Introduction

XINA provides an integrated system for running and tracking asynchronous tasks.

There are currently two task implementation options. They may either be a generic process registered with the XINA Run utility, or implemented in the Amazon Web Services (AWS) Lambda service. Task deployment must be explicitly configured for each XINA instance.

Tasks are executed with the RUN API action. Note that successful completion of a RUN action does not mean the task has executed successfully, just that the task has been queued for execution successfully. Depending on the task type and configuration it may take additional time to start, or may never start if something is misconfigured or disabled.

The RUN action used the following syntax:

 {
  "action" : "run",
  "tasks"  : [ <task JSON object>, ... ]
 }

Each task is defined with the following properties:

Property	Type	Value
name	string	task name
conf	JSON object	task configuration
auto	boolean	indicates if a task was started automatically (optional, default false)
archive	boolean	if true files will be saved locally by XINA Run (optional, default false)
open	boolean	if true not be concluded after completing (lambda only) (optional, default false)
parent	integer	parent task ID (lambda only) (optional)
desc	string	plain text task description (optional)
thread	string	task thread name (XINA Run only) (optional)
ref_id	integer	arbitrary reference ID for client use (optional)
priority	integer	task priority (not yet implemented)
timeout	integer	absolute task timeout (not yet implemented)

The task configuration object is the primary means by which information can be passed to the task. The format required for the object is defined by each task. Note that XINA does not validate the configuration format when receiving the as it does not know what a valid format would be. It only ensures that the configuration is a valid JSON object.