AWS S3 Multipart

The @sme-uploader/aws-s3-multipart plugin can be used to upload files directly to an S3 bucket using S3’s Multipart upload strategy. With this strategy, files are chopped up in parts of 5MB+ each, so they can be uploaded concurrently. It is also very reliable: if a single part fails to upload, only that 5MB chunk has to be retried.

const AwsS3Multipart = require('@sme-uploader/aws-s3-multipart')
uploader.use(AwsS3Multipart, {
  limit: 4,
  companionUrl: 'https://uploader-companion.myapp.net/'
})

Installation

This plugin is published as the @sme-uploader/aws-s3-multipart package.

Install from NPM:

npm install @sme-uploader/aws-s3-multipart

In the CDN package, it is available on the SmeUploader global object:

const AwsS3Multipart = SmeUploader.AwsS3Multipart

Options

The @sme-uploader/aws-s3-multipart plugin has the following configurable options:

limit: 0

The maximum amount of chunks to upload simultaneously. Set to 0 to disable limiting.

retryDelays: [0, 1000, 3000, 5000]

When uploading a chunk fails, automatically try again after the millisecond intervals specified in this array. By default, we first retry instantly; if that fails, we retry after 1 second; if that fails, we retry after 3 seconds, etc.

Set to null to disable automatic retries, and fail instantly if any chunk fails to upload.

companionUrl: null

URL of the Companion instance to use for proxying calls to the S3 Multipart API.

This will be used by the default implementations of the upload-related functions below. If you provide your own implementations, a companionUrl is unnecessary.

companionHeaders: {}

Custom headers that should be sent along to Companion on every request.

This will be used by the default implementations of the upload-related functions below. If you provide your own implementations, these headers are not sent automatically.

getChunkSize(file)

A function that returns the minimum chunk size to use when uploading the given file.

The S3 Multipart plugin uploads files in chunks. Each chunk requires a signing request (prepareUploadPart()). To reduce the amount of requests for large files, you can choose a larger chunk size, at the cost of having to re-upload more data if one chunk fails to upload.

S3 requires a minimum chunk size of 5MB, and supports at most 10,000 chunks per multipart upload. If getChunkSize() returns a size that’s too small, SME Uploader will increase it to S3’s minimum requirements.

createMultipartUpload(file)

A function that calls the S3 Multipart API to create a new upload. file is the file object from SME Uploader’s state. The most relevant keys are file.name and file.type.

Return a Promise for an object with keys:

• uploadId - The UploadID returned by S3.
• key - The object key for the file. This needs to be returned to allow it to be different from the file.name.

The default implementation calls out to Companion’s S3 signing endpoints.

listParts(file, { uploadId, key })

A function that calls the S3 Multipart API to list the parts of a file that have already been uploaded. Receives the file object from SME Uploader’s state, and an object with keys:

• uploadId - The UploadID of this Multipart upload.
• key - The object key of this Multipart upload.

Return a Promise for an array of S3 Part objects, as returned by the S3 Multipart API. Each object has keys:

• PartNumber - The index in the file of the uploaded part.
• Size - The size of the part in bytes.
• ETag - The ETag of the part, used to identify it when completing the multipart upload and combining all parts into a single file.

The default implementation calls out to Companion’s S3 signing endpoints.

prepareUploadPart(file, partData)

A function that generates a signed URL to upload a single part. Receives the file object from SME Uploader’s state. The partData argument is an object with keys:

• uploadId - The UploadID of this Multipart upload.
• key - The object key in the S3 bucket.
• body - A Blob of this part’s contents.
• number - The index of this part in the file (PartNumber in S3 terminology).

Return a Promise for an object with keys:

• url - The presigned URL to upload a part. This can be generated on the server using the S3 SDK like so:

  sdkInstance.getSignedUrl('uploadPart', { Bucket: 'target', Key: partData.key, UploadId: partData.uploadId, PartNumber: partData.number, Body: '', // Empty, because it is uploaded later Expires: 5 * 60, }, (err, url) => { /* there's the url! */ })

• headers - (Optional) Custom headers that should be sent to the S3 presigned URL.

abortMultipartUpload(file, { uploadId, key })

A function that calls the S3 Multipart API to abort a Multipart upload, and delete all parts that have been uploaded so far. Receives the file object from SME Uploader’s state, and an object with keys:

• uploadId - The UploadID of this Multipart upload.
• key - The object key of this Multipart upload.

This is typically called when the user cancels an upload. Cancellation cannot fail in SME Uploader, so the result of this function is ignored.

The default implementation calls out to Companion’s S3 signing endpoints.

completeMultipartUpload(file, { uploadId, key, parts })

A function that calls the S3 Multipart API to complete a Multipart upload, combining all parts into a single object in the S3 bucket. Receives the file object from SME Uploader’s state, and an object with keys:

• uploadId - The UploadID of this Multipart upload.
• key - The object key of this Multipart upload.
• parts - S3-style list of parts, an array of objects with ETag and PartNumber properties. This can be passed straight to S3’s Multipart API.

Return a Promise for an object with properties:

• location - (Optional) A publically accessible URL to the object in the S3 bucket.

The default implementation calls out to Companion’s S3 signing endpoints.

S3 Bucket Configuration

S3 buckets do not allow public uploads by default. In order to allow SME Uploader to upload to a bucket directly, its CORS permissions need to be configured.

This process is described in the AwsS3 documentation.

While the SME Uploader AWS S3 plugin uses POST requests while uploading files to an S3 bucket, the AWS S3 Multipart plugin uses PUT requests when uploading file parts. Additionally, the ETag header must also be whitelisted:

<CORSRule>
  <!-- Change from POST to PUT if you followed the docs for the AWS S3 plugin ... -->
  <AllowedMethod>PUT</AllowedMethod>

  <!-- ... keep the existing MaxAgeSeconds and AllowedHeader lines and your other stuff ... -->

  <!-- ... and don't forget to add this tag. -->
  <ExposeHeader>ETag</ExposeHeader>
</CORSRule>