View previous topic :: View next topic
|
Author |
Message |
s.ayodhyawasi
New User
Joined: 16 Sep 2006 Posts: 13
|
|
|
|
In my JOB, step1 is a cobol program which generate job summary report for spool. I want to capture it in a dataset for processing in next step. I am not suppose to catalog it, since we have standard to send the job summary from program into spool.
I found the below example to capture the JESMSGLG using ISFAFD.
www.ibmmainframeforum.com/viewtopic.php?f=16&t=3199&start=20#p13052
Please suggest if it can be modified to capture the spool data for a particular ddname of previous step.
Thanks for help. |
|
Back to top |
|
|
dbzTHEdinosauer
Global Moderator
Joined: 20 Oct 2006 Posts: 6966 Location: porcelain throne
|
|
|
|
since it is a JOB and not a STEP summary,
why don't you generate the output of the cobol program to a temp (&&...) ds,
and then input the temp in your next step (simultaneously outputing to the spool summary (who dreams up this stupidity, anyway).
then you will not need to learn anything new
(in this case good, since if the JOB abends, you want the restart summary anyway)
and potentially difficult to implement.
why everyone needs to read the spool is just a technical exercise for little purpose. |
|
Back to top |
|
|
s.ayodhyawasi
New User
Joined: 16 Sep 2006 Posts: 13
|
|
|
|
thanks for quick reply.
If I save the job summary of step1 in temp dataset, use it in next step and generate simultaneously to spool as well, then the job summary will appear in next step ( though in same job but not in step1) and it's not as per standard of generating job summary of program in same step.
I was thinking there could be a way using ISFAFD, by changing some parameer to capture ddname spool data instead of JESMSGLG. In this way i don't have to change anything in step1.
thanks |
|
Back to top |
|
|
dick scherrer
Moderator Emeritus
Joined: 23 Nov 2006 Posts: 19244 Location: Inside the Matrix
|
|
|
|
Hello,
Quote: |
I am not suppose to catalog it, since we have standard to send the job summary from program into spool |
So, do as DBZ suggests - create a temporary dataset. Do what is needed with the dataset, copy it to the spool, delete the dataset without ever cataloging it.
You now have followed the "not catalog" rule and the output is in the spool where it is supposed to be.
Quote: |
In this way i don't have to change anything in step1 |
Why would this be prefereable? Also, changing a sysout to a dataset is extremely trivial. . . Why add code you don't know how to write (that will need to be supported long-term) and additional jcl when there is a very simple solution. |
|
Back to top |
|
|
s.ayodhyawasi
New User
Joined: 16 Sep 2006 Posts: 13
|
|
|
|
Quote:
You now have followed the "not catalog" rule and the output is in the spool where it is supposed to be.
the solution DBZ suggested will move the summary report to next step's spool from step1. |
|
Back to top |
|
|
dbzTHEdinosauer
Global Moderator
Joined: 20 Oct 2006 Posts: 6966 Location: porcelain throne
|
|
|
|
it would be easier to add an additional output file in the pgm of step one,
that would duplicate the output (of sysout) which could be a temp ds,
for input to step two.
but, I don't imagine you can change the pgm in step 1, can you? |
|
Back to top |
|
|
s.ayodhyawasi
New User
Joined: 16 Sep 2006 Posts: 13
|
|
|
|
yes, I can't change the program in step1 |
|
Back to top |
|
|
dbzTHEdinosauer
Global Moderator
Joined: 20 Oct 2006 Posts: 6966 Location: porcelain throne
|
|
|
|
my, it must be comforting to creators of inflexible standards,
to know that no one will ever question anything. |
|
Back to top |
|
|
dick scherrer
Moderator Emeritus
Joined: 23 Nov 2006 Posts: 19244 Location: Inside the Matrix
|
|
|
|
Hello,
Quote: |
the solution DBZ suggested will move the summary report to next step's spool from step1. |
Yes, but it will still be in the sysout for the job. . .
I suspect there is also lack of understanding as to the intent of the standard. . . |
|
Back to top |
|
|
|