The key exec_workflow¶
This key is string and allow to set all your workflow configuration process with the ID of the Jobs
The Jobs have an alphanumerical string [a-zA-Z0-9_] so let's suppose that we have 4 Jobs with IDs 1,2,3 and 4 for examples.
The exec_workflow control if your workflow is sequential or parallel with commas , and colons : and you can play with this.
For example, if exec_workflow value is 1:2:3:4 means that your workflow is:
One job at a time in sequential form
If exec_workflow value is 1,2:3:4 means that your workflow is:
Jobs with IDs 1 and 2 executes in parallel
If exec_workflow value is 1,2:3,4 means that your workflow is:
Jobs with IDs 1 and 2 executes in parallel and when both completes then 3 and 4 execute in parallel
The workflow defined in exec_workflow global key does not take into account the events. These are configured in every job. See Handle Jobs events. But for preview, let's see this example:
Imagine there are 4 Jobs configured in your package, with sames ID but here we going to use events for calling the other jobs. This is the configuration: Job 1 call 2 and 3 when ocurrs a SUCCESS event, and after calls 4.
So the configuration for exec_workflow is 1:4 and the job 1 will have configured his local exec_workflow for event_success with value 2,3. The workflow is:
4 is called after 1 and all his events complete. More examples in Handle Jobs events