CDN is capable of configuring basic cache. Cache validity can be configured according to rules such as specified service types, directories, and specific URLs to regularly purge resources cached on nodes, pull the latest resources from the origin server, and cache them again.
In addition, CDN can purge cache for specified URLs or directories in batches:
Last-Modify
information of the resource will be pulled from the origin server. If it is the same as that of the cached resource, the cached resource will be directly returned; otherwise, the updated resource will be pulled from the origin server and cached again. If you select Purge all resources, when the user accesses a resource under the corresponding directory, the latest version of the resource will be directly pulled from the origin server and cached again.Note:
After a purge is successfully executed, the corresponding resource on the node will not have a valid cache. When the user initiates an access request again, the node will pull the required resource from the origin server and cache it on the node. If you submit a large number of purge tasks, many caches will be cleared, resulting in a surge in origin-pull requests and high pressure on the origin server.
When a resource is overwritten by a new resource with the same name on the origin server, you can submit a request to purge the URL/directory and clear all caches so users can directly access the latest version of the resource. This will prevent users from accessing the legacy version of the resource cached on the node.
When illegal resources (such as resources related to pornography, drug, or gambling) are found on your origin server, they may still be accessible even after you delete them on the origin server because of node cache. To protect your network environment security, you can delete the cached resources through URL purge.
Log in to the CDN console, click Purge and Prefetch on the left sidebar, and submit a Purge URL or Purge Directory task:
Note:
If the URL encode feature is enabled, Chinese characters will be converted to URL encoded format during URL and directory purge.
In the History tab, you can query tasks by a specified time period, term, and task type. Term queries only support querying with a domain name or a complete purged URL/directory:
Note:
The console can return up to 10,000 operation records at a time, which can be exported to Excel. If you have a high number of purge tasks, please query and export them in batches.
URL purge:
http://
or https://
protocol identifier when submitting a purge task.http://*.test.com/
cannot be purged. Even if you connect a wildcard domain name to CDN, you need to submit the corresponding subdomain names for purge.Directory Purge:
http://
or https://
protocol identifier when submitting a purge task.http://*.test.com/
cannot be purged. Even if you connect a wildcard domain name to CDN, you need to submit the corresponding subdomain names for purge.Sub-user permissions configuration:
The acceleration domain name is purge-test-1251991073.file.myqcloud.com
, the origin server is Tencent Cloud Object Storage (COS), and resources on the origin server are as follows:
1.txt
and 2.txt
respectively. Nodes to be hit can be determined based on X-Cache-Lookup: Hit From Distank3
and Server: NWS_SPMid
, resources will be directly returned by the nodes:1.txt
with a file that has the same name, and the file's last modified time changes, while 2.txt
stays the same:1.txt
resource will be accessed:Last-Modified
of 1.txt
has been changed, the request will be forwarded to the origin server. As 2.txt
has not been changed, even after a directory purge task is submitted, it will still be hit and returned:
Was this page helpful?