Function Compute:Manage pipelines

Configure a pipeline

To configure a pipeline, you must configure the following parameters: Pipeline Trigger Mode, YAML, DingTalk Chatbot Notification Settings, and Pipeline Execution Environment. When you create an application and an environment, you can configure only the Pipeline Trigger Mode and Pipeline Execution Environment parameters. After you create an application and an environment, you can click Modify to the right of Pipeline Configurations on the Pipeline Management tab to modify the preceding four parameters.

Pipeline Trigger Mode

Serverless Application Center allows you to specify a method to trigger Git events of a pipeline. Serverless Application Center uses webhooks to receive Git events. When an event that meets specified trigger conditions is received, Serverless Application Center creates and executes the pipeline based on the YAML file that you specify. The following trigger modes are supported:

• Triggered by branch: The environment must be associated with a specific branch and all push events in the branch are matched.

• Triggered by tag: All tag creation events of a specific tag expression are matched.

• Triggered by branch merging: Merge and pull request events from the source branch to the destination branch with which the environment is associated are matched.

Pipeline Execution Environment

You can set this parameter to Default Environment or Custom Environment.

Default Environment

If you set the Pipeline Execution Environment parameter to Default Environment, Serverless Application Center fully manages pipeline resources and Function Compute bears fees generated during pipeline executions and you do not have to pay any fees. Each task of the pipeline runs on an independent and sandboxed virtual machine. Serverless Application Center ensures the isolation of the pipeline execution environment. The following items describe the limits that are imposed on the default environment:

• Instance specifications: 4 vCPUs and 8 GB of memory.

• Temporary disk space: 10 GB.

• Task execution timeout period: 15 minutes.

• Region: If you create an environment by using a template or source code from GitHub, specify the Singapore region. If you create an environment by using Gitee, GitLab, or Codeup, specify the China (Hangzhou) region.

• Network: You cannot use fixed IP addresses or CIDR blocks. You cannot configure an IP address whitelist to access specific websites. You cannot access resources in a virtual private cloud (VPC).

Custom Environment

If you set the Pipeline Execution Environment parameter to Custom Environment, the pipeline is executed in your Alibaba Cloud account. Compared with the default environment, a custom environment provides more custom capabilities. Serverless Application Center fully manages tasks in the custom environment and schedules Function Compute instances in real time to run the pipeline based on your authorization. Similar to the default environment, a custom environment provides the serverless feature, which eliminates the need to maintain infrastructure.

Custom environments provide the following custom capabilities:

• Region and network: You can specify the region and the VPC for the execution environment to access internal code repositories, artifact repositories, image repositories, and Maven internal repositories. For more information about supported regions, see Supported regions.

• Instance specification: You can specify CPU and memory specifications for the execution environment. For example, you can specify instances with larger specifications to speed up the build process.

  Note

  The ratio of vCPU capacity to memory capacity (in GB) ranges from 1:1 to 1:4.

• Persistence storage: You can configure File Storage NAS (NAS) or Object Storage Service (OSS) mounting. For example, you can use NAS to cache files to speed up the build process.

• Logging: You can specify a Simple Log Service project and Logstore to persist the execution logs of a pipeline.

• Timeout period: You can specify custom timeout periods for pipeline tasks. Default value: 600. Maximum value: 86400. Unit: seconds.

YAML

Serverless Application Center is integrated with Serverless Devs. You can use YAML files of Serverless Devs to define resource configurations of your applications. The default YAML file is named s.yaml. You can also specify another YAML file. After you specify the YAML file, you can use it in the pipeline by using the following methods:

• When you use the deployment plug-in @serverless-cd/s-deploy, the plug-in automatically uses the specified YAML file. The -t/--template command is added to the operation instructions of Serverless Devs. In the following example, the specified YAML file is demo.yaml, and the command executed by the plug-in is s deploy -t demo.yaml.

- name: deploy
  context:
    data:
      deployFile: demo.yaml
      steps:
      - plugin: '@serverless-cd/s-setup'
      - plugin: '@serverless-cd/checkout'
      - plugin: '@serverless-cd/s-deploy'
  taskTemplate: serverless-runner-task

• If you use scripts for execution, you can use ${{ ctx.data.deployFile }} to reference a specific YAML file. In the following example, the s plan command is run by using the specified file if you specify a YAML file. Otherwise, the s plan command is run by using the default s.yaml file.

- name: pre-check
  context:
    data:
      steps:
      - run: s plan -t ${{ ctx.data.deployFile || s.yaml }}
      - run: echo "s plan finished."
  taskTemplate: serverless-runner-task

DingTalk Chatbot Notification Settings

After you enable this feature, configure the following parameters: Webhook address, Signing key, Notify only on failure, and custom message. You can manage the tasks that require notifications in a centralized manner. You can also use the YAML file of the pipeline to enable notifications for each task. After you configure notifications, you can configure notifications for each task on the Pipeline Details section.

Pipeline Details section

In the Pipeline Details section, you can configure the pipeline flow, tasks in the pipeline, and relationships between the tasks. The platform automatically generates a default pipeline process, based on which you can modify configurations.

You can manage pipelines by using YAML. You can select Hosting on Platform or Read from Repository as the pipeline path.

Hosting on Platform

By default, the YAML files of pipelines are configured in and managed by Serverless Application Center in a centralized manner. After you update the YAML file of a pipeline, the new configurations take effect in the next deployment.

Read from Repository

The YAML file of the pipeline is stored in a Git repository. After you modify and save the configurations in the Function Compute console, Serverless Application Center commits the modifications to the Git repository in real time. This does not trigger execution of the pipeline. When an event in your code repository triggers the execution of the pipeline, Serverless Application Center creates and executes the pipeline by using the specified YAML file in the Git repository.

You can select Read from Repository in the upper part of the Pipeline Details section and enter the name of the YAML file. The following figure shows an example.

The Pipeline Details section consists of a tool area on the left and a YAML editing area on the right.

• YAML editing area: You can edit the YAML file of a pipeline to modify the pipeline. For more information, see Use a YAML file to describe a pipeline.

• The tool area provides auxiliary tools for editing YAML files. The following tabs are included:

  • Process Preview tab: You can preview and modify the pipeline process.

  • Task Template tab: You can view common YAML templates that are provided.

In the upper-right corner of the subsection, the Save, Fullscreen, and Reset buttons are displayed.

• You can click Save to save all modifications in the YAML file. Then, all saved data is synchronized to the YAML file of the pipeline.

• You can click Fullscreen to display the editing subsection as a full screen.

• You can click Reset to undo the most recent modifications to the YAML file.

View historical executions of a pipeline

You can click the execution version of a pipeline to view the execution details of the pipeline. Then, you can use the details to view the execution logs and status of the pipeline and troubleshoot issues.

Update the runtime of a pipeline

You can specify the runtime version of a pipeline by using the runtime-setup plug-in or modifying variables in the YAML file.