All Products
Document Center


Last Updated:Jul 04, 2024

Prefetches content from origin servers to points of presence (POPs). This reduces loads on origin servers because users can directly hit cache upon their first visits.

Operation description

  • Alibaba Cloud CDN supports POST requests in which parameters are sent as a form.
  • You can call the RefreshObjectCaches operation to refresh content and call the PushObjectCache operation to prefetch content.
  • By default, each Alibaba Cloud account can submit up to 1,000 URLs per day. If the daily peak bandwidth value of your workloads exceeds 200 Mbit/s, you can submit a ticket to increase your daily quota. Alibaba Cloud reviews your application and then increases the quota accordingly.
  • You can specify at most 100 URLs in each prefetch request.
  • For each Alibaba Cloud account, the prefetch queue can contain up to 50,000 URLs. Content is prefetched based on the time when the URLs are submitted. The URL that is submitted the earliest has the highest priority. If the number of URLs in the queue reaches 50,000, you cannot submit more URLs until the number drops below 50,000.
  • You can call this operation up to 50 times per second per account.
  • For more information about how to automate refresh or prefetch tasks, see Run scripts to refresh and prefetch content.


  • After a prefetch task is submitted and completed, the POPs immediately start to retrieve resources from the origin server. Therefore, a large number of refresh tasks cause a large number of concurrent download tasks. This increases the number of requests that are redirected to the origin server. The back-to-origin routing process consumes more bandwidth resources and the origin server may be overwhelmed.
  • The time required for a prefetch task to complete is proportional to the size of the prefetched file. In actual practice, most prefetch tasks require 5 to 30 minutes to complete. A task with a smaller average file size requires less time.
  • To allow RAM users to perform this operation, you must first grant them the required permissions. For more information, see Authorize a RAM user to prefetch and refresh resources.


OpenAPI Explorer automatically calculates the signature value. For your convenience, we recommend that you call this operation in OpenAPI Explorer.

Authorization information

There is currently no authorization information disclosed in the API.

Request parameters


The URLs based on which content is prefetched. Format: accelerated domain name/files to be prefetched.

Note Separate URLs with line feeds (\n or \r\n). Each object path can be up to 1,024 characters in length.\

The accelerated region where content is to be prefetched. Valid values:

  • domestic**: Chinese mainland**
  • overseas**: regions outside the Chinese mainland**

If you do not set this parameter, content in the accelerated region of the domain name is prefetched.

  • If the accelerated region is set to Mainland China Only, content in regions in the Chinese mainland is prefetched.
  • If the accelerated region is set to Global, content in all regions is prefetched.
  • If the accelerated region is set to Global (Excluding Mainland China), content in regions outside the Chinese mainland is prefetched.

Specifies whether to prefetch content to POPs. Valid values:

  • true: prefetches content to POPs.
  • false: prefetches content to regular POPs. Regular POPs can be L2 POPs or L3 POPs. Default value: false.

The custom header for prefetch in the JSON format.

{ "Accept-Encoding": [ "gzip" ] }

Response parameters


The ID of the prefetch task. If multiple tasks are returned, the IDs are separated by commas (,). The task IDs are merged based on the following rules:

  • If the tasks are set for the same accelerated domain name, submitted within the same second, and prefetch content from URLs instead of directories, the tasks IDs are merged into the same task ID (RushTaskId).
  • If the number of tasks that are set for the same accelerated domain name, submitted within the same second, and prefetch content from URLs instead of directories exceeds 500, every 500 task IDs are merged into the same task ID (RushTaskId).

The ID of the request.



Sample success responses


  "PushTaskId": "9524xxxx",
  "RequestId": "16A96B9A-F203-4EC5-8E43-CB92E68F4CD8"

Error codes

HTTP status codeError codeError messageDescription
400SingleRequest.OverLimitA maximum of 1000 URLs are supported for each request.-
400QuotaExceeded.PreloadYour preload attempts have exceeded the daily limit.The maximum number of URL prefetches on the current day is exceeded.
400InvalidObjectPath.MalformedThe specified ObjectPath is invalid.-
400InvalidExtensiveDomain.ValueNotSupportedThe specified ExtensiveDomain is not supported.-
400PreloadQueueFullThe warming queue is full,please try again later.-
400TooManyRequestsToo many requests, please try again later-
400PreloadQueueFullPreload queue is full, please try again later.-
400QuotaPerMinuteExceeded.RefreshYou have exceeded the prescribed preload limits per minute.-
400InvalidObjectPath.ExceedsMaximumThe maximum number of urls is exceeded.-
400InvalidCustomHeaderParse preload header failed.-
400InvalidCustomHeaderUnsupported Preload headers included.-

For a list of error codes, visit the Service error codes.

Change history

Change timeSummary of changesOperation
2023-07-25The Error code has changedView Change Details