Storage::append() downloading content then appending

[code-creeper]

Hi, I am a laravel freamwork user from 3 years.

I want to append content in a file using s3.

But laravel Storage::append() download the content placed in the RAM then concatenate the new content with it the replace it with the file already placed in s3, please tell me If I am wrong or tell me any other way to do append file in s3 bucket for support chunk upload because cloude fare has limit my request 100mb.

Thanks,

[RizqyNurfauzella]

Your interpretation of Laravel's Storage::append() is correct: for an existing file, it reads the current contents and writes the combined contents back. It also inserts a separator (a newline by default), so it is not suitable for assembling binary upload chunks.

Use S3's multipart-upload API for this:

1. Your authenticated Laravel endpoint starts a CreateMultipartUpload and stores the upload ID with the intended object key.
2. Upload each chunk using UploadPart, retaining the same upload ID and assigning distinct part numbers.
3. Store each returned ETag and part number, then call CompleteMultipartUpload with that ordered list.
4. Abort an abandoned upload with AbortMultipartUpload.

Non-final parts must be at least 5 MiB. For your stated 100 MB request limit, 8 MiB parts are one possible choice, provided the total stays within S3's 10,000-part limit. You can also issue short-lived, scoped presigned UploadPart URLs so the browser uploads directly to S3. Configure bucket CORS for that flow, including exposing ETag; keep AWS credentials on the server.

This avoids downloading and rewriting the growing object after every chunk. Complete the parts as one multipart upload, rather than calling Storage::append() on a completed object.

References: Laravel append implementation, AWS multipart workflow, AWS part limits.

The implementation and documented API behavior were checked; this has not been tested against your bucket configuration.