Learning about content distribution

Technical description of the content distribution process

How content distribution is initiated

The actions that make up the distribution event are initiated by the core application. A job is not the actual distribution event. It is a set of predefined parameters that control the content distribution event. A job is created through the user interface, and the information is stored in the database. After a job is created, the database has all the information needed to run the job, including the scheduling information, but the distribution event cannot start until the core application initiates the process.

To do this, a scheduler service polls the database. When the scheduler finds a ready-to-run indicator for a job, it notifies the core application, which initiates the content distribution process as defined in the job.

Note
When a job is run manually from the user interface (as opposed to running at a scheduled time), the user interface communicates the ready-to-run status to the database and notifies the scheduler.

The directory structure on the source

When a job is first run, the core application creates a subdirectory under the workspace named with the job ID. Under the job ID, two permanent directories and a temporary directory are created:

The temporary directory contains a mapping of the source. It is the location from which the data for the assignment is retrieved for transport to the destination devices. The temporary directory is created each time an assignment is created. It is deleted after the distribution event is finished.
The Shadow directory contains a mapping of the source. The backup copies of the data are stored there.
The JournalFiles directory contains a file for each assignment run for a job. The file contains the command buffer for that assignment.