MediaMTX / rtsp-simple-server is a ready-to-use and zero-dependency server and proxy that allows users to publish, read and proxy live video and audio streams.

Live streams can be published to the server with:

protocol	variants	video codecs	audio codecs
WebRTC	Browser-based, WHIP	AV1, VP9, VP8, H264	Opus, G722, G711
RTSP clients	UDP, TCP, RTSPS	AV1, VP9, VP8, H265, H264, MPEG-4 Video (H263, Xvid), MPEG-1/2 Video, M-JPEG and any RTP-compatible codec	Opus, MPEG-4 Audio (AAC), MPEG-1/2 Audio (MP3), G722, G711, LPCM and any RTP-compatible codec
RTSP servers and cameras	UDP, UDP-Multicast, TCP, RTSPS	AV1, VP9, VP8, H265, H264, MPEG-4 Video (H263, Xvid), MPEG-1/2 Video, M-JPEG and any RTP-compatible codec	Opus, MPEG-4 Audio (AAC), MPEG-1/2 Audio (MP3), G722, G711, LPCM and any RTP-compatible codec
RTMP clients (OBS Studio)	RTMP, RTMPS, Enhanced RTMP	AV1, H265, H264	MPEG-4 Audio (AAC), MPEG-1/2 Audio (MP3)
RTMP servers and cameras	RTMP, RTMPS, Enhanced RTMP	H264	MPEG-4 Audio (AAC), MPEG-1/2 Audio (MP3)
HLS servers and cameras	Low-Latency HLS, MP4-based HLS, legacy HLS	H265, H264	Opus, MPEG-4 Audio (AAC)
UDP/MPEG-TS streams	Unicast, broadcast, multicast	H265, H264	Opus, MPEG-4 Audio (AAC)
Raspberry Pi Cameras		H264

And can be read from the server with:

protocol	variants	video codecs	audio codecs
WebRTC	Browser-based, WHEP	AV1, VP9, VP8, H264	Opus, G722, G711
RTSP	UDP, UDP-Multicast, TCP, RTSPS	AV1, VP9, VP8, H265, H264, MPEG-4 Video (H263, Xvid), MPEG-1/2 Video, M-JPEG and any RTP-compatible codec	Opus, MPEG-4 Audio (AAC), MPEG-1/2 Audio (MP3), G722, G711, LPCM and any RTP-compatible codec
RTMP	RTMP, RTMPS, Enhanced RTMP	H264	MPEG-4 Audio (AAC), MPEG-1/2 Audio (MP3)
HLS	Low-Latency HLS, MP4-based HLS, legacy HLS	H265, H264	Opus, MPEG-4 Audio (AAC)

Features:

• Publish live streams to the server
• Read live streams from the server
• Proxy streams from other servers or cameras, always or on-demand
• Streams are automatically converted from a protocol to another. For instance, it's possible to publish a stream with RTSP and read it with HLS
• Serve multiple streams at once in separate paths
• Authenticate users; use internal or external authentication
• Redirect readers to other RTSP servers (load balancing)
• Query and control the server through an HTTP API
• Reload the configuration without disconnecting existing clients (hot reloading)
• Read Prometheus-compatible metrics
• Run external commands when clients connect, disconnect, read or publish streams
• Natively compatible with the Raspberry Pi Camera
• Compatible with Linux, Windows and macOS, does not require any dependency or interpreter, it's a single executable

Note about rtsp-simple-server

rtsp-simple-server has been rebranded as MediaMTX. The reason is pretty obvious: this project started as a RTSP server but has evolved into a much more versatile media server (i like to call it a "media broker", a message broker for media streams), that is not tied to the RTSP protocol anymore. Nothing will change regarding license, features and backward compatibility.

Table of contents

• Installation
  • Standard
  • Docker
  • OpenWRT
• Basic usage
• General
  • Configuration
  • Authentication
  • Encrypt the configuration
  • Proxy mode
  • Remuxing, re-encoding, compression
  • Save streams to disk
  • On-demand publishing
  • Start on boot
    • Linux
    • Windows
  • HTTP API
  • Metrics
  • pprof
  • Compile from source
• Publish to the server
  • From a webcam
  • From a Raspberry Pi Camera
  • From OBS Studio
  • From OpenCV
  • From a UDP stream
  • From the browser
• Read from the server
  • From VLC and Ubuntu
• RTSP protocol
  • General usage
  • TCP transport
  • UDP-multicast transport
  • Encryption
  • Redirect to another server
  • Fallback stream
  • Corrupted frames
  • Decrease latency
• RTMP protocol
  • General usage
  • Encryption
• HLS protocol
  • General usage
  • Browser support
  • Embedding
  • Low-Latency variant
  • HLS on Apple devices
  • Decrease latency
• WebRTC protocol
  • General usage
  • Usage inside a container or behind a NAT
  • Embedding
• Standards
• Links

Installation

Standard

1. Download and extract a precompiled binary from the release page.

2. Start the server:

   ./mediamtx
   

Docker

Download and launch the image:

docker run --rm -it --network=host aler9/rtsp-simple-server

The --network=host flag is mandatory since Docker can change the source port of UDP packets for routing reasons, and this doesn't allow the server to find out the author of the packets. This issue can be avoided by disabling the UDP transport protocol:

docker run --rm -it -e MTX_PROTOCOLS=tcp -p 8554:8554 -p 1935:1935 -p 8888:8888 -p 8889:8889 aler9/rtsp-simple-server

Please keep in mind that the Docker image doesn't include FFmpeg. if you need to use FFmpeg for an external command or anything else, you need to build a Docker image that contains both rtsp-simple-server and FFmpeg, by following instructions here.

OpenWRT

1. In a x86 Linux system, download the OpenWRT SDK corresponding to the wanted OpenWRT version and target from the OpenWRT website and extract it.

2. Open a terminal in the SDK folder and setup the SDK:

   ./scripts/feeds update -a
   ./scripts/feeds install -a
   make defconfig
   

3. Download the server Makefile and set the server version inside the file:

   mkdir package/mediamtx
   wget -O package/mediamtx/Makefile https://raw.githubusercontent.com/bluenviron/mediamtx/main/openwrt.mk
   sed -i "s/v0.0.0/$(git ls-remote --tags --sort=v:refname https://github.com/bluenviron/mediamtx | tail -n1 | sed 's/.*\///; s/\^{}//')/" package/mediamtx/Makefile
   

4. Compile the server:

   make package/mediamtx/compile -j$(nproc)
   

5. Transfer the .ipk file from bin/packages/*/base to the OpenWRT system and install it with:

   opkg install [ipk-file-name].ipk
   

Basic usage

1. Publish a stream. For instance, you can publish a video/audio file with FFmpeg:

   ffmpeg -re -stream_loop -1 -i file.ts -c copy -f rtsp rtsp://localhost:8554/mystream
   

   or GStreamer:

   gst-launch-1.0 rtspclientsink name=s location=rtsp://localhost:8554/mystream filesrc location=file.mp4 ! qtdemux name=d d.video_0 ! queue ! s.sink_0 d.audio_0 ! queue ! s.sink_1
   

   To publish from other hardware / software, take a look at the Publish to the server section.

2. Open the stream. For instance, you can open the stream with VLC:

   vlc --network-caching=50 rtsp://localhost:8554/mystream
   

   or GStreamer:

   gst-play-1.0 rtsp://localhost:8554/mystream
   

   or FFmpeg:

   ffmpeg -i rtsp://localhost:8554/mystream -c copy output.mp4
   

General

Configuration

All the configuration parameters are listed and commented in the configuration file.

There are 3 ways to change the configuration:

1. By editing the mediamtx.yml file, that is

   • included into the release bundle

   • available in the root folder of the Docker image (/mediamtx.yml); it can be overridden in this way:

     docker run --rm -it --network=host -v $PWD/mediamtx.yml:/mediamtx.yml aler9/rtsp-simple-server
     

   The configuration can be changed dynamically when the server is running (hot reloading) by writing to the configuration file. Changes are detected and applied without disconnecting existing clients, whenever it's possible.

2. By overriding configuration parameters with environment variables, in the format MTX_PARAMNAME, where PARAMNAME is the uppercase name of a parameter. For instance, the rtspAddress parameter can be overridden in the following way:

   MTX_RTSPADDRESS="127.0.0.1:8554" ./mediamtx
   

   Parameters that have array as value can be overriden by setting a comma-separated list. For example:

   MTX_PROTOCOLS="tcp,udp"
   

   Parameters in maps can be overridden by using underscores, in the following way:

   MTX_PATHS_TEST_SOURCE=rtsp://myurl ./mediamtx
   

   This method is particularly useful when using Docker; any configuration parameter can be changed by passing environment variables with the -e flag:

   docker run --rm -it --network=host -e MTX_PATHS_TEST_SOURCE=rtsp://myurl aler9/rtsp-simple-server
   

3. By using the HTTP API.

Authentication

Edit mediamtx.yml and replace everything inside section paths with the following content:

paths:
  all:
    publishUser: myuser
    publishPass: mypass

Only publishers that provide both username and password will be able to proceed:

ffmpeg -re -stream_loop -1 -i file.ts -c copy -f rtsp rtsp://myuser:mypass@localhost:8554/mystream

It's possible to setup authentication for readers too:

paths:
  all:
    publishUser: myuser
    publishPass: mypass

    readUser: user
    readPass: userpass

If storing plain credentials in the configuration file is a security problem, username and passwords can be stored as sha256-hashed strings; a string must be hashed with sha256 and encoded with base64:

echo -n "userpass" | openssl dgst -binary -sha256 | openssl base64

Then stored with the sha256: prefix:

paths:
  all:
    readUser: sha256:j1tsRqDEw9xvq/D7/9tMx6Jh/jMhk3UfjwIB2f1zgMo=
    readPass: sha256:BdSWkrdV+ZxFBLUQQY7+7uv9RmiSVA8nrPmjGjJtZQQ=

WARNING: enable encryption or use a VPN to ensure that no one is intercepting the credentials.

Authentication can be delegated to an external HTTP server:

externalAuthenticationURL: http://myauthserver/auth

Each time a user needs to be authenticated, the specified URL will be requested with the POST method and this payload:

{
  "ip": "ip",
  "user": "user",
  "password": "password",
  "path": "path",
  "protocol": "rtsp|rtmp|hls|webrtc",
  "id": "id",
  "action": "read|publish",
  "query": "query"
}

If the URL returns a status code that begins with 20 (i.e. 200), authentication is successful, otherwise it fails.

Please be aware that it's perfectly normal for the authentication server to receive requests with empty users and passwords, i.e.:

{
  "user": "",
  "password": "",
}

This happens because a RTSP client doesn't provide credentials until it is asked to. In order to receive the credentials, the authentication server must reply with status code 401, then the client will send credentials.

Encrypt the configuration

The configuration file can be entirely encrypted for security purposes.

An online encryption tool is available here.

The encryption procedure is the following:

1. NaCL's crypto_secretbox function is applied to the content of the configuration. NaCL is a cryptographic library available for C/C++, Go, C# and many other languages;

2. The string is prefixed with the nonce;

3. The string is encoded with base64.

After performing the encryption, put the base64-encoded result into the configuration file, and launch the server with the MTX_CONFKEY variable:

MTX_CONFKEY=mykey ./mediamtx

Proxy mode

MediaMTX is also a proxy, that is usually deployed in one of these scenarios:

• when there are multiple users that are reading a stream and the bandwidth is limited; the proxy is used to receive the stream once. Users can then connect to the proxy instead of the original source.
• when there's a NAT / firewall between a stream and the users; the proxy is installed on the NAT and makes the stream available to the outside world.

Edit mediamtx.yml and replace everything inside section paths with the following content:

paths:
  proxied:
    # url of the source stream, in the format rtsp://user:pass@host:port/path
    source: rtsp://original-url

After starting the server, users can connect to rtsp://localhost:8554/proxied, instead of connecting to the original url. The server supports any number of source streams, it's enough to add additional entries to the paths section:

paths:
  proxied1:
    source: rtsp://url1

  proxied2:
    source: rtsp://url1

It's possible to save bandwidth by enabling the on-demand mode: the stream will be pulled only when at least a client is connected:

paths:
  proxied:
    source: rtsp://original-url
    sourceOnDemand: yes

Remuxing, re-encoding, compression

To change the format, codec or compression of a stream, use FFmpeg or GStreamer together with MediaMTX. For instance, to re-encode an existing stream, that is available in the /original path, and publish the resulting stream in the /compressed path, edit mediamtx.yml and replace everything inside section paths with the following content:

paths:
  all:
  original:
    runOnReady: ffmpeg -i rtsp://localhost:$RTSP_PORT/$RTSP_PATH -pix_fmt yuv420p -c:v libx264 -preset ultrafast -b:v 600k -max_muxing_queue_size 1024 -f rtsp rtsp://localhost:$RTSP_PORT/compressed
    runOnReadyRestart: yes

Save streams to disk

To save available streams to disk, you can use the runOnReady parameter and FFmpeg:

paths:
  mypath:
    runOnReady: ffmpeg -i rtsp://localhost:$RTSP_PORT/$RTSP_PATH -c copy -f segment -strftime 1 -segment_time 60 -segment_format mpegts saved_%Y-%m-%d_%H-%M-%S.ts
    runOnReadyRestart: yes

In the configuratio above, streams are saved into TS files, that can be read even if the system crashes, while MP4 files can't.

On-demand publishing

Edit mediamtx.yml and replace everything inside section paths with the following content:

paths:
  ondemand:
    runOnDemand: ffmpeg -re -stream_loop -1 -i file.ts -c copy -f rtsp rtsp://localhost:$RTSP_PORT/$RTSP_PATH
    runOnDemandRestart: yes

The command inserted into runOnDemand will start only when a client requests the path ondemand, therefore the file will start streaming only when requested.

Start on boot

Linux

Systemd is the service manager used by Ubuntu, Debian and many other Linux distributions, and allows to launch MediaMTX on boot.

Download a release bundle from the release page, unzip it, and move the executable and configuration in the system:

sudo mv mediamtx /usr/local/bin/
sudo mv mediamtx.yml /usr/local/etc/

Create the service:

sudo tee /etc/systemd/system/mediamtx.service >/dev/null << EOF
[Unit]
Wants=network.target
[Service]
ExecStart=/usr/local/bin/mediamtx /usr/local/etc/mediamtx.yml
[Install]
WantedBy=multi-user.target
EOF

Enable and start the service:

sudo systemctl daemon-reload
sudo systemctl enable mediamtx
sudo systemctl start mediamtx

Windows

Download the WinSW v2 executable and place it into the same folder of mediamtx.exe.

In the same folder, create a file named WinSW-x64.xml with this content:

<service>
  <id>mediamtx</id>
  <name>mediamtx</name>
  <description></description>
  <executable>%BASE%/mediamtx.exe</executable>
</service>

Open a terminal, navigate to the folder and run:

WinSW-x64 install

The server is now installed as a system service and will start at boot time.

HTTP API

The server can be queried and controlled with an HTTP API, that must be enabled by setting the api parameter in the configuration:

api: yes

The API listens on apiAddress, that by default is 127.0.0.1:9997; for instance, to obtain a list of active paths, run:

curl http://127.0.0.1:9997/v2/paths/list

Full documentation of the API is available on the dedicated site.

Metrics

A metrics exporter, compatible with Prometheus, can be enabled with the parameter metrics: yes; then the server can be queried for metrics with Prometheus or with a simple HTTP request:

wget -qO- localhost:9998/metrics

Obtaining:

# metrics of every path
paths{name="[path_name]",state="[state]"} 1
paths_bytes_received{name="[path_name]",state="[state]"} 1234

# metrics of every HLS muxer
hls_muxers{name="[name]"} 1
hls_muxers_bytes_sent{name="[name]"} 187

# metrics of every RTSP connection
rtsp_conns{id="[id]"} 1
rtsp_conns_bytes_received{id="[id]"} 1234
rtsp_conns_bytes_sent{id="[id]"} 187

# metrics of every RTSP session
rtsp_sessions{id="[id]",state="idle"} 1
rtsp_sessions_bytes_received{id="[id]",state="[state]"} 1234
rtsp_sessions_bytes_sent{id="[id]",state="[state]"} 187

# metrics of every RTSPS connection
rtsps_conns{id="[id]"} 1
rtsps_conns_bytes_received{id="[id]"} 1234
rtsps_conns_bytes_sent{id="[id]"} 187

# metrics of every RTSPS session
rtsps_sessions{id="[id]",state="[state]"} 1
rtsps_sessions_bytes_received{id="[id]",state="[state]"} 1234
rtsps_sessions_bytes_sent{id="[id]",state="[state]"} 187

# metrics of every RTMP connection
rtmp_conns{id="[id]",state="[state]"} 1
rtmp_conns_bytes_received{id="[id]",state="[state]"} 1234
rtmp_conns_bytes_sent{id="[id]",state="[state]"} 187

# metrics of every WebRTC session
webrtc_sessions{id="[id]"} 1
webrtc_sessions_bytes_received{id="[id]",state="[state]"} 1234
webrtc_sessions_bytes_sent{id="[id]",state="[state]"} 187

pprof

A performance monitor, compatible with pprof, can be enabled with the parameter pprof: yes; then the server can be queried for metrics with pprof-compatible tools, like:

go tool pprof -text http://localhost:9999/debug/pprof/goroutine
go tool pprof -text http://localhost:9999/debug/pprof/heap
go tool pprof -text http://localhost:9999/debug/pprof/profile?seconds=30

Compile from source

Standard

Install Go ≥ 1.20, download the repository, open a terminal in it and run:

go build .

The command will produce the mediamtx binary.

Raspberry Pi

The server can be compiled with native support for the Raspberry Pi Camera. Compilation must happen on a Raspberry Pi Device, with the following dependencies:

• Go ≥ 1.20
• libcamera-dev
• libfreetype-dev
• xxd
• patchelf

Download the repository, open a terminal in it and run:

cd internal/rpicamera/exe
make
cd ../../../
go build -tags rpicamera .

The command will produce the mediamtx binary.

Compile for all supported platforms

Compilation for all supported platform can be launched by using:

make binaries

The command will produce tarballs in folder binaries/.

Publish to the server

From a webcam

To publish the video stream of a generic webcam to the server, edit mediamtx.yml and replace everything inside section paths with the following content:

paths:
  cam:
    runOnInit: ffmpeg -f v4l2 -i /dev/video0 -pix_fmt yuv420p -preset ultrafast -b:v 600k -f rtsp rtsp://localhost:$RTSP_PORT/$RTSP_PATH
    runOnInitRestart: yes

If the platform is Windows:

paths:
  cam:
    runOnInit: ffmpeg -f dshow -i video="USB2.0 HD UVC WebCam" -pix_fmt yuv420p -c:v libx264 -preset ultrafast -b:v 600k -f rtsp rtsp://localhost:$RTSP_PORT/$RTSP_PATH
    runOnInitRestart: yes

Where USB2.0 HD UVC WebCam is the name of your webcam, that can be obtained with:

ffmpeg -list_devices true -f dshow -i dummy

After starting the server, the webcam can be reached on rtsp://localhost:8554/cam.

From a Raspberry Pi Camera

MediaMTX natively support the Raspberry Pi Camera, enabling high-quality and low-latency video streaming from the camera to any user. There are a couple of requisites:

1. The server must run on a Raspberry Pi, with Raspberry Pi OS bullseye or newer as operative system. Both 32 bit and 64 bit operative systems are supported.

2. Make sure that the legacy camera stack is disabled. Type sudo raspi-config, then go to Interfacing options, enable/disable legacy camera support, choose no. Reboot the system.

If you want to run the standard (non-containerized) version of the server:

1. Make sure that the following packages are installed:

   • libcamera0 (at least version 0.0.2)
   • libfreetype6

2. download the server executable. If you're using 64-bit version of the operative system, make sure to pick the arm64 variant.

3. edit mediamtx.yml and replace everything inside section paths with the following content:

   paths:
     cam:
       source: rpiCamera
   

If you want to run the server with Docker, you need to use the latest-rpi image (that already contains libcamera) and set some additional flags:

docker run --rm -it \
--network=host \
--privileged \
--tmpfs /dev/shm:exec \
-v /run/udev:/run/udev:ro \
-e MTX_PATHS_CAM_SOURCE=rpiCamera \
aler9/rtsp-simple-server:latest-rpi

After starting the server, the camera can be reached on rtsp://raspberry-pi:8554/cam or http://raspberry-pi:8888/cam.

Camera settings can be changed by using the rpiCamera* parameters:

paths:
  cam:
    source: rpiCamera
    rpiCameraWidth: 1920
    rpiCameraHeight: 1080

All available parameters are listed in the sample configuration file.

From OBS Studio

OBS Studio can publish to the server by using the RTMP protocol. In Settings -> Stream (or in the Auto-configuration Wizard), use the following parameters:

• Service: Custom...
• Server: rtmp://localhost
• Stream key: mystream

If credentials are in use, use the following parameters:

• Service: Custom...
• Server: rtmp://localhost
• Stream key: mystream?user=myuser&pass=mypass

If you want to generate a stream that can be read with WebRTC, open Settings -> Output -> Recording and use the following parameters:

• FFmpeg output type: Output to URL
• File path or URL: rtsp://localhost:8554/mystream
• Container format: rtsp
• Check show all codecs (even if potentically incompatible
• Video encoder: h264_nvenc (libx264)
• Video encoder settings (if any): bf=0
• Audio track: 1
• Audio encoder: libopus

The use the button Start Recording (instead of Start Streaming) to start streaming.

From OpenCV

To publish a video stream from OpenCV to the server, OpenCV must be compiled with GStreamer support, by following this procedure:

sudo apt install -y libgstreamer1.0-dev libgstreamer-plugins-base1.0-dev gstreamer1.0-plugins-ugly gstreamer1.0-rtsp python3-dev python3-numpy
git clone --depth=1 -b 4.5.4 https://github.com/opencv/opencv
cd opencv
mkdir build && cd build
cmake -D CMAKE_INSTALL_PREFIX=/usr -D WITH_GSTREAMER=ON ..
make -j$(nproc)
sudo make install

You can check that OpenCV has been installed correctly by running:

python3 -c 'import cv2; print(cv2.getBuildInformation())'

And verifying that the output contains GStreamer: YES.

Videos can be published with VideoWriter:

import cv2
import numpy as np
from time import sleep, time

fps = 15
width = 800
height = 600
colors = [
    (0, 0, 255),
    (255, 0, 0),
    (0, 255, 0),
]

out = cv2.VideoWriter('appsrc ! videoconvert' + \
    ' ! x264enc speed-preset=ultrafast bitrate=600 key-int-max=' + str(fps * 2) + \
    ' ! video/x-h264,profile=baseline' + \
    ' ! rtspclientsink location=rtsp://localhost:8554/mystream',
    cv2.CAP_GSTREAMER, 0, fps, (width, height), True)
if not out.isOpened():
    raise Exception("can't open video writer")

curcolor = 0
start = time()

while True:
    frame = np.zeros((height, width, 3), np.uint8)

    # create a rectangle
    color = colors[curcolor]
    curcolor += 1
    curcolor %= len(colors)
    for y in range(0, int(frame.shape[0] / 2)):
        for x in range(0, int(frame.shape[1] / 2)):
            frame[y][x] = color

    out.write(frame)
    print("frame written to the server")

    now = time()
    diff = (1 / fps) - now - start
    if diff > 0:
        sleep(diff)
    start = now

From a UDP stream

The server supports ingesting UDP/MPEG-TS packets (i.e. MPEG-TS packets sent with UDP). Packets can be unicast, broadcast or multicast. For instance, you can generate a multicast UDP/MPEG-TS stream with:

gst-launch-1.0 -v mpegtsmux name=mux alignment=1 ! udpsink host=238.0.0.1 port=1234 \
videotestsrc ! video/x-raw,width=1280,height=720 ! x264enc speed-preset=ultrafast bitrate=3000 key-int-max=60 ! video/x-h264,profile=high ! mux. \
audiotestsrc ! audioconvert ! avenc_aac ! mux.

Edit mediamtx.yml and replace everything inside section paths with the following content:

paths:
  udp:
    source: udp://238.0.0.1:1234

After starting the server, the stream can be reached on rtsp://localhost:8554/udp.

From the browser

Open the page into the browser:

http://localhost:8889/mystream/publish

Read from the server

From VLC and Ubuntu

The VLC shipped with Ubuntu 21.10 doesn't support playing RTSP due to a license issue (see here and here).

To overcome the issue, remove the default VLC instance and install the snap version:

sudo apt purge -y vlc
snap install vlc

Then use it to read the stream:

vlc rtsp://localhost:8554/mystream

RTSP protocol

General usage

RTSP is a standardized protocol that allows to publish and read streams; in particular, it supports different underlying transport protocols, that are chosen by clients during the handshake with the server:

• UDP: the most performant, but doesn't work when there's a NAT/firewall between server and clients. It doesn't support encryption.
• UDP-multicast: allows to save bandwidth when clients are all in the same LAN, by sending packets once to a fixed multicast IP. It doesn't support encryption.
• TCP: the most versatile, does support encryption.

The default transport protocol is UDP. To change the transport protocol, you have to tune the configuration of your client of choice.

TCP transport

The RTSP protocol supports the TCP transport protocol, that allows to receive packets even when there's a NAT/firewall between server and clients, and supports encryption (see Encryption).

You can use FFmpeg to publish a stream with the TCP transport protocol:

ffmpeg -re -stream_loop -1 -i file.ts -c copy -f rtsp -rtsp_transport tcp rtsp://localhost:8554/mystream

You can use FFmpeg to read that stream with the TCP transport protocol:

ffmpeg -rtsp_transport tcp -i rtsp://localhost:8554/mystream -c copy output.mp4

You can use GStreamer to read that stream with the TCP transport protocol:

gst-launch-1.0 rtspsrc protocols=tcp location=rtsp://localhost:8554/mystream ! fakesink

You can use VLC to read that stream with the TCP transport protocol:

vlc --rtsp-tcp rtsp://localhost:8554/mystream

UDP-multicast transport

The RTSP protocol supports the UDP-multicast transport protocol, that allows a server to send packets once, regardless of the number of connected readers, saving bandwidth.

This mode must be requested by readers when handshaking with the server; once a reader has completed a handshake, the server will start sending multicast packets. Other readers will be instructed to read existing multicast packets. When all multicast readers have disconnected from the server, the latter will stop sending multicast packets.

If you want to use the UDP-multicast protocol in a Wireless LAN, please be aware that the maximum bitrate supported by multicast is the one that corresponds to the lowest enabled WiFi data rate. For instance, if the 1 Mbps data rate is enabled on your router (and it is on most routers), the maximum bitrate will be 1 Mbps. To increase the maximum bitrate, use a cabled LAN or change your router settings.

To request and read a stream with UDP-multicast, you can use FFmpeg:

ffmpeg -rtsp_transport udp_multicast -i rtsp://localhost:8554/mystream -c copy output.mp4

or GStreamer:

gst-launch-1.0 rtspsrc protocols=udp-mcast location=rtsps://ip:8554/...

or VLC (append ?vlcmulticast to the URL):

vlc rtsp://localhost:8554/mystream?vlcmulticast

Encryption

Incoming and outgoing RTSP streams can be encrypted with TLS (obtaining the RTSPS protocol). A TLS certificate is needed and can be generated with OpenSSL:

openssl genrsa -out server.key 2048
openssl req -new -x509 -sha256 -key server.key -out server.crt -days 3650

Edit mediamtx.yml, and set the protocols, encryption, serverKey and serverCert parameters:

protocols: [tcp]
encryption: optional
serverKey: server.key
serverCert: server.crt

Streams can be published and read with the rtsps scheme and the 8322 port:

ffmpeg -i rtsps://ip:8322/...

If the client is GStreamer, disable the certificate validation:

gst-launch-1.0 rtspsrc tls-validation-flags=0 location=rtsps://ip:8322/...

At the moment VLC doesn't support reading encrypted RTSP streams. A workaround consists in launching an instance of MediaMTX on the same machine in which VLC is running, using it for reading the encrypted stream with the proxy mode, and reading the proxied stream with VLC.

Redirect to another server

To redirect to another server, use the redirect source:

paths:
  redirected:
    source: redirect
    sourceRedirect: rtsp://otherurl/otherpath

Fallback stream

If no one is publishing to the server, readers can be redirected to a fallback path or URL that is serving a fallback stream:

paths:
  withfallback:
    fallback: /otherpath

Corrupted frames

In some scenarios, when reading RTSP from the server, decoded frames can be corrupted or incomplete. This can be caused by multiple reasons:

• the packet buffer of the server is too small and can't keep up with the stream throughput. A solution consists in increasing its size:

  readBufferCount: 1024
  

• The stream throughput is too big and the stream can't be sent correctly with the UDP transport. UDP is more performant, faster and more efficient than TCP, but doesn't have a retransmission mechanism, that is needed in case of streams that need a large bandwidth. A solution consists in switching to TCP:

  protocols: [tcp]
  

  In case the source is a camera:

  paths:
    test:
      source: rtsp://..
      sourceProtocol: tcp
  

• The stream throughput is too big to be handled by the network between server and readers. Upgrade the network or decrease the stream bitrate by re-encoding it.

Decrease latency

The RTSP protocol doesn't introduce any latency by itself. Latency is usually introduced by clients, that put frames in a buffer to compensate network fluctuations. In order to decrease latency, the best way consists in tuning the client. For instance, latency can be decreased with VLC by decreasing the Network caching parameter, that is available in the Open network stream dialog or alternatively ca be set with the command line:

vlc --network-caching=50 rtsp://...

RTMP protocol

General usage

RTMP is a protocol that allows to read and publish streams, but is less versatile and less efficient than RTSP (doesn't support UDP, encryption, doesn't support most RTSP codecs, doesn't support feedback mechanism). It is used when there's need of publishing or reading streams from a software that supports only RTMP (for instance, OBS Studio and DJI drones).

At the moment, only the H264 and AAC codecs can be used with the RTMP protocol.

Streams can be published or read with the RTMP protocol, for instance with FFmpeg:

ffmpeg -re -stream_loop -1 -i file.ts -c copy -f flv rtmp://localhost/mystream

or GStreamer:

gst-launch-1.0 -v flvmux name=s ! rtmpsink location=rtmp://localhost/mystream filesrc location=file.mp4 ! qtdemux name=d d.video_0 ! queue ! s.video d.audio_0 ! queue ! s.audio

Credentials can be provided by appending to the URL the user and pass parameters:

ffmpeg -re -stream_loop -1 -i file.ts -c copy -f flv rtmp://localhost:8554/mystream?user=myuser&pass=mypass

Encryption

RTMP connections can be encrypted with TLS, obtaining the RTMPS protocol. A TLS certificate is needed and can be generated with OpenSSL:

openssl genrsa -out server.key 2048
openssl req -new -x509 -sha256 -key server.key -out server.crt -days 3650

Edit mediamtx.yml, and set the rtmpEncryption, rtmpServerKey and rtmpServerCert parameters:

rtmpEncryption: optional
rtmpServerKey: server.key
rtmpServerCert: server.crt

Streams can be published and read with the rtmps scheme and the 1937 port:

rtmps://localhost:1937/...

Please be aware that RTMPS is currently unsupported by VLC, FFmpeg and GStreamer. However, you can use a proxy like stunnel or nginx to allow RTMP clients to access RTMPS resources.

HLS protocol

General usage

HLS is a protocol that allows to embed live streams into web pages. It works by splitting streams into segments, and by serving these segments with the HTTP protocol. Every stream published to the server can be accessed by visiting:

http://localhost:8888/mystream

where mystream is the name of a stream that is being published.

Browser support

Although the server can produce HLS with a variety of video and audio codecs (that are listed at the beginning of the README), not all browsers can read all codecs. You can check what codecs your browser can read by visiting this page:

https://jsfiddle.net/4msrhudv

If you want to increase the compatibility of the stream in order to support most browsers, you have to re-encode it by using the H264 and AAC codecs, for instance by using FFmpeg:

ffmpeg -i rtsp://original-source -pix_fmt yuv420p -c:v libx264 -preset ultrafast -b:v 600k -c:a aac -b:a 160k -f rtsp rtsp://localhost:8554/mystream

Embedding

The simples way to embed a HLS stream into a web page consists in using an iframe tag:

<iframe src="http://mediamtx-ip:8888/mystream" scrolling="no"></iframe>

For more advanced options, you can create and serve a custom web page by starting from the source code of the default page.

Low-Latency variant

Low-Latency HLS is a recently standardized variant of the protocol that allows to greatly reduce playback latency. It works by splitting segments into parts, that are served before the segment is complete.

LL-HLS is enabled by default. Every stream published to the server can be read with LL-HLS by visiting:

https://localhost:8888/mystream

If the stream is not shown correctly, try tuning the hlsPartDuration parameter, for instance:

hlsPartDuration: 500ms

HLS on Apple devices

In order to correctly display Low-Latency HLS streams in Safari running on Apple devices (iOS or macOS), a TLS certificate is needed and can be generated with OpenSSL:

openssl genrsa -out server.key 2048
openssl req -new -x509 -sha256 -key server.key -out server.crt -days 3650

Set the hlsEncryption, hlsServerKey and hlsServerCert parameters in the configuration file:

hlsEncryption: yes
hlsServerKey: server.key
hlsServerCert: server.crt

Keep also in mind that not all H264 video streams can be played on Apple Devices due to some intrinsic properties (distance between I-Frames, profile). If the video can't be played correctly, you can either:

• re-encode it by following the guide

• disable the Low-latency variant of HLS and go back to the legacy variant:

  hlsVariant: mpegts
  

Decrease latency

in HLS, latency is introduced since a client must wait for the server to generate segments before downloading them. This latency amounts to 500ms-3s when the low-latency HLS variant is enabled (and it is by default), otherwise amounts to 1-15secs.

To decrease the latency, you can:

• try decreasing the hlsPartDuration parameter;

• try decreasing the hlsSegmentDuration parameter;

• The segment duration is influenced by the interval between the IDR frames of the video track. An IDR frame is a frame that can be decoded independently from the others. The server changes the segment duration in order to include at least one IDR frame into each segment. Therefore, you need to decrease the interval between the IDR frames. This can be done in two ways:

  • if the stream is being hardware-generated (i.e. by a camera), there's usually a setting called Key-Frame Interval in the camera configuration page

  • otherwise, the stream must be re-encoded. It's possible to tune the IDR frame interval by using ffmpeg's -g option:

    ffmpeg -i rtsp://original-stream -pix_fmt yuv420p -c:v libx264 -preset ultrafast -b:v 600k -max_muxing_queue_size 1024 -g 30 -f rtsp rtsp://localhost:$RTSP_PORT/compressed
    

WebRTC protocol

General usage

Every stream published to the server can be read with WebRTC by visiting:

http://localhost:8889/mystream

Usage inside a container or behind a NAT

If the server is hosted inside a container or is behind a NAT, additional configuration is required in order to allow the two WebRTC parts (the browser and the server) to establish a connection (WebRTC/ICE connection).

A first method consists into forcing all WebRTC/ICE connections to pass through a single UDP server port, by using the parameters:

# public IP of the server
webrtcICEHostNAT1To1IPs: [192.168.x.x]
# any port of choice
webrtcICEUDPMuxAddress: :8189

The NAT / container must then be configured in order to route all incoming UDP packets on port 8189 to the server. If you're using Docker, this can be achieved with the flag:

docker run --rm -it \
-p 8189:8189/udp
....
aler9/rtsp-simple-server

If the UDP protocol is blocked by a firewall, all WebRTC/ICE connections can be forced to pass through a single TCP server port:

# public IP of the server
webrtcICEHostNAT1To1IPs: [192.168.x.x]
# any port of choice
webrtcICETCPPMuxAddress: :8189

The NAT / container must then be configured in order to redirect all incoming TCP packets on port 8189 to the server. If you're using Docker, this can be achieved with the flag:

docker run --rm -it \
-p 8189:8189
....
aler9/rtsp-simple-server

Finally, if none of these methods work, you can force all WebRTC/ICE connections to pass through a TURN server, like coturn, that must be configured externally. The server address and credentials must be set in the configuration file:

webrtcICEServers: [turn:user:pass:host:port]

Where user and pass are the username and password of the server. Note that port is not optional.

If the server uses a secret-based authentication (for instance, coturn with the use-auth-secret option), it must be configured in this way:

webrtcICEServers: [turn:AUTH_SECRET:secret:host:port]

where secret is the secret of the TURN server. MediaMTX will generate a set of credentials by using the secret, and credentials will be sent to clients before the WebRTC/ICE connection is established.

Embedding

The simples way to embed a WebRTC stream into a web page consists in using an iframe tag:

<iframe src="http://mediamtx-ip:8889/mystream" scrolling="no"></iframe>

For more advanced options, you can create and serve a custom web page by starting from the source code of the default read page and source code of the publish page.

Standards

• RTSP

  • RTSP/RTP/RTCP standards

• HLS

  • HLS standards

• RTMP

  • RTMP
  • Enhanced RTMP

• WebRTC

  • WebRTC: Real-Time Communication in Browsers
  • WebRTC HTTP Ingestion Protocol (WHIP)
  • WebRTC HTTP Egress Protocol (WHEP)

• Video and audio codecs

  • Codec standards

• Other

  • Golang project layout

Links

Related projects

• gortsplib (RTSP library used internally)
• gohlslib (HLS library used internally)
• pion/webrtc (WebRTC library used internally)
• pion/sdp (SDP library used internally)
• pion/rtp (RTP library used internally)
• pion/rtcp (RTCP library used internally)
• notedit/rtmp (RTMP library used internally)
• go-astits (MPEG-TS library used internally)
• go-mp4 (MP4 library used internally)