How can I trigger an event upon successful completion of a job?

[Frank Korfias]

Currently code looks like this:

 

JOB DXRFADSD                                 
  CCCHK RC(1:9999) OK CONTINUE               
  RUN ANY                                    
  NOTIFY JOBEND EVENT(DXGROUP.DXRFADSD_JOBEND)
  RELEASE DXRFALTU                           
ENDJOB                                       

 

What I am trying to do is ensure that a list of jobs continues to run even if this job fails, which is why I have the OK CONTINUE, at least that is my understanding of its use. I need the event DXGROUP.DXRFADSD_JOBEND to only be triggered upon successful completion of job DXRFADSD. I am getting mixed search results as to whether this is the correct way to go.

 

Thanks,

Frank

[powdo03]

Hi Frank, 

First I would like to verify what you are trying to accomplish. 

If job DXRFADSD gets any return code you want it to end successfully on the screen. 

If DXRFADSD ends with a 0 then you want to trigger an event. 

 

Does this sound correct? 

 

I will start with the simple case.

DPTST001 will always end successfully. Note 4095 is the highest RC possible. 

 

DPTST002 and DPTST003 only run if DPTST001 ends with a 0  If DPTST001 gets a return code > 0 then DPTST002 and DPTST003 will BYPASS 

 

JOB DPTST001
RUN DAILY
    CCCHK RC (0:4095) OK CONTINUE
    RELEASE ADD(DPTST002) COND(RC(0))
ENDJOB

JOB DPTST002
    RUN DAILY
    RELEASE DPTST003
ENDJOB

JOB DPTST003
    RUN DAILY
ENDJOB

 

You mentioned triggering an EVENT...... That would look something like this. 

 

JOB DPTST001 
    RUN DAILY 
    CCCHK RC (0:4095) OK CONTINUE 
    RELEASE ADD(DPTST002) COND(RC(0)) 
ENDJOB 

 

JOB DPTST002 LINK PROCESS
RUN DAILY 
ESP TRIGGER DXGROUP.DXRFADSD_JOBEND
ENDJOB 

 

IMPORTANT NOTE:

In this scenario if DPTST001 gets a SOC7 it is going to fail and stop these jobs. The above scenario is only coded for return code failures. 

[Frank Korfias]

Good morning,

 

Partially correct, if DXRFADSD gets a return code greater than 0 I want it to fail, but I still want it to release the next job in line. If it gets a return code of 0, then I also want it to trigger an event. I am guessing I need to change from OK CONTINUE to FAIL CONTINUE. Here is what I have now:

 

NOTIFY OVERDUE FAILURE ABEND MAILBOX(SECURITY)

 

JOB DXRFADSD

RUN ANY

  CCCHK RC(1:4095) FAIL CONTINUE

  RELEASE ADD(RELFADSD) COND(RC(0))

  RELEASE DXRFALTU

ENDJOB

 

JOB DXRFALTU

RUN ANY

  CCCHK RC(1:4095) FAIL CONTINUE

  RELEASE ADD(RELFALTU) COND(RC(0))

  RELEASE DXRFCERT

ENDJOB

 

JOB DXRFCERT

RUN ANY

  CCCHK RC(1:4095) FAIL CONTINUE

  RELEASE DXRFCGRP

ENDJOB

 

JOB DXRFCGRP

RUN ANY

  CCCHK RC(1:4095) FAIL CONTINUE

  RELEASE ADD(RELFCGRP) COND(RC(0))

  RELEASE DXRFPERM

ENDJOB

 

JOB DXRFPERM

RUN ANY

  CCCHK RC(1:4095) FAIL CONTINUE

  RELEASE DXRFRDEF

ENDJOB

 

JOB DXRFRDEF

RUN ANY

  CCCHK RC(1:4095) FAIL CONTINUE

  RELEASE DXRFSITE

ENDJOB

 

JOB DXRFSITE

RUN ANY

  CCCHK RC(1:4095) FAIL

ENDJOB

 

JOB RELFADSD

RUN ANY

  ESP TRIGGER DXGROUP.DXRFADSD_JOBEND

ENDJOB

 

JOB RELFALTU

RUN ANY

  ESP TRIGGER DXGROUP.DXRFALTU_JOBEND

ENDJOB

 

JOB RELFCGRP

RUN ANY

  ESP TRIGGER DXGROUP.DXRFCGRP_JOBEND

ENDJOB

 

 

So basically I need the following jobs to run on a daily basis no matter what their return code is:

DXRFADSD

DXRFALTU

DXRFCERT

DXRFCGRP

DXRFPERM

DXRFRDEF

DXRFSITE

 

 

Should the following jobs have a return code of 0, they should each trigger a separate event:

DXRFADSD

DXRFALTU

DXRFCGRP

[powdo03]

Hi Frank, 

NOTES - A couple of notes so we are on the same page. 

1) In the statement below the CONTINUE means the next step in the JCL will run. This is not related to the next job in the schedule. 

"CCCHK RC(1:4095) FAIL CONTINUE" 

 

2) To make the next job run the RELEASE statement needs to be coded correctly.  To make it release whether it completes successfully or not(unconditionally) it should look like the statement below. 

RELEASE ADD(DPTST002(U))

 

3) I added the statement below to cover the default case. 

CCCHK RC (0) OK CONTINUE

 

4) I am assuming that someone is going to come back and restart the failed jobs. It would not be good to require someone to clean this up manually each time a job fails. 

 

I think it would look closer to the schedule below. 

 

NOTIFY OVERDUE FAILURE ABEND MAILBOX(SECURITY)

JOB DXRFADSD

  RUN ANY

  CCCHK RC(1:4095) FAIL CONTINUE

  CCCHK RC (0) OK CONTINUE

  RELEASE ADD(DXRFALTU(U))

ENDJOB

 

JOB DXRFALTU

RUN ANY

  CCCHK RC(1:4095) FAIL CONTINUE

  CCCHK RC (0) OK CONTINUE

  RELEASE ADD(DXRFCERT(U))

ENDJOB

 

 

JOB DXRFCERT

RUN ANY

  CCCHK RC(1:4095) FAIL CONTINUE

  CCCHK RC (0) OK CONTINUE

  RELEASE ADD(DXRFCGRP(U))

ENDJOB

 

.........

..........

........

For these jobs the TRIGGER can't be in the same job since you want to trigger it after it ends and only in the case of RC=0 

JOB RELFADSD

RUN ANY

  RELEASE ADD(LINK1) COND(RC(0))

ENDJOB

 

JOB LINK1

  RUN ANY

  ESP TRIGGER DXGROUP.DXRFADSD_JOBEND

ENDJOB

 

[Frank Korfias]

Thank you. I will get it coded up and start testing and let you know how it goes.

[Frank Korfias]

Works great, thank you so much!

[powdo03]

Great!!!.... Thanks for the update......