使用自定义版本ID将对象上传到Amazon S3

Question

I want to upload an object to Amazon versioned bucket (using Java AWS SDK) and set a custom version to this object (goal is to set the same version to all objects, uploaded at once)

PutObjectResult por = amazonS3Client.putObject(...);
por.setVersionId("custom_version");

So, is it the right way to set a version to the uploaded object?

Does this code lead to 2 separate requests to Amazon?

What if Internet is broken while por.setVersionId(..) is being called?

Why por.setVersionId(..) does not throw an exception such as SdkClientException if this method really is trying to set a version ID on Amazon server?

Answer 1

setVersionId would be something the SDK library itself uses to populate the versionId returned by the service when the object is created, so that you can retrieve it if you want to know what it is.

Version IDs in S3 are system-generated opaque strings that uniquely identify a specific version of an object. You can't assign them.

The documentation uses some unfortunate examples like "111111" and "222222," which do not resemble real version-ids. There's a better example further down the page, where you'll find this:

Unique version IDs are randomly generated, Unicode, UTF-8 encoded, URL-ready, opaque strings that are at most 1024 bytes long. An example version ID is 3/L4kqtJlcpXroDTDmJ+rmSpXd3dIbrHY+MTRCxf3vjVBH40Nr8X8gdRQBpUMLUo.

Only Amazon S3 generates version IDs.

They cannot be edited.

You don't get an error here because all this method does is set the versionId inside the PutObjectResult object in local memory after the upload has finished. It succeeds, but serves no purpose.

To store user-defined metadata with objects, such as your release/version-id, you'd need to use object metadata ( x-amz-meta-* ) or the new object tagging feature in S3.