Can I stream a file upload to S3 without a content-length header?

Question

I'm working on a machine with limited memory, and I'd like to upload a dynamically generated (not-from-disk) file in a streaming manner to S3. In other words, I don't know the file size when I start the upload, but I'll know it by the end. Normally a PUT request has a Content-Length header, but perhaps there is a way around this, such as using multipart or chunked content-type.

S3 can support streaming uploads. For example, see here:

http://blog.odonnell.nu/posts/streaming-uploads-s3-python-and-poster/

My question is, can I accomplish the same thing without having to specify the file length at the start of the upload?

Solution 1

You have to upload your file in 5MiB+ chunks via S3's multipart API. Each of those chunks requires a Content-Length but you can avoid loading huge amounts of data (100MiB+) into memory.

• Initiate S3 Multipart Upload.
• Gather data into a buffer until that buffer reaches S3's lower chunk-size limit (5MiB). Generate MD5 checksum while building up the buffer.
• Upload that buffer as a Part, store the ETag (read the docs on that one).
• Once you reach EOF of your data, upload the last chunk (which can be smaller than 5MiB).
• Finalize the Multipart Upload.

S3 allows up to 10,000 parts. So by choosing a part-size of 5MiB you will be able to upload dynamic files of up to 50GiB. Should be enough for most use-cases.

However: If you need more, you have to increase your part-size. Either by using a higher part-size (10MiB for example) or by increasing it during the upload.

First 25 parts:   5MiB (total:  125MiB)
Next 25 parts:   10MiB (total:  375MiB)
Next 25 parts:   25MiB (total:    1GiB)
Next 25 parts:   50MiB (total: 2.25GiB)
After that:     100MiB

This will allow you to upload files of up to 1TB (S3's limit for a single file is 5TB right now) without wasting memory unnecessarily.

---

A note on your link to Sean O'Donnells blog:

His problem is different from yours - he knows and uses the Content-Length before the upload. He wants to improve on this situation: Many libraries handle uploads by loading all data from a file into memory. In pseudo-code that would be something like this:

data = File.read(file_name)
request = new S3::PutFileRequest()
request.setHeader('Content-Length', data.size)
request.setBody(data)
request.send()

His solution does it by getting the Content-Length via the filesystem-API. He then streams the data from disk into the request-stream. In pseudo-code:

upload = new S3::PutFileRequestStream()
upload.writeHeader('Content-Length', File.getSize(file_name))
upload.flushHeader()

input = File.open(file_name, File::READONLY_FLAG)

while (data = input.read())
  input.write(data)
end

upload.flush()
upload.close()

Solution 2

Putting this answer here for others in case it helps:

If you don't know the length of the data you are streaming up to S3, you can use S3FileInfo and its OpenWrite() method to write arbitrary data into S3.

var fileInfo = new S3FileInfo(amazonS3Client, "MyBucket", "streamed-file.txt");

using (var outputStream = fileInfo.OpenWrite())
{
    using (var streamWriter = new StreamWriter(outputStream))
    {
        streamWriter.WriteLine("Hello world");
        // You can do as many writes as you want here
    }
}

Solution 3

You can use the gof3r command-line tool to just stream linux pipes:

$ tar -czf - <my_dir/> | gof3r put --bucket <s3_bucket> --key <s3_object>

Solution 4

If you are using Node.js you can use a plugin like s3-streaming-upload to accomplish this quite easily.

Solution 5

Refer more on HTTP multi-part enitity requests. You can send a file as chunks of data to the target.

View more solutions

Comments

[Radim]

The smart_open Python library does that for you (streamed read and write).

[Ermiya Eskandary]

10 years later & the AWS S3 SDKs still don't have a managed way to do this - as someone who is hugely invested in the AWS ecosystem, it's very disappointing to see this in comparison to object management SDKs provided by other cloud providers. This is a core feature missing.

[Steve K]

Is there a Java equivalent of these classes?

[sigget]

A java implementation of this in the form of an OutputStream exists in s3distcp github.com/libin/s3distcp/blob/master/src/main/java/com/amaz‌​on/…

[Alex Hall]

I've created an open source library dedicated to this at github.com/alexmojaki/s3-stream-upload

[at0mzk]

isnt the length of "Hello world" known? does it work if the input is a stream?

[Landon Kuhn]

Where did you find the 5MiB limit?

[chrismarx]

Looks like you can also use the cli now with pipe - github.com/aws/aws-cli/pull/903

[Admin]

is there a way to just do tar -czf - <my_dir/> | aws s3 --something-or-other ?

[xiaochuanQ]

not supported in dotnet core, since the synchronous nature of Amazon.S3.IO apis, per Microsoft.

[Tushar Kolhe]

@AlexHall any python implementation?

[Alex Hall]

@TusharKolhe googling "python stream multipart upload s3" I found stackoverflow.com/questions/31031463/… and stackoverflow.com/questions/52825430/… and it looks like there were more results

[Tushar Kolhe]

@AlexHall thanx i figured out the way, this is the actual problem that i m trying to solve stackoverflow.com/questions/61696155/…. In case of a file already on disk i m able to do this.. but i want to upload streaming frames