Skip to main content
Version: v1.7

Overview

Workflows, as part of the Application, can glue together additional delivery processes and specify arbitrary delivery environments. In short, Workflow provides customized control flow and flexibility based on the original delivery model of Kubernetes(Apply). For example, Workflow can be used to implement complex operations such as pause, manual approval, if conditions, waiting status, data flow, multi-environment gray release, A/B testing, etc.

A workflow consists of multiple steps, and typical workflow steps include step groups (containing a series of sub-steps), human review, multi-cluster publishing, notifications, etc. You can view all built-in workflow steps provided by default in KubeVela in built-in workflow steps. If the built-in workflow steps don't meet your needs, you can also custom workflow steps.

Workflow consists of steps, you can either use KubeVela's built-in workflow steps like suspend, notification, deploy, step-group etc, or customize your own WorkflowStepDefinition.

In fact, if you only use components in the Application and do not declare a workflow, KubeVela will automatically create a default workflow for deploying the components when running the Application.

In VelaUX, you can feel the workflow more intuitively. As shown in the figure: The following is a workflow where the first step succeeds and the second fails:

velaux-workflow

Execution order

In the workflow, all the steps will be executed sequentially and the next step will be executed after the previous one succeeded. If a step is of type step-group, it can contain a series of sub-steps, all of which are executed together when the step group is executed.

In KubeVela 1.5.0+, you can explicitly specify the execution mode of steps, such as:

workflow:
mode:
steps: StepByStep
subSteps: DAG

There're two modes of execution: StepByStep and DAG.

If you do not explicitly declare the execution mode, by default steps are executed sequentially in StepByStep and subSteps are executed in parallel in DAG.

State of Application and Workflow

ApplicationWorkflowDescription
runningWorkflowexecutingWhen the workflow is executing, the status of the application is runningWorkflow
workflowSuspendingsuspendingWhen the workflow is suspended, the status of the application is workflowSuspending
workflowTerminatedterminatedWhen the workflow is terminated, the status of the application is workflowTerminated
workflowFailedfailedWhen a step in the workflow fails and the workflow is completed, the status of the application is workflowFailed
runningsucceededWhen all steps in the workflow are executed successfully, the status of the application is running

Core features

Workflow has powerful process control capabilities, including: