When the scheduler processes a CHANGE_STATUS event to set a box job to the INACTIVE status, it overwrites the status of jobs that are set to the ON_NOEXEC status with the INACTIVE status.
This behavior causes the scheduler to execute jobs that are supposed to bypass execution the next time the starting conditions of the box are met.
WAAE 11.3.5 on AIX
After sending a CHANGE_STATUS/INACTIVE to a box containing jobs in the ON_NOEXEC state, re-send JOB_ON_NOEXEC events to the jobs to restore its non-execution state.
Or apply one of this :
- fix T54Y065
- CA WAAE 11.3.6 SP3
- CA WAAE 11.3.5: Post INCR4 + CUM1