Upload Limit
Definition
An upload limit governs one transaction: the request, or set of requests, that carries a file from your machine to the host. It appears in several forms — a maximum request body size enforced by the web server, a cap on how many files one batch may contain, a per-minute rate limit on the upload endpoint, or a timeout that ends a transfer taking too long. It is easily confused with the file size limit, which asks how large one finished object may be, and with the storage quota, which asks how much you may hold altogether. A 100 MB request cap and a 100 MB per-file rule look identical right up until you drop ten 30 MB files in at once. The usual way past a request-level ceiling is not a bigger plan but a different method: break the transfer into pieces.
Why It Matters
Upload failures waste the most expensive minutes in the whole process, because they happen after the bytes have already travelled. A 400 MB archive over a 5 Mbps upstream connection takes roughly eleven minutes; finding out at the end that the request cap was 256 MB means starting again with nothing salvaged. Batch limits fail more quietly — a folder of 600 images dropped into a box that accepts 500 may take the first 500 and report success, and you notice the gap weeks later. Knowing which shape of limit you are against tells you whether to compress, to split, or to move to an API that can resume.
How It Works
A browser upload is normally a multipart form POST, and the server checks the declared Content-Length against its configured maximum before it reads the body. Over the limit, it answers 413 Payload Too Large and closes the connection — often before the transfer has finished, which is why a big upload can die at ninety per cent with no warning. Rate limits are counted per API key or per IP across a rolling window and answered with 429 Too Many Requests plus a Retry-After header saying how long to wait. A chunked upload sidesteps the body cap altogether by cutting the file into pieces of a few megabytes, sending each as its own request, and asking the host to reassemble them; when a piece fails you resend that piece, not the file. Resumable protocols add an offset on top, so an interrupted transfer picks up where it stopped.
Real-World Example
A studio tries to push a 900 MB video archive to 99helpers as one ZIP and watches it fail twice on hotel Wi-Fi. Sent as a chunked transfer instead, the same archive goes up in 5 MB pieces, and when the connection drops at piece 140 only that piece is resent. The finished file lands at studio-reel.99helpers.site with the address it would have had either way. What changed was twenty minutes of retries that never happened.
Common Mistakes
- ✕Reading a per-file figure as a per-request one — ten 30 MB files in one batch is a 300 MB request, and the server measures the request
- ✕Retrying a failed large upload unchanged — the same ceiling is still in place and the second attempt dies at the same point
- ✕Ignoring Retry-After on a 429 and hammering the endpoint — the window resets each time and the block lasts far longer than waiting would have
Related Terms
File Size Limit
A file size limit is the largest a single file may be for a host to accept and store it. It is measured one file at a time, so it says nothing about how many files you may keep or how much traffic they will generate.
Storage Quota
A storage quota is the total volume of data your account may hold at any one moment. It is a standing measurement of what sits on disk, and downloading a file does not change it.
Chunked Upload
Chunked upload is the technique of splitting one large file into smaller pieces and sending each piece as its own request. If the connection drops, only the piece in flight is lost rather than the whole transfer.
Multipart Upload
Multipart upload is the object-storage protocol, originally from Amazon S3 and now copied widely, for sending one large object as numbered parts that share an upload ID. A final completion call lists every part number with its ETag and turns the parts into a single object.
HTTP Status Code
An HTTP status code is the three-digit number a server returns at the head of every response, saying what happened to the request. It is the first thing on the first line — a 200 means here is the file, a 404 means there is nothing at that address.
Put a file online in seconds
Drop in a document, an image, a page or a whole static website and share the link — free, with no build step and no server to set up.
Host a file free →