|
View previous topic :: View next topic
|
| Author |
Message |
vbhat
New User
Joined: 29 Apr 2005
Posts: 38
|
|
|
|
1) I have 3 steps S1,S2 and S3 respectively
In s1 i am creating a temporary data set and i am using it in s2 and s3 steps if s3 abends can i restart s3?
pls help me
thanks
vbhat |
|
| Back to top |
|
 |
David P
Active User
Joined: 11 Apr 2005
Posts: 106
Location: Cincinnati Ohio
|
|
|
|
Hi Vbhat,
temp data set remain only till the job is executing once the execution is over they are not available.
Since you are creating the temp data set in the first step and second and third step you are using that, in case any of your second or third step abends you will have to start from the first step as the temp data set will
only be available after the completion of first step.
So the answer to your question is you cannot restart from step 3. you will have to execute step1 first and create the temp data set again.
let me know in case of any clarification.
David. |
|
| Back to top |
|
|
vbhat
New User
Joined: 29 Apr 2005
Posts: 38
|
|
|
|
| thanks david |
|
| Back to top |
|
|
MGIndaco
Active User
Joined: 10 Mar 2005
Posts: 432
Location: Milan, Italy
|
|
|
|
I think that you can...
Not all the temporary dataset are deleted when the job is ended... and not all the temporary dataset are preceded by a && 
Try with this:
| Code: |
//S1 EXEC PGM=MYPGM
//DD1 DD DISP=(,PASS),LRECL=XX,RECFM=XX,SPACE...,
// DSN=MYUSER.SEQ
//S2 EXEC PGM=MYPGM
//DD1 DD DISP=(OLD,PASS,CATLG),LRECL=XX,RECFM=XX,SPACE...,
// DSN=MYUSER.SEQ
//S3 EXEC PGM=MYPGM
//DD1 DD DISP=(OLD,DELETE,CATLG),LRECL=XX,RECFM=XX,SPACE...,
// DSN=MYUSER.SEQ |
If I'm not wrong, and I'm not... if your second or third step have an abend you can restart from the same point where you have stopped. |
|
| Back to top |
|
|
M Alam
New User
Joined: 13 Jun 2007
Posts: 1
Location: pune
|
|
|
|
MGIndaco,
Since the dataset was not Cataloged or kept in step S1 in first run, how will in second run, the dataset will be found by step S3? Please clarify.
Thanks |
|
| Back to top |
|
|
murmohk1
Senior Member
.jpg)
Joined: 29 Jun 2006
Posts: 1436
Location: Bangalore,India
|
|
|
|
Alam,
| Quote: |
| Since the dataset was not Cataloged or kept in step S1 in first run, how will in second run, the dataset will be found by step S3? Please clarify. |
Once the temp dataset is passed to next step, the disp of the dataset is determined in that step.
In MGIndaco code, step S2 has DISP=(OLD,PASS,CATLG). So when the job abends in this step, the dataset remains intact. |
|
| Back to top |
|
|
dick scherrer
Moderator Emeritus
Joined: 23 Nov 2006
Posts: 19244
Location: Inside the Matrix
|
|
|
|
Hello,
By definition, a cataloged dataset is not a temporary dataset - it may possibly be a transient dataset, but it is not a temporary dataset.
True "temporary datasets" cannot exist once a job completes - whether successful, abended, or an in-run jcl error. |
|
| Back to top |
|
|
murmohk1
Senior Member
Joined: 29 Jun 2006
Posts: 1436
Location: Bangalore,India
|
|
|
|
Dick,
This was taken from QW -
| Quote: |
PASS
The data set will be passed to a subsequent step in the same job. Its
final disposition will be determined by that step. |
|
|
| Back to top |
|
|
dick scherrer
Moderator Emeritus
Joined: 23 Nov 2006
Posts: 19244
Location: Inside the Matrix
|
|
|
|
Hello,
Yes, but that quote has nothing to do with true temporary datasets. If you PASS a dataset with DSN=&&stuff.data.file it will not exist when the job ends. If you have a dataset with DSN=my.data.file, it exist after the job completes, but that is not a tgemporary dataset.
The dsn determines whether the dataset is a temporary dataset, not the usage intention of the author. |
|
| Back to top |
|
|
|
|
