Known limitations¶

This documentation tracks any compability issues that the object storage service in Binero cloud exhibits compared to other implementations and interoperability between the S3 and Swift APIs.

Browser Uploads and Chunked Uploads in S3 API¶

The authentication backend used for Binero cloud does not implement all required details from the AWSv4 specification and therefore the Browser Uploads and Chunked Uploads features in the S3 API is unsupported.

This is normally not an issue but we’ve had customers utilizing the official AWS C# SDK encountering the issue with Chunked Uploads because UseChunkEncoding in the PutObjectRequest methods defaults to true, this needs to be set to false.

Using the Swift API to download a S3 API multipart upload¶

A single upload of a file is limited to 5 GB by design and to upload a larger file than that the multipart upload features in for example the S3 API is used. This is usually handled by the client or library that you use and is not something that you need to thing about, this splits your file into multiple parts and then uploads them concurrently and the parts get’s assembled in the platform.

If you upload a large file using the S3 API and then try to download that using the Swift API you will get an error like Error downloading [object-name]: md5sum != etag indicating the operation failed, but the object is successfully downloaded. We recommend that you select one API to use for your application and not combining the usage unless absolutely neccesary.