented by a gaggle of junior "software engineers" that know nothing of programming or software development (as evinced by your -- very common among junior "software engineers" -- of "generation" with "version"). I suggest that your organization hire someone who knows what he's doing. |
|
| Back to top |
|
|
Nic Clouston
Global Moderator
Joined: 10 May 2007
Posts: 2455
Location: Hampshire, UK
|
|
|
|
| You realise that for the naming convention proposed for the GDG output you need to set up a GDG base for EVERY base name? And that it would hold approximately ONE generation on each base name. |
|
| Back to top |
|
|
Anuj Dhawan
Superior Member
Joined: 22 Apr 2006
Posts: 6250
Location: Mumbai, India
|
|
|
|
| A Back-up is well, a back-up -- so why this much fuss about gdg/non-gdg. And a GDG is just another QSAM file for which you already have a process in place... |
|
| Back to top |
|
|
expat
Global Moderator
Joined: 14 Mar 2007
Posts: 8797
Location: Welsh Wales
|
|
|
|
| Strange, I thought the DFdss was a standard part of the SMS subsystem and comes FREE from IBM |
|
| Back to top |
|
|
Anuj Dhawan
Superior Member
Joined: 22 Apr 2006
Posts: 6250
Location: Mumbai, India
|
|
|
|
| expat wrote: |
| Strange, I thought the DFdss was a standard part of the SMS subsystem and comes FREE from IBM |
Ditto. |
|
| Back to top |
|
|
enrico-sorichetti
Superior Member
Joined: 14 Mar 2007
Posts: 10873
Location: italy
|
|
|
|
to all the people wasting time trying to help ...
what is that You do not understand here ?
the requirement forbids <DFDSS> and prescribes a dataset name with the date
probably time to lock the topic as per
www.ibmmainframes.com/viewtopic.php?t=20820&highlight=
Add a DATE and/or TIMESTAMP to a dataset. This Forum recommends that you consult with your Scheduling team first. Otherwise, this topic has already been posted many times. Please search for the answer that best suits your needs. |
|
| Back to top |
|
|
|
|
