I have a requirement where I don't want to repeat the steps coded under HEADER3. Instead the job with one step has to be repeated for every value of the input and submitted through internal reader.
Please advise how to do it. When I move the "Sections=(81,3," above HEADER1, its giving me error.
To explain clearly this is the code I have to create a job and add one step for each value of the INPFIELD from input file FILE1.
I want this code to be modified to generate a 'job and one step' for each value of the INPFIELD and submit it to internal reader. I dont multiple steps in the same job for each field. I want one job for each INPFIELD.
I would also want to make the "FRMTS=2018-07-25-00.00.00.00000" , "TOTS=2018-07-25-00.00.00.000000" and "VALUE=P001,VERS=VERS1" parameters dynamic by passing symbolic parameters. When I pass symbolic parameters, DFSORT is not evaluating them.
In each step of the job I'd be calling another proc which has around 20 steps. So if I create 40 steps for each value of the input file, it would cross the 255 limit for a job.
Then I thought of having a single job with one step created at once and submit it immediately.
Please advise if there is a way around to to do this.
Joined: 17 Oct 2006 Posts: 2481 Location: @my desk
You could do this using a bit of rexx code, if you would like to do that way. Using the sort utility, it might be a bit more complex. You may have to build a file with all the jobs-one after the other, then read this file, route each job into different OUTFIL dd data sets dynamically.
Can't we write the built JCL statements to INTRDR directly without writing to a file ? That way its submitted upon writing to INTRDR which is what we want.
I am able to submit the JCL statements to INTRDR built for first value of the file. I just have to repeat the BUILD for each value of the input file. Please advise on this.
Joined: 14 Mar 2007 Posts: 8796 Location: Welsh Wales
Why not use ISPF file tailoring for the purpose it was designed for, tailoring pre defined JCL skeletons for certain situations. Gives you far better control when things change.
Why not use ISPF file tailoring for the purpose it was designed for, tailoring pre defined JCL skeletons for certain situations. Gives you far better control when things change.
So far, the task seems to be too trivial for FT with the whole ISPF environment to get involved.
Joined: 17 Oct 2006 Posts: 2481 Location: @my desk
vnktrrd,
I did not realize you can have multiple job cards in the same data set and once that is routed to internal reader, those would get submitted as individual jobs, until a friend of mine discussed about this with me. You can remove the HEADER1 and move the job card to the HEADER3 and build a single output with multiple job cards. You might as well talk to your scheduling guys if they can set something up instead of you building this submission process.
