The correct action is to resume the job from Activity Monitor. Cohesity NetBackup documentation states that Activity Monitor provides a Resume action for jobs that are suspended or in an Incomplete state; the documented procedure is to open Activity Monitor > Jobs, select the incomplete job, and choose Actions > Resume Job. This is the supported way to make the same incomplete backup continue before the next scheduled backup starts. It is especially important in a multiple data streams policy, where each stream appears as an individual job and the failed stream should be resumed rather than replaced by a new manual backup. The Move backup job from incomplete state to done state property only controls how long a failed backup can remain incomplete before Activity Monitor shows it as Done; reducing that value would shorten the recovery window, not rerun the backup. Cohesity also notes that a resumed job reuses the same job ID, whereas a restarted job receives a new job ID. Running a manual client or mount-point backup would create a separate backup operation and would not verify completion of the existing incomplete job. References/topics:
Activity Monitor job actions, Resume incomplete jobs, Checkpoint restart, and Clean-up host properties.