Use the Enable option to "Submit a Job for Execution" and the Disable option to "Temporary Withdraw a Job from Execution". Job can either be created in ENABLED state by using Save & Enable option or in DISABLED state by using the Save button. You must enable the disabled job in order to run it. To enable a disabled job, simply go to Jobs, then from the Quick Actions Menu of a job, click Enable. The Save & Enable option should be chosen only when you are confident that job is correctly configured, because this job will begin executing as soon as you "Save" the job. This option is available only when creating a Job the first time, not during editing the job.
To enable a job, a job must include at least one patch, a QQL, or one pre-action or post-action. Also, it must target at least one included asset or one included tag.
For the Run-Once jobs i.e. "On-Demand" and "Non-recurring Scheduled", you can submit them for execution only once, since they are going to run only once. The only difference between the "On-Demand" and "Scheduled" job is that the first option initiates execution of the job right away, while the second option waits till the scheduled time or past it.
In summary, Run-Once jobs can only be enabled once. They cannot be disabled. Run-Once jobs can be edited only before you enable them. Once you Enable them, the job configuration can not be modified.
Run-Once jobs ideally do not expire. They are guaranteed to run once, even if you enable them past their schedule time. The only exception where such jobs expire is if you "Set Duration" within "Patch Window" to anything other than "None".
Remember that patch window cannot be greater than 7 days i.e. 168 hours. If you think you would like it to be more than 168 hours, we would suggest you turn off the "Patch Window" option itself, by changing it to 'None'.
Recurring Jobs on the other hand can be enabled and disabled multiple times. They can also be edited in both enabled as well as disabled state, but the changes will reflect in the next run only. The next run is a loaded term as it depends on multiple parameters of a job configuration. Some of these parameters are explained below:
- The scheduled time of the job (minus 3 hours) should be a point in time later in the day.
- If the job is scheduled for the Agent time zone, which is the default value, then it is possible that for some endpoints the run may have executed and for some it is pending.
Those pending for execution more than 3 hours later from the time of saving configuration (and re-enabling the job, if disabled) will reflect change of behavior in the current run while the ones where it has already been executed or pending for execution less than 3 hours later will reflect in the next run.
Depending on the schedule of the job i.e. daily, weekly, etc, next run could be few days or weeks later. Technically, it could be few hours later too but we would recommend you to configure schedule cycles for the job a day apart, at the least.