|
View previous topic :: View next topic
|
| Author |
Message |
ibmcoolz
New User
Joined: 20 Sep 2006
Posts: 6
|
|
|
|
I am just trying to get a clarification on GDGs. Suppose I create a GDG
dataset in the first step of my JCL for eg.
I create a GDG dataset TEST.ABC.G0001V00 in Step 010 in my jcl.
using DSN=TEST.ABC(+1)
I wish to refer to it in the same job, in say, Step 030 , a subsequent step.
Can i refer this dataset using DSN=TEST.ABC(+0),DISP=SHR in Step 030. |
|
| Back to top |
|
 |
superk
Global Moderator
Joined: 26 Apr 2004
Posts: 4652
Location: Raleigh, NC, USA
|
|
|
|
| No, it's +1. |
|
| Back to top |
|
|
Robert Sample
Global Moderator
Joined: 06 Jun 2008
Posts: 8697
Location: Dubuque, Iowa, USA
|
|
|
|
| The new generation won't be cataloged until the job finishes, so everywhere you need the generation in the job, you will need (+1) as Kevin says. |
|
| Back to top |
|
|
dick scherrer
Moderator Emeritus
Joined: 23 Nov 2006
Posts: 19244
Location: Inside the Matrix
|
|
|
|
Hello,
When you have a question, post a new topic for your your question rather than posting a reply to a topic that has been inactive for 4 years.
d |
|
| Back to top |
|
|
gcicchet
Senior Member
Joined: 28 Jul 2006
Posts: 1702
Location: Australia
|
|
|
|
Hi Robert,
I disagree with
| Quote: |
| new generation won't be cataloged until the job finishes |
.
I'm able to see the GDG's cataloged when a job is still running.
Gerry |
|
| Back to top |
|
|
radhakrishnan82
Active User
Joined: 31 Mar 2005
Posts: 435
Location: chennai, India
|
|
|
|
For GDG it will always say as 'ROLLED IN' and not 'RETAINED' as for the normal datasets. The 'Rolledin' is 'ROLLEDOUT' at job termination and then only the version is available.
The GDG, version control is handled by Job termination routines. Even though we create a +1, it becomes a 0 level, after Job termination.So that is the reason why you keep +1 in the job, wherever the dataset is refered in the same job and in the same run.
Hope this helps. |
|
| Back to top |
|
|
ibmcoolz
New User
Joined: 20 Sep 2006
Posts: 6
|
|
|
|
Thanks for all your responses. I agree that when we create a new GDG we must refer it with a +1 in a subsequent step in the same job.
But I was told that a particular installation allows usage of a +0 to refer GDG created in an earlier step in the job. Maybe an exception!! |
|
| Back to top |
|
|
dick scherrer
Moderator Emeritus
Joined: 23 Nov 2006
Posts: 19244
Location: Inside the Matrix
|
|
|
|
Hello,
| Quote: |
| But I was told that a particular installation allows usage of a +0 to refer GDG created in an earlier step in the job |
If the job is terminated early for some reason and re-started later, the reference to the new generation (cataloged by the terminated run) would then be zero not +1 as would be coded in the normal jcl.
Possibly, this is the situation where someone mentioned the zero would be used. |
|
| Back to top |
|
|
radhakrishnan82
Active User
Joined: 31 Mar 2005
Posts: 435
Location: chennai, India
|
|
|
|
| Quote: |
| But I was told that a particular installation allows usage of a +0 to refer GDG created in an earlier step in the job. Maybe an exception!! |
can you tell me what installation is that?
As for as I know, during aborts
1)Once the job aborts,then the job gets completely disconnected with the submitted id.In this case,after job aborts,if you need to restart the job then it is same as submitting a new job. eg., version (0) is applicable.It is same as submitting a new job.hence (0) is applicable for GDG's (only during the restart).
2)The job moves to hold once the job aborts.If you need to restart after the abort then the control lies within the job and it is entirely different from the job that follows the process of the first step. For scheduling systems,the version (+1) is applicable during restart.Scheduling systems belong to the second case where in which your job
flow STOPS during aborts. |
|
| Back to top |
|
|
radhakrishnan82
Active User
Joined: 31 Mar 2005
Posts: 435
Location: chennai, India
|
|
|
|
| I didn't noticed the posting by Dick. I agree with Dick. |
|
| Back to top |
|
|
|
|
