Open topic with navigation

Jobs

A Job is an organized configuration of one or more ordered tasks. The tasks in a job may be organized into several Phases, where a phase must run to completion before the next phase can begin. Jobs may be scheduled to run using the Director UI, the Server Console UI, or an external scheduler using the Command Line Interface.

A task is the execution of a Snapshot, a Process, an Export, a Results Book Export or an External Task.

Jobs may also include Triggers. Triggers can be used to stop real-time processes or to start other jobs, either before the other tasks in a phase start running, or once they are all complete.

It is possible to configure a Job Notification email to be sent out on completion of a job. The email will show the status of the job (that is, whether it succeeded or failed), and will also contain details of any warnings or errors encountered during execution of the job. See the topic Job Notifications for more details.

For a full guide to the various options available when configuring a job, see Execution Options.

Defining and scheduling a job

Defining a job provides full access to all runtime options when running a process, that is, all of the Process Execution Preferences and:

• The ability to schedule the Job to run in the future, and on a recurring basis
• The ability to define and save a Job that consists of many tasks, such as multiple snapshots, processes, and exports
• The ability to run External Tasks before or after these tasks (for example, to transfer a new file into the server's landing area for processing)

To define a job, right click the Jobs section of the Project Browser (within each Project) and create a new job. Set up a job and add an ordered set of tasks. You can then schedule the job to run at a convenient time.

For example, the following job is designed to perform full screening of a client's customers against watchlists, in three phases.

In the first phase, new lists are downloaded using File Download tasks.

In the second phase, the customer data and watchlists are snapshotted, prepared for matching, and matched.

Finally, in the third phase, a Results Book Export writes summary results to a file:

Click on each task within the job to inspect or edit the configuration of the task.

Phases

Phases allow you to control the order in which tasks are executed in a job. For example, if you want to ensure that a task completes before the next task starts, you should use two phases. The later phases will only execute after the former phases have completed.

Within a phase, all configured tasks will run as quickly as possible, and wherever possible will run in parallel, sharing the available threads. It is possible to run several 'chained' process in the same phase, in which case they will effectively run as a single process.

Note that the enablement or disablement of a phase can always be overridden using a Run Profile when running a job from the Server Console UI, or using the 'runopsjob' command of the OEDQ command line interface.

Conditional execution

The execution of a phase can be made conditional on the outcome of the previous phase. There are three execution options when configuring a phase within a job:

• Execute regardless - this means that the phase will execute regardless of whether or not a previous phase completed successfully. This may be useful if you want to ensure that a process runs on the latest available set of data even if that data could not be refreshed (for example, because it is downloaded from the Internet and a connection was unavailable).
• Execute on success - this means that the phase will only execute if all the previous phases completed successfully. This is useful if you want to ensure that processing time and resources are not wasted running tasks that may not error, but which may be using out-of-date data.
• Execute on failure - this means that the phase will only execute if the previous phase did not complete successfully. This allows you to take action in the event of a failure in part of a job - for example to run a different snapshot if another snapshot failed to run.

Note that if an error occurs in any phase, the error will stop all 'Execute on success' phases unless an 'Execute regardless' or 'Execute on failure' phase runs before it which has the 'Clear Error?' button ticked.

Triggers

Triggers can be added to a job from the Tool Palette on the right of the screen. Two types of trigger are available: Run Job triggers and Shutdown Web Services triggers.

• Run Job Triggers are used to start a job from within another job.
• Shutdown Web Services Triggers are used to shut down real-time processes.

For more information on triggers and their use in jobs, see Using Triggers.

Command line interface

Jobs can be run using OEDQ's Command Line interface. See the Command Line Interface guide for full details.

Deleting a job

A job can be deleted by clicking on it in the Project Browser and pressing the Delete key, or by right-clicking on it in the Project Browser and selecting 'Delete'.  Deleting a job does not delete the processes that the job contained.