After a workflow runs, Realtime Compute for Apache Flink creates a workflow instance for each execution. This topic explains how to view workflow instance history, inspect a task instance, and take action when a task fails or needs to be rerun.
Prerequisites
Before you begin, make sure you have:
-
A workflow that has been created and run. For more information, see Create a workflow and Run a workflow.
Navigate to a workflow details page
Both procedures in this topic start from the workflow details page. Complete these steps once before viewing workflow instances or task instances.
-
Log on to the Realtime Compute for Apache Flink console.
-
Find the workspace you want to manage and click Console in the Actions column.
-
In the left-side navigation pane, click O&M > Workflows.
-
Find the workflow and click its name to open the details page.
View workflow instance history
On the workflow details page, the Overview tab lists all workflow instances. For each instance, the tab shows the execution time and status.
To view the diagram of a specific workflow instance, click the value in the ID column in the Instance History section. The Diagrams tab opens and displays the workflow topology. Each task instance node is color-coded to reflect its state. For a full color-to-state reference, see Workflow status.
Alternatively, click the Diagrams tab directly to view the workflow instance diagram.
View and manage a task instance
-
On the Overview tab, click the value in the ID column for a workflow instance in the Instance History section. The Diagrams tab opens.
-
Click the task instance you want to inspect. A dialog box appears with the task instance details.
From the dialog box, take any of the following actions:
Action When to use Description Show Task Logs Diagnosing task behavior View the operational logs of the task instance. Show Related Job Details Inspecting the underlying job Redirects to the Deployments page to view deployment details. Re-Run Rerunning a task after fixing an issue Reruns the task. Select Execute the same operation for downstream tasks to rerun all downstream tasks in the same operation. Set as Success Skipping a failed task intentionally Sets the state of a failed task to SUCCESS. Select Execute the same operation for downstream tasks to mark all downstream tasks as SUCCESS as well. Force Stop Terminating a task instance Terminates the task instance.
What's next
-
To understand Workflows concepts such as states and task dependencies, see Workflows (public preview).
-
To isolate and manage resources with queues, see Manage resource queues.
-
To create a batch deployment of the SQL, JAR, or Python type, see Deploy a job.