Currently, files that users delete manually by mistake cannot be restored. However, you can enable versioning on your bucket to help recover future data lost due to accidental deletion or application failure. It is because versioning allows you to upload and store multiple versions of an object with the same name in your bucket, so that you can extract, delete, or restore a specified version of the object. For more information, see Setting Versioning.
An incomplete multipart upload is generated when you suspend or cancel an object upload. COS allows you to upload a large file greater than 5 GB in multiple file parts. During a multipart upload job, uploaded parts will be stored as an incomplete multipart upload and cannot be downloaded if you don’t call the Abort Multipart Upload or Complete Multipart Upload API.
Like objects, incomplete multipart uploads consume your storage capacity, and incur storage usage fees.
You can delete an incomplete multipart upload directly by using the COS console. For directions, see Deleting Incomplete Multipart Upload. Alternatively, you can regularly clear incomplete multipart uploads by configuring a lifecycle.
No, it won’t. It just deletes the unsuccessful uploads.
Check your browser or CDN for cached data. You can use the curl and wget commands to avoid browser caching. The cached CDN data can be purged in the CDN Console if you use a CDN domain name for access.
Check the configuration of the CDN-acceleration domain name by following the steps below.
a. If the bucket permission is private-read:
CDN Authentication | CDN Acceleration Domain Name | COS Domain Name | Scenarios |
---|---|---|---|
Disabled (default) | Denies access | COS authentication is required | Direct access to CDN domain names is allowed to protect the data on origin server. |
Enabled | URL authentication is required | COS authentication is required | Full stack strict SSL secured connection. Hotlink protection for CDN authentication is supported. |
b. If the bucket permission is public-read:
CDN Authentication | CDN Acceleration Domain Name | COS Domain Name | Scenarios |
---|---|---|---|
Disabled (default) | Allows access | Allows access | Allows public access to the entire website via CDN or origin server |
Enabled | Requires URL authentication | Allows access | Hotlink protection is enabled for access via CDN access, but not for access via origin server (not recommended) |
After confirming that the above configurations are correct, check the protocol used to access the CDN-acceleration domain name and the forced HTTPS configuration of the static website:
If the problem persists after all the above steps are completed, contact us for further troubleshooting by submitting a ticket.
Cross-origin access is to request resources over HTTP from a domain for another domain. The difference in any one of the protocol, domain name, and port will result in two different domains. For console operation steps, see Setting Cross Origin Resource Sharing or the Best Practice documentation Setting Cross Origin Resource Sharing.
Set the origin to *
when configuring CORS. For more information, see the Best Practice documentation Setting Cross Origin Resource Sharing.
Configure the CORS rule as shown below and try using a different browser to test whether it works. For more information, see Setting Cross Origin Resource Sharing.
If you are using a CDN acceleration domain name, configure CORS in the CDN console. For more information, see CDN Custom Response Header Configuration.
The V5 Console (XML) supports fuzzy match of second-level domain names, and the V4 Condole (JSON) does not support second-level wildcard domain names.
COS supports custom headers. For more information, see Custom Headers.
An origin-pull address is used for data migration. If a resource requested by the user does not exist on COS, it is pulled from the origin-pull address in real time.
Yes. COS will automatically pull and create the directory.
You can realize this by using COS with SCF. For more information, see Acquire Image on COS and Create a Thumbnail.
COS allows you to view the number and size of objects in the current folder. For more information, see Viewing Folder Details.
Once you enable Versioning on your bucket, you can store multiple versions of an object in the bucket, and extract, delete or recover a specified version of the object. For detailed directions, see Setting Versioning.
To do this, you can enable inventory, and check your generated inventory report. For more information, please see Enabling Inventory.
Was this page helpful?