Smoke testing

更新时间:
复制 MD 格式

To ensure your scheduling nodes execute as expected, we recommend performing smoke testing. This process verifies that a node's parameter substitution logic and execution results are correct, preventing basic configuration errors from affecting the production environment.

Supported node types

Smoke testing can be performed during task deployment for the following node types:

Category

Node type

Data Integration

batch synchronization

MaxCompute

MaxCompute SQL, MaxCompute Script, PyODPS 2, PyODPS 3, MaxCompute MR, SQL script template

Hologres

Hologres SQL

EMR

EMR Hive, EMR Impala, EMR MR, EMR Presto, EMR Shell, EMR Spark, EMR Trino, EMR Kyuubi

CDH

CDH Hive, CDH Spark, CDH Spark SQL, CDH MR, CDH Presto, CDH Impala

Lindorm

Lindorm Spark, Lindorm Spark SQL

ClickHouse

ClickHouse SQL

ADB

ADB Spark, ADB Spark SQL

Data Quality

data quality monitoring

General

shell node, Function Compute, SSH

Algorithm

PAI DLC, PAI Designer

Enable mandatory smoke testing

A workspace administrator can enable mandatory smoke testing. When this feature is enabled, all tasks in the workspace must pass a smoke test before they can be deployed to the production environment.

  1. Go to the Workspaces page in the DataWorks console. In the top navigation bar, select a desired region. Find the desired workspace and choose Shortcuts > Data Studio in the Actions column.

  2. In the left sidebar of Data Studio, click image > Data Studio Settings.

  3. On the Data Studio settings page, switch to the Security Settings and Others tab. In the Smoke test section, turn on the Mandatory Smoke Testing switch.

Perform a smoke test

  1. After you finish developing the code for a node, click Publish in the toolbar.

  2. Initiate the smoke test.

    1. In a Create Deploy Task, which has a single compute resource, you must click Create Deploy Task before you can click Initiate Smoke Testing.

    2. For a Initiate Smoke Testing: The development and production environments are isolated. Before deploying to the production environment, click Initiate Smoke Testing. The test runs in the development environment.

      This button is located in the fourth stage of the deployment pipeline: the production checker.

  3. On the smoke testing page, select a data timestamp.

    Note

    If you select today or yesterday as the business date for a smoke test, the task will wait until its scheduled time to run. For example: The current time is 2024-06-02 12:00, and the task's scheduled run time is 15:00. If you select 2024-06-01 as the business date for the test, the task's actual run time is 2024-06-02 15:00. In this case, the smoke test task will have a waiting period because its scheduled time of 15:00 has not yet been reached.

View smoke testing records

  1. Go to the smoke testing records page.

    • Method 1: In the left sidebar of Data Studio, click image.

    • Method 2: On the deployment page, after you initiate a smoke test, click image.

      This eye-shaped icon is located to the right of the Initiate Smoke Testing button in the production checker stage. The tooltip for the icon is View smoke testing records.

  2. On the Smoke Testing Records page, you can view details such as Tested At, Test Status, Tested By, and Data Timestamp.

  3. In a smoke test record, click Logs to view the execution log. For a record with a Status of Running, you can click Stop to end the run.