Function Compute:Use asynchronous invocations to more reliably handle time-consuming, resource-intensive, or error-prone requests, Asynchronous invocation

Common scenarios

The following items list common scenarios of asynchronous invocations:

• Audio and video processing

  You can use Function Compute to process audio and video files. During the processes, time-consuming tasks such as encoding, decoding, and transcoding are involved. You can trigger these tasks in asynchronous mode so that the tasks run in the backend and the frontend can proceed without waiting. This improves user experience. In addition, you can invoke multiple functions in asynchronous mode to implement concurrent processing and reduce the task processing time. For example, you can split your audio and video project into multiple tasks and process these tasks in parallel if your project is large. You can also convert a video to multiple formats in parallel.

• Data ETL processing

  In an extract, transform, and load (ETL) process, multiple independent operations, such as database querying, file reading and writing, and data cleansing, are involved. You can use asynchronous invocations to allow these operations to be performed in parallel, reducing processing time and improving system performance. For time-consuming tasks, such as processing of large-scale data sets and transformations of complex data, asynchronous invocations allow these tasks to run in the backend without waiting at the frontend. This improves user experience.

• Developing web applications

  Function Compute can be used with other cloud services to quickly build web applications. Asynchronous invocations can be used to prevent page freezing caused by lengthy backend response time when users submit forms, perform queries, and load a large amount of data. In this case, users can continue to interact with other parts of the page without experiencing delays. In high-concurrency scenarios, such as when a large number of users access an application at the same time, asynchronous invocations can distribute request loads and prevent services from being overloaded.

Deferred invocation

In specific scenarios, you may want Function Compute to defer a function invocation after you submit an asynchronous invocation request. In this case, you can use the Function Compute API or an SDK to defer the function invocation.

Add the HTTP request header x-fc-async-delay to the code. Valid values: (0,3600). Unit: seconds. Function Compute invokes the function when the period specified by x-fc-async-delay elapses.

Retry policy

Error handling and retry mechanisms are supported in asynchronous invocations. A task can be rescheduled without affecting the entire process if the task fails. Function Compute automatically retries an asynchronous invocation if the invocation fails.

Retry mechanism

The following table describes the default retry mechanism for common errors.

Configure a retry policy

You can specify the maximum number of retries and the maximum time to live (TTL) for messages in Function Compute.

1. Log on to the Function Compute console. In the left-side navigation pane, click Functions.

2. In the top navigation bar, select a region. On the Functions page, click the function that you want to manage.

3. On the function details page, click the Configurations tab. In the left-side navigation pane, click Asynchronous Mode Configurations. In the Asynchronous Mode Configurations section, click Modify.

4. In the Asynchronous Mode Configurations panel, configure the following parameters and click Deploy.

   Parameter	Description
   Task Mode	Specify whether to enable the asynchronous task mode. For more information, see Asynchronous tasks.
   Maximum Retries	Specify the maximum number of retries for messages in asynchronous invocations. Valid values: [0,8]. By default, Function Compute retries three times for messages that fail to be asynchronously triggered. You can reduce or increase the number of asynchronous invocation retries based on your business requirements.
   Maximum Message Lifetime	Specify the maximum lifetime of messages in asynchronous invocations. Valid values: [1,604800]. Default value: 86400. Unit: seconds. This duration is calculated from the time when an asynchronous invocation is triggered. If the maximum message lifetime elapses, the message is discarded. Discarded messages are tracked by the Asynchronous Invocation Trigger Events metric in CloudMonitor. For more information, see Monitoring metrics.

Result callback

After receiving an asynchronous invocation request, Function Compute persists the request and immediately returns a response without waiting for the request execution to complete. If you want to retain a request that fails to be executed after it is retried for the specified maximum number of times or if you want to notify downstream applications of the asynchronous invocation results, you can configure result callback. After you configure a destination service for asynchronous invocations, Function Compute automatically triggers a callback to the corresponding service based on the execution result.

How it works

The following figure shows the process of a result callback.

Scenarios

• Save discarded events for later use

  If an asynchronous request fails to be executed after the system retry the request based on the retry policy, Function Compute discards the request. If a destination is configured for failed invocations, Function Compute pushes the context information of the failed request to the destination service such as ApsaraMQ for RocketMQ for subsequent processing. You can also set the destination service to another function. Function Compute automatically pushes the context information of the failed request to this function to execute custom error handling logic.

• Notify downstream services of execution results

  You can configure a destination for successful invocations. After a request is successfully executed, Function Compute pushes the context information of the request to the downstream destination service. For example, if you use Function Compute to automatically decompress ZIP files that are uploaded to Object Storage Service (OSS), you can configure result callback for the decompression function if you want to receive notifications after decompression.

Supported destination services for asynchronous invocations

Take note of the following items when you configure the destination service for asynchronous invocations:

• Event content sent to the destination for the asynchronous invocation

  The following sample code provides an example of event content when the destination for the asynchronous invocation is Simple Message Queue (formerly MNS), Function Compute, or ApsaraMQ for RocketMQ.

  {
    "timestamp": 1660120276975,
    "requestContext": {
      "requestId": "xxx",
      "functionArn": "acs:fc:{regionid}:{accountid}:functions/xxxx",
         "condition": "FunctionResourceExhausted", 
          "approximateInvokeCount": 3
    },
    "requestPayload": "",
    "responseContext": {
      "statusCode": 200,
      "functionError": ""
    },
    "responsePayload": ""
  }

  Table 1. Parameter description

  Parameter	Description
  timestamp	The timestamp of the invocation.
  requestContext	The context of the request.
  requestContext.requestId	The request ID of the asynchronous invocation.
  requestContext.functionArn	The Alibaba Cloud Resource Name (ARN) of the function that is asynchronously invoked.
  requestContext.condition	The error code of the invocation.
  requestContext.approximateInvokeCount	The number of execution times of the asynchronous invocation. A value greater than 1 indicates that Function Compute retried the execution.
  requestPayload	The original payload of the function request.
  responseContext	The response context.
  responseContext.statusCode	The returned system status code. A status code other than 200 indicates that a system error occurred.
  responseContext.functionError	The error message of the invocation.
  responsePayload	The original payload that is returned after the function is executed.

  The following sample code provides an example of event content when the destination for asynchronous invocations is EventBridge. For more information, see Overview.

  {
      "datacontenttype": "application/json",
      "aliyunaccountid": "143xxxx",
      "data": {
          "requestContext": {
              "condition": "",
              "approximateInvokeCount": 1,
              "requestId": "0fcb7f0c-xxxx",
              "functionArn": "acs:fc:{regionid}:{accountid}:functions/xxxx"
          },
          "requestPayload": "",
          "responsePayload": "",
          "responseContext": {
              "functionError": "",
              "statusCode": 200
          },
          "timestamp": 1660120276975
      },
      "subject": "acs:fc:{regionid}:{accountid}:functions/xxxx",
      "source": "acs:fc",
      "type": "fc:AsyncInvoke:succeeded",
      "aliyunpublishtime": "2021-01-03T09:44:31.233Asia/Shanghai",
      "specversion": "1.0",
      "aliyuneventbusname": "xxxxxxx",
      "id": "ecc4865xxxxxx",
      "time": "2021-01-03T01:44:31Z",
      "aliyunregionid": "cn-shanghai-vpc",
      "aliyunpublishaddr": "199.99.xxx.xxx"
  }

• Payload limits

  The following items list the limits of payload sizes for different destination services:

  • Simple Message Queue (formerly MNS): 64 KB

  • Function Compute: 128 KB

  • EventBridge: 64 KB

  • ApsaraMQ for RocketMQ: 4 MB

• Loop avoidance

  When you configure a destination for asynchronous invocations, make sure that loop invocations do not occur. For example, a loop occurs if you configure Function B as the destination for successful asynchronous invocations of Function A and Function A as the destination for successful asynchronous invocations of Function B. An asynchronous invocation and execution of Function A may begin an infinite loop in which Function A and Function B continuously invoke each other.

FAQ

How do I trigger an asynchronous invocation of a function?

You can use one of the following methods to asynchronously invoke a Function Compute function:

• Log on to the Function Compute console, find the desired function, and then select Invoke the function in asynchronous mode on the Tests tab.

  

• Call the InvokeFunction operation and set the value of the x-fc-invocation-type parameter to Async.

• Use Serverless Devs to configure asynchronous invocations. Set the value of the invocation-type parameter to Async. For more information, see Invoke a function.

• Create a trigger that supports asynchronous invocations. For more information, see Trigger tasks by using events.

What to do next

If you want to obtain the status information of each phase of asynchronous requests, you can enable the task mode. For more information, see Asynchronous tasks.