A downloader that hides its ceilings wastes your time twice: once waiting, once guessing why it failed. Every number below was read from the server that answers your request, and each one comes with the reason it exists.
The short version
- Too long or too big. Over 30 minutes, or over 500 MB. Both are hard ceilings with no override.
- Too many, too fast. Ten link lookups and six downloads per address per ten minutes. Two separate counters.
- Too many at once. One download runs at a time, ten may wait. The eleventh waiter is refused.
- Not this platform. Anything that is not a Dailymotion video page or a dai.ly short link, including YouTube.
- Not reachable. Private, deleted, region-blocked, or an offline channel. No login exists to get past it.
The two request limits are separate and both run on a ten-minute window. Ten link lookups and six downloads per address. Reloading a page that seems stuck spends lookups, not downloads — which is why a jittery connection can leave you able to see the versions and unable to start one.
The twelve limits, with the numbers
1. 30 minutes is the longest video accepted
A video longer than 1800 seconds is refused as over the limit. The check runs against the duration the source reports, before any media is transferred, so a long file is turned away in a moment instead of filling the disk and then being discarded. There is no way to raise this from the page: it is a fixed ceiling, not a preference.
2. 500 MB is the largest file accepted
A single downloaded file above 500 MB is refused. The ceiling is written into three places, and only the last of them decides in practice. The parse stage refuses a candidate whose size the source has already declared — except that in every sample measured here the source declared no size at all, so that branch has never run. The extraction tool is also handed the ceiling. What actually enforces it is the real byte count after the file is written: over 500 MB and the file is deleted, the task ends as too large. That is why this failure can arrive at the end of a download rather than at the start — a declared size is a claim, not a fact, and here there is usually no claim to read.
3. One download runs at a time
Media processing is serial. A second request does not fail, it waits. That is a deliberate trade: the server has one disk and a limited amount of memory, and running two extractions together makes both slower and more likely to die halfway. In practice a queue means your download starts a little later rather than failing outright.
4. The waiting queue holds ten
Beyond ten waiting tasks the queue is full and a new request is refused rather than stacked indefinitely. A queue with no ceiling is not a queue, it is a slow-motion outage: everyone waits, nobody is told, and the site looks broken instead of busy.
5. A download gives up after 10 minutes
Each task has a hard ten-minute wall clock. Extraction, remuxing and the disk write all have to finish inside it. A stalled source or a very large file ends as a timeout rather than hanging forever with a progress bar that never moves.
6. Reading a link gives up after 120 seconds
Resolving a page into a list of available versions has its own two-minute ceiling, separate from the download ceiling. If the source does not answer within it, the attempt is reported as a timeout. This is the step most likely to fail on a badly behaved source, because it depends entirely on how quickly that source replies.
7. 10 link lookups per address every 10 minutes
Reading a link — the step that lists available versions — is limited to ten attempts per client address in any ten-minute window. Past that the answer is HTTP 429 with a retry delay, and the page shows how long is left. Refresh-hammering a page that seems stuck is the usual way to hit it.
8. 6 downloads per address every 10 minutes
Starting a download is limited separately and more tightly: six per client address per ten minutes. The two counters are independent, so you can be allowed to look but not to download, or the reverse. This is the limit most people meet on a real binge, not the size or duration ceilings.
9. Only two Dailymotion address forms are accepted
A video page under www.dailymotion.com, or a dai.ly short link. Every other address is refused before anything is fetched. That includes YouTube — not because YouTube could not technically be read, but because this tool is for Dailymotion and a wider gate would be a promise this page cannot keep. It also excludes lookalike domains, addresses with credentials in them, non-standard ports, and anything that is not http or https.
10. Private, deleted and offline videos are not reachable
A video that is private, removed, region-blocked, or on a channel that is currently offline cannot be read, and the answer says so rather than pretending the link was malformed. A public link is the only kind that works. There is no login, so there is no way to reach anything behind one.
11. Separate audio-only streams are not merged for you
Some quality levels are published by the source as two files: one picture without sound, one sound without picture. Where a version has no audio track of its own, it is offered as it is rather than silently stitched together with a second stream. The result plays, but it may be silent — and telling you that up front is better than letting you discover it after a two-minute download.
12. Downloads stop when the disk gets low
New tasks are refused once free disk space falls below 5 GB. This is a safety valve, not a policy: a full disk does not degrade gracefully, it takes the whole site down for everyone. Temporary files are cleaned up an hour after a task finishes, which is what keeps the reserve from draining in the first place.
What each failure actually means
Seven outcomes look similar from the outside and only two of them are worth retrying. Telling them apart saves a lot of pointless reloading.
1Over the limitThe duration or the size is beyond the ceiling. A different video is the only fix.
2Too largeThe finished file was over the 500 MB ceiling when its bytes were counted. Try a lower quality version of the same video.
3Queue fullTen tasks are already waiting. Nothing is wrong with your link; the server is busy. Waiting a minute and retrying usually works.
4Disk fullFree space fell under the reserve. This is temporary and affects everyone, not just you.
5Not found or privateThe source itself says the video is unavailable. Retrying will not change it.
6Rate limitedYou have used this window's allowance. The answer includes how many seconds to wait.
7Expired / not readyAn earlier step timed out or was cleaned up. Start again from the link.
Worth knowing if you are scripting against this: a refusal and a failure are not the same status. A malformed or foreign link and an expired job come back as client errors, while the queue, the disk reserve and the rate limit come back as HTTP 429 with a retry delay. Treating them as one class of error is how a retry loop turns a temporary limit into a permanent block.
What is not stored
No account, so there is nothing to log into and no history to keep. A download exists as a temporary file on the server for up to an hour after the task finishes, then it is removed. Nothing is kept as a library, nothing is served back to anyone else, and there is no list of what has been downloaded.
What this tool is not
Not affiliated with Dailymotion
VidMotion is independent. It is not affiliated with, sponsored by or endorsed by Dailymotion, and a refusal or a failure says nothing about what Dailymotion itself would allow.
Not a licence to reuse
A video being publicly visible is not permission to save or republish it. Copyright is unchanged by whether a link resolves. Only download what you own or have permission to keep.
Not a quality improver
You get one of the versions the source actually publishes. A 480p source does not become 1080p, and nothing is upscaled, sharpened or re-encoded to look better than the original.