Problems with current plan batch jobs

The recovery actions you take are crucial any time a CP batch job ends unsuccessfully, for example:

• A DP batch job abnormally ends or is canceled.
• A system outage occurs while a DP batch job is running.
• In an end-to-end with fault tolerance capabilities environment, a problem related to UNIX™ system services (HFS, zFS, OMVS) occurs during DP batch processing.

Before taking any recovery action, do the following:

1. Check if the NCP you have is valid for a restart, looking for message EQQN057I in the controller message log. If you do not find it, do not restart the controller with JTOPTS CURRPLAN(NEW). In fact the controller issues message EQQN057I when recognizing the NCP, created by the DP process, as new version of the CP.
2. If you must restart the controller after unexpected events, such as system crash or controller abend, before starting the controller temporarily set in the in JTOPTS statement:
   • JOBSUBMIT(NO).
   • If the end-to-end scheduling with fault tolerance capabilities is active, FTWJSUB(NO).
3. With controller running, compare the current plan end value reported in LTP (panel 2.4 EQQLSTAP) with the planning period end value reported in CP (panel 6.6 EQQGSCPP). If the EQQLSTAP value is more recent than EQQGSCPP value, it is a symptom of a synchronization problem during the Symphony file creation process.