RTSP explained (and why you can’t put RTSP on a web page)

If you own an IP camera, you have probably seen an address in its settings that starts with rtsp://. Plug it into VLC or a video recorder and the picture appears. Paste it into a web page and nothing happens. That gap between “works in VLC” and “works on my website” is the single most common streaming question we get, and the answer explains a lot about how camera video works.

What RTSP is

RTSP is a control protocol, standardized by the IETF: version 1.0 in RFC 2326 (1998) and version 2.0 in RFC 7826 (2016). RFC 7826 describes it as an application-level protocol for the setup and control of the delivery of data with real-time properties. Think of it as the remote control for a camera’s stream:

• DESCRIBE: “What streams do you have?” The camera answers with a description of its video and audio (codec, resolution).
• SETUP: “Send the video this way.” The client and camera agree how the media will travel.
• PLAY: “Start sending.” Video begins to flow.
• TEARDOWN: “Stop.” The session ends.

The video itself usually travels by RTP (Real-time Transport Protocol), either over UDP for low delay or interleaved inside the same TCP connection as the RTSP commands. The default port for RTSP is 554; RFC 7826 also defines rtsps over TLS on port 322, though many consumer cameras do not support it.

The parts of an RTSP address

An RTSP address follows the same pattern as a web address, with a few camera-specific parts. In placeholder form it looks like rtsp://[camera-address]:[port]/[stream-path]:

• Scheme: rtsp (or rtsps for the encrypted version).
• Camera address: on a home or business network, a private address that only works inside that network.
• Port: 554 unless changed.
• Stream path: chosen by the maker, often different for the main stream and the sub stream.

Some software lets you put a username and password inside the address. That is convenient on your own computer, but it means anyone who sees the address sees the password, which is one more reason never to paste an RTSP address anywhere public.

RTSP 1.0 vs 2.0

Almost all cameras and software in use today speak RTSP 1.0 (RFC 2326). RTSP 2.0 (RFC 7826, published in December 2016) cleaned up the protocol, changed how some methods and headers work and is not backward compatible beyond basic version negotiation. In practice, when a camera spec sheet says “RTSP,” assume 1.0. The differences rarely matter to camera owners; what matters is that the camera offers RTSP at all, and lets you choose the stream and authentication settings.

RTSP and recorders

Network video recorders (NVRs) and video management software use RTSP to record cameras continuously. A recorder pulls each camera’s main stream, stores it on its hard drives and serves playback to your phone through the maker’s app. Because the recorder is the only thing pulling from the camera, it also protects the camera from being overloaded. If you want both recording and a public stream, the usual setup is: recorder pulls the main stream for recording, and an encoder or the camera itself pushes a separate stream out to YouTube. Our how live streaming works page shows where each piece sits.

Why RTSP is great inside your network

• Low delay: video flows straight from the camera, often well under a second or two behind real time.
• Universal among cameras: nearly every IP camera offers it, and the ONVIF interface standards use RTSP for streaming (Profiles S and T).
• Efficient: the camera sends the stream once to each client, with no re-encoding.
• Two streams: cameras usually offer a high-quality main stream and a lighter sub stream on different RTSP paths.

That is why network video recorders, security software, OBS and media servers all speak RTSP.

Why you can’t put RTSP on a web page

There are four separate reasons, and any one of them is enough:

1. Browsers do not include an RTSP client. Modern browsers play video delivered over HTTP(S), such as HLS and DASH segments through Media Source Extensions, and real-time video through WebRTC. RTSP is not on the list. Older plugin-based workarounds (Flash, ActiveX, browser-embedded VLC) are gone.
2. Every viewer would connect to your camera. A camera can serve only a handful of RTSP clients before it runs out of processing power or upload bandwidth. A popular webcam may have thousands of viewers at once.
3. Viewers would need to reach your camera from the internet. That means port forwarding port 554, which exposes the camera to the whole internet. Automated scanners look for exactly that. See port forwarding vs P2P.
4. Credentials would be public. RTSP URLs often carry a username and password, or rely on weak authentication. Putting one in a web page hands those to every visitor.

How to get a camera’s RTSP stream onto the web

The fix is always the same idea: something that speaks RTSP pulls the stream inside your network and pushes or serves it in a web format.

For a public webcam, pushing to YouTube is usually the best mix of cost, reach and safety. Our step-by-step guide is how to stream an IP camera to YouTube 24/7, and how to embed a live camera on your website covers the web page side.

RTSP compared with other protocols

A real-world example

Picture a small hotel with a rooftop camera facing the sea. The owner can watch it in VLC on the front-desk computer using its RTSP address, and the hotel’s recorder keeps a week of footage. Now they want the view on their website’s home page. They have three good choices: switch on the camera’s own RTMP push to YouTube and embed the YouTube player; run OBS on the front-desk PC to pull RTSP and push to YouTube; or pay a camera-streaming service that gives them an embeddable player without YouTube branding. What they should not do is forward port 554 on the router and paste the RTSP address into the website. It would not play anyway, and it would expose the camera to the internet.

Once the YouTube stream is running, they can also submit it to LiveLocation so travelers find it on our map.

Finding your camera’s RTSP address (for your own camera)

On a camera you own, the RTSP path is in the maker’s manual or support site, and often in the camera’s network settings. Some makers require you to switch RTSP on first, and some battery and cloud cameras do not offer it at all. Tools that use ONVIF can also discover a camera’s stream addresses automatically on your own network. Use a dedicated low-privilege user account for RTSP rather than the admin account, and keep the address private.

RTSP problems and fixes

• Stream works in VLC on Wi-Fi but stutters: RTP over UDP can lose packets on busy Wi-Fi. Try “RTSP over TCP” in your software.
• Video plays but no sound: the camera may send audio in a codec your software does not support, or audio may be off (often a good idea for privacy).
• Only a few clients can connect: normal. Use a recorder or media server to pull once and redistribute.
• Stream freezes after hours: often a camera firmware issue; update firmware and set the encoder to reconnect automatically.

Do these next

1. Learn how RTMP pushes your stream out safely.
2. Send your camera to YouTube step by step.
3. Lock the camera down so RTSP stays private.

Last reviewed October 7, 2026 by the LiveLocation team. General information, not legal or electrical advice.