Skip to main content

Tasks

A task is Usercube's way to configure and use a given executable that performs a given IGA action.

Anatomy of a Task​

Each of Usercube's IGA actions is contained in a standard Windows executable file that can be launched using PowerShell.

The choice of a simple standard format for Usercube's building blocks makes it very easy to pick and choose them a la carte to configure the solution.

Tasks are used to insert these blocks into Usercube's configuration, in order to be launchable via the UI, or even scheduled to be launched automatically periodically.

For example, Usercube's tasks include synchronization, computation of entitlement assignments, or provisioning of varied managed systems. See the list of all available tasks.

See how to configure tasks.

Data Consistency​

Every task is written as a transactional process. This means that a task cannot be executed partially. It is either fully executed, or not executed at all. It guarantees data consistency as data cannot be harmed by a half-executed task.

Every task is written as an idempotent function. This means that, for a given input, applying a task one time will produce the same result as applying it several times. It guarantees data consistency as it prevents the potential side-effects of a retry which might occur following a network error, or a task failure.

Every task is designed as a single responsibility process. This principle ensures that two distinct tasks do not have an effect on similar pieces of the system. This guarantees data consistency by avoiding incompatible changes to be committed by different tasks at the same time. For the same reasons, a given task cannot be executed twice simultaneously.

Task Modes​

Two distinct modes exist to execute tasks inside jobs:

  • In complete mode, tasks process whole inputs with all data.
  • In incremental mode, tasks only consider the changes that occurred since their last execution. This mode is not available for all tasks.

Both modes can be performed considering potential filters if said tasks involve a specific selection of data instead of whole inputs. The difference between these modes lies in the consideration of all data for the complete mode, versus only the last changes for the incremental mode.