Creating a Manual Workflow from Scratch


This section shows how to create a manually triggered manual workflow.

Before You Start

Ensure that you have created the codes and scripts needed for running the data as resources. For more information, see Creating a Resource.

Note

A manual workflow cannot be changed to periodic workflow after creation.

Step 1: Create a Workflow

  1. In the EnOS Management Console, click Batch Data Processing > Workflow Designer from the left navigation menu. Then, click + above the workflow list or click Create Workflow in the blank page to create a workflow.
  2. In the New Workflow window, provide the following settings and click OK.
    • Mode: Choose whether to create a workflow from scratch or by importing a configuration file.
    • Name: Enter the name of workflow.
    • Scheduling Type: Select Manual scheduling.
    • Description: Provide a description for the workflow.
    • Select Dir: Select the directory to store the workflow.

Step 2: (Optional) Add References

If the workflow is dependent on another workflow or task, reference it as the root node.

  • To reference a workflow, drag the Flow Ref under Reference into the workflow panel, select the workflow, and click Create.
  • To reference a task, drag the Task Ref under Reference into the workflow panel, select the task, and click Create.

Note

  • To reference a task node, you must set the task node to allow reference in the Node Scheduling Config of the task node.
  • A task node that allows reference must be a periodic task.
  • A workflow or task node is referenced by selecting its instance ID, which can be found on the Workflow Operation page (clicking the task name to view the instance ID).

Step 3: Add Task Nodes

Add a Data Synchronization Task Node

  1. From the Task Node panel, drag the Data Sync task node into the workflow panel.
  2. In the New Task Node window, enter the name and description of the task. Note that a task is unique within the same workflow.
  3. Select the type of the data to be synchronized: Structured data or File stream.
  4. Click Create.
  5. Double click the node that you just created and complete the configuration of the data synchronization task. For the detailed steps, see Data Synchronization Overview.
  6. Click Node Scheduling Config at the right edge of the configuration panel and provide the following scheduling settings.
    • Timeout: The timeout value.
    • Retry Times: The number of reconnection attempts to try after timeout.
    • Retry Interval: The interval between each reconnection attempt.
  7. If parameters are used in the task, click Parameter Settings at the right edge to specify the parameter values. You can use constants, system variables, or custom variables for a parameter. For more information, see Setting Parameters.
  8. Click the Running Mode at the right edge of the configuration panel and set the container resources that are required by running the task.
  9. Click Save and click Back to Workflow Panel.

Add a SHELL Task Node

  1. From the Task Node panel, drag the SHELL task node into the workflow panel.
  2. In the New Task Node window, enter the name and description of the task. Note that task is unique within the same workflow.
  3. Click Create.
  4. Double click the node that you just created.
  5. Enter the command to run, select the resource, and the resource version (the Shell script to be run must be uploaded through the Task Resources page).
  6. Click Node Scheduling Config at the right edge of the configuration panel and complete the scheduling attribute configuration.
  7. If parameters are used in the task, click Parameter Settings at the right edge to specify the parameter values. You can use constants, system variables, or custom variables for a parameter. For more information, see Setting Parameters.
  8. Click the Running Mode at the right edge of the configuration panel and set the running mode of the Shell task:
    • Single Task: Set the container resources that are required by running the task.
    • Multiple Tasks: Set the container resources that are required by running the task, the maximum concurrency, and specify the source of the distribution key (from external files or custom key).
  9. Click Save and click Back to Workflow Panel.

Add an External APP Task Node

  1. From the Task Node panel, drag the External APP task node into the workflow panel.
  2. In the New Task Node window, enter the name and description of the task. Note that task is unique within the same workflow.
  3. Click Create.
  4. Double click the node that you just created.
  5. Enter the domain name of the application and the command to run (if extra parameters are needed in the command, configure the parameter in the Parameter Config at the right edge of the configuration panel).
  6. Click Node Scheduling Config at the right edge of the configuration panel and complete the scheduling attribute configuration.
  7. If parameters are used in the task, click Parameter Settings at the right edge to specify the parameter values. You can use constants, system variables, or custom variables for a parameter. For more information, see Setting Parameters.
  8. Click Save and click Back to Workflow Panel.

Add a PYTHON Task Node

  1. From the Task Node panel, drag the PYTHON task node into the workflow panel.
  2. In the New Task Node window, enter the name and description of the task. Note that task is unique within the same workflow.
  3. Click Create.
  4. Double click the node that you just created.
  5. In the Command field, enter the command to run, and select the corresponding resource and the resource version (The Python script to be run must be already uploaded through the Task Resources page. For more information about how to access data sources by Python, see Accessing Data Sources by Python).
  6. Click Node Scheduling Config at the right edge of the configuration panel and complete the scheduling attribute configuration.
  7. If parameters are used in the task, click Parameter Settings at the right edge to specify the parameter values. You can use constants, system variables, or custom variables for a parameter. For more information, see Setting Parameters.
  8. Click the Running Mode at the right edge of the configuration panel and set the running mode of the Python task.
    • Image: You can choose the default, Showflake or Pgsql image service to manipulate the data in the database.
    • Single Task: Set the container resources that are required by running the task.
    • Multiple Tasks: Set the container resources that are required by running the task, the maximum concurrency, and specify the source of the distribution key (from external files or custom key).
  9. Click Save and click Back to Workflow Panel.

Step 4: Add Relations

  1. From the Component panel, click Relation.
  2. Move the cursor to the upstream task node, click and drag the cursor, and an arrowhead will appear.
  3. Drag the arrowhead to the downstream task node.
  4. Repeat steps 2 and 3 until you finish connecting all the nodes according to your design.

Note

References must always be the root node.

Step 5: Configure Scheduling Settings

  1. Click Scheduling Config from the right edge of the configuration panel.
  2. Provide the basic settings for the below.
    • Task Name: If you want to change the workflow name, enter the new name.
    • Queue: (Optional) Enter the name of batch processing queue resource for running the workflow (which can be requested through Resource Management).
    • Owner: Select the workflow owner from the list of users in the organization who have access to the Data Synchronization service. The workflow creator is added by default.
      • As the creator, you cannot delete yourself.
      • You can add other owners in the same organization.
      • The same owner is not allowed to create another workflow with the same name.
    • Description: (Optional) Provide a description.
    • Alert Mode: Select how to alert the workflow owner. Email is always selected by default, and cannot be unselected.
      • E-mail: an alert e-mail is sent to the owner when an instance meets the alert conditions. When the task fails, a copy of the alert is also sent to the affected downstream task owner, if an owner is assigned.
      • SMS: To use this feature, the user’s phone number must be verified during user registration. Note that the SMS alert is sent to only the owner when an instance meets the alert conditions.

Step 6: Specify Parameters

If parameters are used when you configure the data source and target, specify the parameter values by clicking Parameter Config at the right edge of the configuration panel. For more information, see Setting Parameters.

Next Step

Click Pre-run to test the result of the workflow.


After a workflow starts to run, an instance will be generated. You can then trace the details about the instance through Workflow Operation. For more information, see Monitoring Manual Workflow Instances.