OpenVidu configuration

This page lists all available configuration properties for an OpenVidu deployment, as well as their possible values and the default ones.
These properties may be set:

  • In any official production deployment of OpenVidu CE, Pro or Enterprise: in the .env config file at OpenVidu installation path, default to /opt/openvidu
  • In the official development OpenVidu docker container: passing them as environment variables with flag -e PROPERTY=value

Configuration parameters for OpenVidu CE 🔗

These configuration parameters apply to OpenVidu CE, OpenVidu Pro and OpenVidu Enterprise.

Parameter Description Default value
DOMAIN_OR_PUBLIC_IP Domain name where OpenVidu Server will be reachable. If you do not have one, the public IP address of the machine. Clients will use this to connect to OpenVidu Server. For example:
  • openvidu.example.com
  • 192.0.2.2
OPENVIDU_SECRET Secret used to connect to OpenVidu Server. This value is required when using the REST API or any server client (openvidu-java-client, openvidu-node-client), as well as when connecting to openvidu-server dashboard. Can only contain alphanumeric characters, hyphens and underscores (matching regex [a-zA-Z0-9_-]+).
CERTIFICATE_TYPE Which type of certificate you want to use in your OpenVidu deployment:
  • selfsigned
  • owncert
  • letsencrypt
 selfsigned
HTTPS_PORT Secure port where OpenVidu Server will listen for connections. All of OpenVidu services will be reachable at this port. Any client or SDK will have to connect to OpenVidu at this port. 443
HTTP_PORT Insecure port from which OpenVidu will automatically redirect any request to the secure port.
For example, with
HTTP_PORT=80 and HTTPS_PORT=443
a request to
http://DOMAIN_OR_PUBLIC_IP/
will be automatically redirected to
https://DOMAIN_OR_PUBLIC_IP/
WARNING: if you have set CERTIFICATE_TYPE=letsencrypt, you cannot change the default value 80, as Let's Encrypt needs this port opened to generate your certificate. This applies at least for the first time you execute OpenVidu, when your certificate is generated. You should be able to change this value later, once Let's Encrypt has successfully run for the first time.		 80
OPENVIDU_CDR Whether to enable Call Detail Record or not. false
OPENVIDU_CDR_PATH System path where to write CDR files /opt/openvidu/cdr
OPENVIDU_RECORDING Whether to enable the Recording module or not. false
OPENVIDU_RECORDING_PATH System path where to store the video files of recorded sessions. /opt/openvidu/recordings
OPENVIDU_RECORDING_PUBLIC_ACCESS Whether to allow free HTTP access to recorded sessions or not.
  • If true, OPENVIDU_RECORDING_PATH will be publicly reachable at /recordings/, so any client can connect to
    https://DOMAIN_OR_PUBLIC_IP/openvidu/recordings/RECORDING_ID/RECORDING_NAME.EXTENSION
    and reach the recorded files.
  • If false, this path will be secured with Basic Auth the same way that REST API is
    ("OPENVIDUAPP:OPENVIDU_SECRET")
For example, for an OpenVidu Server configured with
  • DOMAIN_OR_PUBLIC_IP=example.com
  • OPENVIDU_RECORDING=true
  • OPENVIDU_RECORDING_PATH=/my/path
  • OPENVIDU_RECORDING_PUBLIC_ACCESS=true
a session with ID foo that has been recorded may generate a video file locally stored in the host machine under /my/path/foo/foo.mp4 and reachable by any client connecting to
https://example.com/openvidu/recordings/foo/foo.mp4		 false
OPENVIDU_RECORDING_NOTIFICATION Which users should receive the recording events (recordingStarted, recordingStopped) in the client side:
  • all: Every user connected to the session.
  • publisher_moderator: Users with role PUBLISHER or MODERATOR.
  • moderator: Only users with role MODERATOR.
  • none: No user will receive these events.
 publisher_moderator
OPENVIDU_RECORDING_CUSTOM_LAYOUT System path where OpenVidu Server should look for custom recording layouts. /opt/openvidu/custom-layout
OPENVIDU_RECORDING_AUTOSTOP_TIMEOUT Timeout in seconds for recordings to automatically stop (and the session involved to be closed) when conditions are met. See Automatic stop of recordings to learn more. 120
OPENVIDU_RECORDING_COMPOSED_URL Overrides the default URL that the recording module will use to connect to the session being recorded. This only affects COMPOSED recordings with video.
By default, the URL is built with DOMAIN_OR_PUBLIC_IP and HTTPS_PORT, so in certain cases where the host does not allow hairpinning (connecting to itself using its own public domain), this can be a solution.
For example, for OpenVidu CE using https://PRIVATE_IP/dashboard or OpenVidu Pro using https://PRIVATE_IP/inspector (where PRIVATE_IP is the internal IP of the Master Node), this setting may allow COMPOSED recording to work fine with default or custom layouts hosted in the same machine without accessing the Internet.
OPENVIDU_RECORDING_DEBUG Enable debug mode for recording. Affects COMPOSED and COMPOSED_QUICK_START recording output modes. Generates log files with the recording process details inside this particular recording folder (a folder with the recording identifier inside OPENVIDU_RECORDING_PATH). false
OPENVIDU_WEBHOOK Whether to enable the OpenVidu Webhook service or not. false
OPENVIDU_WEBHOOK_ENDPOINT HTTP endpoint where OpenVidu Server will send the POST messages with Webhook events.
OPENVIDU_WEBHOOK_HEADERS Array of strings with the HTTP headers that OpenVidu Server will append to each POST message of Webhook events.
For example, you may configure a Basic Auth header, setting this property to [\"Authorization:\ Basic\ bmFtZTpwYXNz\"]		 [ ]
OPENVIDU_WEBHOOK_EVENTS Array of strings with the type of events you want OpenVidu Server to send to your Webhook.
["sessionCreated",
"sessionDestroyed",
"participantJoined",
"participantLeft",
"webrtcConnectionCreated",
"webrtcConnectionDestroyed",
"recordingStatusChanged",
"filterEventDispatched",
"signalSent",
"mediaNodeStatusChanged",
"autoscaling",
"nodeCrashed",
"nodeRecovered"]
(All available events)
OPENVIDU_STREAMS_VIDEO_MAX_RECV_BANDWIDTH Maximum video bandwidth sent from clients to OpenVidu Server, in kbps. 0 means unconstrained. 1000
OPENVIDU_STREAMS_VIDEO_MIN_RECV_BANDWIDTH Minimum video bandwidth sent from clients to OpenVidu Server, in kbps. 0 means unconstrained. 300
OPENVIDU_STREAMS_VIDEO_MAX_SEND_BANDWIDTH Maximum video bandwidth sent from OpenVidu Server to clients, in kbps. 0 means unconstrained. 1000
OPENVIDU_STREAMS_VIDEO_MIN_SEND_BANDWIDTH Minimum video bandwidth sent from OpenVidu Server to clients, in kbps. 0 means unconstrained. 300
OPENVIDU_STREAMS_FORCED_VIDEO_CODEC Enforce a specific video codec to be used by all clients, to avoid compatibility issues:
  • MEDIA_SERVER_PREFERRED
    A recommended choice is done for you
  • NONE
    Let each client use their preferred codec
  • VP8
  • VP9
  • H264
More details.		 MEDIA_SERVER_PREFERRED
OPENVIDU_STREAMS_ALLOW_TRANSCODING Allow the media server to perform live transcoding of video streams, ensuring that all codecs match in the session (Kurento only).
More details.		 false
OPENVIDU_SESSIONS_GARBAGE_INTERVAL How often the garbage collector of non active sessions runs. This helps cleaning up Sessions that have been initialized through REST API, and maybe have had Connections created, but have had no final users connected (no Connection of the Session entered active status). Default to 900s (15 mins). 0 to disable the non active sessions garbage collector. 900
OPENVIDU_SESSIONS_GARBAGE_THRESHOLD Minimum time in seconds that a non active session must have been in existence for the garbage collector of non active sessions to remove it. Default to 3600s (1 hour). If non active sessions garbage collector is disabled (property OPENVIDU_SESSIONS_GARBAGE_INTERVAL is set to 0) this property is ignored. 3600
OPENVIDU_WEBRTC_ICE_SERVERS Array of ICE servers to use instead of default Coturn at browser/clients side (comma-separated list of ICE Servers). Examples:
  • ["url=turns:example.turn.com:443,staticAuthSecret=secret"]
  • ["url=turns:example.turn.com:443,username=usertest,credential=userpass"]
More Details		 [ ]


Advanced parameters for OpenVidu CE 🔗

These properties are related to infrastructure and network aspects. OpenVidu is deployed with sane defaults, and these configuration properties will normally not require changes. They are only described here for custom on premises deployments with special network conditions or requirements, and only meant for users with deep knowledge about their deployment environment.
Parameter Description Default value
COTURN_IP This is the IP of the default TURN/STUN Coturn server deployed with OpenVidu.
  • By default it is the autodiscovered Public IPv4. This IP is used by browsers and Kurento Media Server as TURN/STUN service for ICE protocol to autodiscover its own Public IP.
  • Possible values:
    • auto-ipv4
    • auto-ipv6
    • Any valid Public Ipv4 or Public IPv6
 auto-ipv4
COTURN_PORT This is the port used for TURN/STUN Coturn server deployed with OpenVidu in the Master Node.
  • By default it is 3478. This port is used by browsers and Kurento Media Server as TURN/STUN service port for ICE protocol to autodiscover its own Public IP.
  • Possible values: Any value between 1-65535
 3478
OPENVIDU_WEBRTC_ICE_SERVERS Array of ICE servers to use instead of the default Coturn at the client side (comma-separated list of ICE Servers). When OPENVIDU_WEBRTC_ICE_SERVERS is used, Kurento
still uses COTURN_IP (or the public IP of Media Nodes if OPENVIDU_PRO_COTURN_IN_MEDIA_NODES=true) but browsers uses the TURN/STUN servers defined at OPENVIDU_WEBRTC_ICE_SERVERS. Examples:
  • ["url=turns:example.turn.com:443,staticAuthSecret=secret"]
  • ["url=turns:example.turn.com:443,username=usertest,credential=userpass"]
More Details		 [ ]


Configuration parameters for OpenVidu Pro 🔗

These configuration parameters apply to OpenVidu Pro and OpenVidu Enterprise.

Parameter Description Default value
OPENVIDU_PRO_LICENSE License key of your OpenVidu Pro account. This parameter is mandatory to launch OpenVidu Pro clusters.
Sign up now to get your 15 day free trial!
OPENVIDU_EDITION OpenVidu Edition to run. Possible values are: pro
OPENVIDU_PRO_CLUSTER_ID Unique identifier of your cluster. Each OpenVidu Server Pro instance corresponds to one cluster. You can launch as many clusters as you want with your license key. The cluster ID will always be stored to disk so restarting OpenVidu Server Pro will keep the same previous cluster ID if this configuration parameter is not given a distinct value. A random string
OPENVIDU_PRO_CLUSTER_MEDIA_NODES The desired number of Media Nodes on startup. If there are too many existing Media Nodes on startup, they will be automatically dropped until this number is reached. If there are not enough, more will be automatically launched. This property has effect only on AWS deployments and if OPENVIDU_PRO_CLUSTER_AUTOSCALING=false. 1
OPENVIDU_PRO_CLUSTER_RECONNECTION_TIMEOUT Maximum time in seconds that OpenVidu Server will try to reconnect to a crashed Media Node before terminating it. From the perspective of WebHook events, it is the time frame between a Media Node disconnection and a mediaNodeStatusChanged event to terminating. Check out Media Node reconnection configuration for further information. Minimum 3 seconds, maximum infinite time (expressed with -1). Depending on the deployment environment:
-1 (infinite retry) for On Premises deployments with autoscaling disabled.
3 (min value) for any other type of deployment.
OPENVIDU_PRO_CLUSTER_AUTOSCALING Whether to enable autoscaling or not. With autoscaling enabled, the number of Media Nodes will be automatically adjusted according to existing load. Enabling this property makes OPENVIDU_PRO_CLUSTER_MEDIA_NODES useless (instead will be using the value given to property OPENVIDU_PRO_CLUSTER_AUTOSCALING_MIN_NODES). false
OPENVIDU_PRO_CLUSTER_AUTOSCALING_MAX_NODES If OPENVIDU_PRO_CLUSTER_AUTOSCALING=true, the upper limit of Media Nodes that can be reached. Even when the average load exceeds the threshold, no more Media Nodes will be added to cluster. 2
OPENVIDU_PRO_CLUSTER_AUTOSCALING_MIN_NODES If OPENVIDU_PRO_CLUSTER_AUTOSCALING=true, the lower limit of Media Nodes that can be reached. Even when the average load is inferior to the threshold, no more Media Nodes will be removed from the cluster. This property acts as OPENVIDU_PRO_CLUSTER_MEDIA_NODES when autoscaling is enabled. 1
OPENVIDU_PRO_CLUSTER_AUTOSCALING_MAX_LOAD If OPENVIDU_PRO_CLUSTER_AUTOSCALING=true, the upper average load threshold that will trigger the addition of a new Media Node. Percentage value (0 min, 100 max). 70
OPENVIDU_PRO_CLUSTER_AUTOSCALING_MIN_LOAD If OPENVIDU_PRO_CLUSTER_AUTOSCALING=true, the lower average load threshold that will trigger the removal of an existing Media Node. Percentage value (0 min, 100 max). 20
OPENVIDU_PRO_CLUSTER_PATH Path where OpenVidu Server Pro will manage cluster-related files. This includes: looking for custom infrastructure management scripts and storing temporal output files with instances information. /opt/openvidu/cluster
OPENVIDU_PRO_CLUSTER_TEST If true, OpenVidu Server Pro will perform a quick test on startup to check the clustering features:
  1. Launch a Media Node
  2. Connect to it
  3. Drop it
OpenVidu Server Pro logs will inform about the test execution in detail.
Immediately after finishing the test, OpenVidu Server Pro process will terminate with an exit code of 0 if the test was successful, or 1 if it failed.
This property is extremely useful to test that your OpenVidu Pro deployment is working fine.		 false
OPENVIDU_PRO_ELASTICSEARCH This property determines whether OpenVidu Pro or Enterprise will use Elasticsearch to store history of sessions, logs and metrics. If it is set to true, OpenVidu will try to connect to an Elasticsearch instance, either one deployed by default in OpenVidu Pro or a custom one specified in OPENVIDU_PRO_ELASTICSEARCH_HOST in OpenVidu Enterprise. If it is set to false, no history of sessions, logs or metrics will be sent to Elasticsearch. true
OPENVIDU_PRO_ELASTICSEARCH_HOST This property specifies the URL of a custom Elasticsearch instance to be used by OpenVidu Pro or Enterprise. If OPENVIDU_PRO_ELASTICSEARCH is set to true, OpenVidu will try to connect to this URL to send history of sessions, logs and metrics. In OpenVidu Pro, if this property is not defined, OpenVidu will deploy a default Elasticsearch instance in the master node to be used by itself. In OpenVidu Enterprise, if this property is not defined while OPENVIDU_PRO_ELASTICSEARCH is set to true, OpenVidu won't be able to use Elasticsearch. http://localhost:9200
OPENVIDU_PRO_KIBANA_HOST This property specifies the URL of a Kibana instance to be used by OpenVidu Pro or Enterprise to visualize data stored in Elasticsearch. If OPENVIDU_PRO_ELASTICSEARCH is set to true, OpenVidu will use this URL to connect to Kibana. In OpenVidu Pro, if this property is not defined, OpenVidu will deploy a default Kibana instance in the master node to be used by itself. In OpenVidu Enterprise, if this property is not defined while OPENVIDU_PRO_ELASTICSEARCH is set to true, OpenVidu won't be able to use Kibana to visualize data. http://localhost:5601
OPENVIDU_PRO_ELASTICSEARCH_MAX_DAYS_DELETE Maximum number of days that Elasticsearch indexes in rollover state will be kept on disk. 7
OPENVIDU_PRO_STATS_SESSION_INTERVAL Interval in seconds for OpenVidu to start gathering statistics from Media Nodes.
This includes number of sessions in each Media Node, number of WebRTC connections, recordings, etc...
Set it to 0 for no gathering at all.		 5
OPENVIDU_PRO_STATS_SERVER_INTERVAL Interval in seconds for server-related metrics gathering: HTTP requests, statuses, log level information...
Set it to 0 for no gathering at all.		 10
OPENVIDU_PRO_STATS_MONITORING_INTERVAL Interval in seconds for CPU, memory and net usage stats gathering in Master Node. Set it to 0 for no gathering at all. 10
OPENVIDU_PRO_STATS_WEBRTC_INTERVAL Interval in seconds for WebRTC stats gathering from Media Nodes WebRTC endpoints. Set it to 0 for no gathering at all. 30
OPENVIDU_PRO_NETWORK_QUALITY Whether to enable or disable the Network Quality API.
You can monitor and warn users about the quality of their networks with this feature.		 false
OPENVIDU_PRO_NETWORK_QUALITY_INTERVAL If OPENVIDU_PRO_NETWORK_QUALITY=true, how often the network quality algorithm will be invoked for each user, in seconds. 5
OPENVIDU_PRO_SPEECH_TO_TEXT Engine to use in the Speech To Text service.
No value means the Speech To Text service is disabled (this is the deafult configuration). It can be:
  • azure: Use Azure Speech To Text engine
  • aws: Use the Amazon Transcribe engine
  • vosk: Use the open-śource Vosk engine
OPENVIDU_PRO_SPEECH_TO_TEXT_AZURE_KEY If OPENVIDU_PRO_SPEECH_TO_TEXT=azure, Azure key for the Speech To Text service. See Azure Speech To Text docs
OPENVIDU_PRO_SPEECH_TO_TEXT_AZURE_REGION If OPENVIDU_PRO_SPEECH_TO_TEXT=azure, Azure region in which the Speech To Text service is located (e.g. 'westeurope'). See Azure Speech To Text docs
OPENVIDU_PRO_SPEECH_TO_TEXT_VOSK_MODEL_LOAD_STRATEGY If OPENVIDU_PRO_SPEECH_TO_TEXT=vosk, the load strategy for the language models. It can be:
  • on_demand: Vosk models will be loaded into memory just when they are needed, and will be unloaded when they are no longer needed.
  • manual: Vosk models can only be loaded and unloaded with the REST API. See load and unload methods.
 on_demand
OPENVIDU_PRO_SPEECH_TO_TEXT_IMAGE Docker image to use as Speech To Text module. If using OPENVIDU_PRO_SPEECH_TO_TEXT=vosk, this property allows to use a custom image with other Vosk language models than the default ones. See Using custom languages. You can configure private Docker registries using property OPENVIDU_PRO_DOCKER_REGISTRIES
OPENVIDU_PRO_DOCKER_REGISTRIES List of Docker registries to be used by Media Nodes. This is useful for those Docker images that may be overwritten with custom images, which includes the images defined by properties KMS_IMAGE and OPENVIDU_PRO_SPEECH_TO_TEXT_IMAGE. The list is a JSON array of strings, each one with the following properties:
"serveraddress=test.io,username=test,password=test"		 [ ]
OPENVIDU_BROWSER_LOGS How openvidu-browser logs of clients should be sent to Elasticsearch:
  • disabled: Don't send logs.
  • debug: Send all openvidu-browser logs.
  • debug_app: Send openvidu-browser logs and frontend app logs.
 disabled
OPENVIDU_PRO_RECORDING_STORAGE Where to store recordings:
  • local: Use local storage.
  • s3: Use an AWS S3 bucket
See Uploading recordings to AWS S3.		 local
OPENVIDU_PRO_AWS_S3_BUCKET If OPENVIDU_PRO_RECORDING_STORAGE=s3, AWS S3 bucket where to store recordings. May include paths to allow navigating folder structures inside the bucket.
OPENVIDU_PRO_AWS_S3_HEADERS If OPENVIDU_PRO_RECORDING_STORAGE=s3, the collection of HTTP header values that the internal AWS client will use during the upload process. The property is a key-value map of strings, following the format of a JSON object.
For example, to apply server-side encryption with AES-256, this header is mandatory:
{"x-amz-server-side-encryption":"AES256"}
The list of available headers can be found here.		 { }
OPENVIDU_PRO_AWS_S3_SERVICE_ENDPOINT If OPENVIDU_PRO_RECORDING_STORAGE=s3 and you want an S3 provider different to AWS, the service endpoint of the S3 bucket. This allows using any kind of S3 compatible buckets, not only AWS.
For example: https://s3.us-west-002.backblazeb2.com
OPENVIDU_PRO_AWS_S3_WITH_PATH_STYLE_ACCESS If OPENVIDU_PRO_RECORDING_STORAGE=s3, true to always use path-style access for all requests. This may be mandatory to avoid Access Denied errors when accessing non-AWS S3 buckets from domain names and not IPs. See AWS docs Path-style requests.
OPENVIDU_PRO_AWS_ACCESS_KEY This property applies to
OPENVIDU_PRO_RECORDING_STORAGE=s3 and/or
OPENVIDU_PRO_SPEECH_TO_TEXT=aws. It is the AWS long-lived credentials access key. Depending on the service you have enabled:
  • OPENVIDU_PRO_RECORDING_STORAGE=s3: Must have read and write permissions over the bucket defined in property OPENVIDU_PRO_AWS_S3_BUCKET. In this case credentials are optional: if not provided the internal S3 client will try to use the default AWS credentials of the Master Node, if available (see the credentials search order in the Java Doc).
  • OPENVIDU_PRO_SPEECH_TO_TEXT=aws: Must have permissions to manage Amazon Transcribe services. In this case credentials are mandatory.
OPENVIDU_PRO_AWS_SECRET_KEY This property applies to
OPENVIDU_PRO_RECORDING_STORAGE=s3 and/or
OPENVIDU_PRO_SPEECH_TO_TEXT=aws. It is the AWS long-lived credentials secret key. Depending on the service you have enabled:
  • OPENVIDU_PRO_RECORDING_STORAGE=s3: In this case credentials are optional: if not provided then the internal S3 client will try to use the default AWS credentials of the machine, if available (see the credentials search order in the Java Doc).
  • OPENVIDU_PRO_SPEECH_TO_TEXT=aws: In this case credentials are mandatory.
OPENVIDU_PRO_AWS_REGION This property applies to
OPENVIDU_PRO_RECORDING_STORAGE=s3 and/or
OPENVIDU_PRO_SPEECH_TO_TEXT=aws. It is the AWS region hosting the services. Depending on the service you have enabled:
  • OPENVIDU_PRO_RECORDING_STORAGE=s3: AWS region in which the S3 bucket is located (e.g. "eu-west-1"). If not provided, the region will try to be discovered automatically, although this is not always possible.
  • OPENVIDU_PRO_SPEECH_TO_TEXT=aws: AWS region where Amazon Transcribe will operate. In this case the property is always mandatory.
OPENVIDU_PRO_PRIVATE_IP Private IP of Master Node. Media Nodes will use this property to communicate with Master Node. This property is optional and as a general rule should remain empty, as Media Nodes will automatically discover this IP.
It is only suitable for On Premises deployments when Media Nodes are not able to autodiscover the IP address for any kind of not contemplated reason.
OPENVIDU_PRO_LICENSE_HTTP_PROXY OpenVidu Pro or Enterprise sends periodically information about CPU usage and license checks. All of those requests requires HTTP/HTTPS access to the Internet to .

As some environments may use HTTP proxies to access the internet for security reasons, this parameter allows OpenVidu Pro and Enterprise to send license information and CPU usage using a configured proxy in your infrastructure.

An example of this parameter would be: your.proxy:8080
KMS_URIS Array of Media Node URIs (comma-separated list of strings) that should be automatically added to the cluster on startup. If any value is provided, then it must be valid and reachable, or OpenVidu will fail to start.
This property has effect only on On Premises deployments. All of Media Node URIs have the same format:
ws://MEDIA_NODE_IP:8888/kurento
For example, a possible value for this property for a single Media Node might look like this:
["ws://203.0.113.2:8888/kurento"]		 [ ]
ELASTICSEARCH_USERNAME When using a secured external Elastic stack, the name of the Elastic user.
ELASTICSEARCH_PASSWORD When using a secured external Elastic stack, the password of the Elastic user.


Configuration parameters for OpenVidu Enterprise 🔗

These configuration parameters apply only to OpenVidu Enterprise.

Parameter Description Default value
OPENVIDU_ENTERPRISE_MEDIA_SERVER Which media server to use in your deployment (see Kurento vs mediasoup). Possible values are:
• kurento
• mediasoup
 mediasoup
OPENVIDU_ENTERPRISE_S3_CONFIG_AUTORESTART If true, master nodes restart automatically on S3 changes. true
OPENVIDU_WEBRTC_SIMULCAST Whether to enable simulcast for all Publishers or not false


Advanced parameters for OpenVidu Pro and Enterprise 🔗

These properties are related to infrastructure and network aspects. OpenVidu is deployed with sane defaults, and these configuration properties will normally not require changes. They are only described here for custom on premises deployments with special network conditions or requirements, and only meant for users with deep knowledge about their deployment environment.
Parameter Description Default value
COTURN_IP This is the IP of the default TURN/STUN Coturn server deployed with OpenVidu in the Master Node.
  • By default it is the autodiscovered Public IPv4. This IP is used by browsers and Kurento Media Server (mediasoup does not need it) as TURN/STUN service for ICE protocol to autodiscover its own Public IP.
  • This parameter is used when OPENVIDU_PRO_COTURN_IN_MEDIA_NODES=false. When
    OPENVIDU_PRO_COTURN_IN_MEDIA_NODES=true the public IP used for Coturn is the public IPv4 of the Media Node being in use, ignoring any value of COTURN_IP
  • Possible values:
    • auto-ipv4
    • auto-ipv6
    • Any valid Public Ipv4 or Public IPv6
 auto-ipv4
COTURN_PORT This is the port used for TURN/STUN Coturn server deployed with OpenVidu in the Master Node.
  • By default it is 3478. This port is used by browsers and Kurento Media Server (mediasoup does not need it) as TURN/STUN service port for ICE protocol to autodiscover its own Public IP.
  • This parameter is used only if OPENVIDU_PRO_COTURN_IN_MEDIA_NODES=false. When
    OPENVIDU_PRO_COTURN_IN_MEDIA_NODES=true, this property is ignored and OPENVIDU_PRO_COTURN_PORT_MEDIA_NODES is used instead (by default 443)
  • Possible values: Any value between 1-65535
 3478
OPENVIDU_WEBRTC_ICE_SERVERS Array of ICE servers to use instead of default Coturn at the client side (comma-separated list of ICE Servers). When OPENVIDU_WEBRTC_ICE_SERVERS is used, Kurento
still uses COTURN_IP (or the public IP of Media Nodes if OPENVIDU_PRO_COTURN_IN_MEDIA_NODES=true) but browsers uses the TURN/STUN servers defined at OPENVIDU_WEBRTC_ICE_SERVERS. Examples:
  • ["url=turns:example.turn.com:443,staticAuthSecret=secret"]
  • ["url=turns:example.turn.com:443,username=usertest,credential=userpass"]
More Details		 [ ]
COTURN_INTERNAL_RELAY If true, Coturn will autodiscover its own default network gateway and use the default network gateway IP as a relay IP for ICE candidates. This is useful if you want relayed traffic to go through the internal network instead of the public internet false
OPENVIDU_PRO_COTURN_IN_MEDIA_NODES (Experimental) If true, Coturn will be deployed in Media Nodes. As Public IP and Port configuration for Coturn, parameters OPENVIDU_PRO_MEDIA_NODE_PUBLIC_IP_AUTODISCOVER and OPENVIDU_PRO_COTURN_PORT_MEDIA_NODES configure how the Public IP is autodiscovered and what port is used respectively. false
OPENVIDU_PRO_MEDIA_NODE_PUBLIC_IP_AUTODISCOVER This parameter defines how the public IP for Coturn is autodiscovered in Media Nodes when OPENVIDU_PRO_COTURN_IN_MEDIA_NODES=true. This IP is used by browsers and Kurento Media Server (mediasoup does not need it) as TURN/STUN service for ICE protocol. Possible values:
  • auto-ipv4
  • auto-ipv6
 false
OPENVIDU_PRO_COTURN_PORT_MEDIA_NODES This parameter defines what port is used for Coturn in Media Nodes when OPENVIDU_PRO_COTURN_IN_MEDIA_NODES=true. This port is used by browsers and Kurento Media Server (mediasoup does not need it) as TURN/STUN service port for ICE protocol. Possible values: Any value between 1-65535 443
KMS_IMAGE Docker image used in Media Nodes for Kurento in OpenVidu Pro/Enterprise. By default it is the correspondant image of the deployed version (see). If you have created your own Kurento image and it is stored in a private Docker registry, you can configure it using property OPENVIDU_PRO_DOCKER_REGISTRIES
KMS_DOCKER_ENV_* You can add any environment variable to Kurento/mediasoup running in the Media Node. If you want to add an environment variable to this container, you must add a variable using this prefix: KMS_DOCKER_ENV_, followed by the environment variable you want to setup. For example, if you want to setup KMS_MIN_PORT to 50000, it would be KMS_DOCKER_ENV_KMS_MIN_PORT=50000.
AWS_* These parameters are used in OpenVidu Pro and OpenVidu Enterprise for AWS related issues, in AWS deployments of OpenVidu. You should not touch these parameters, as they are automatically initialized on Cloudformation launch.

Special conditions of OpenVidu development container 🔗

When using the official development OpenVidu docker container to develop your app in your LAN network, there are some properties that you can't make use of. In the development container these configuration properties won't have effect, or may have unwanted side effects if declared. Avoid declaring them:

  • HTTP_PORT: in LAN networks redirects or Let's Encrypt don't make sense.
  • CERTIFICATE_TYPE: in LAN networks certificates don't make sense.

Also, take into account that these configuration properties have different default values in the development container:

  • DOMAIN_OR_PUBLIC_IP: default to localhost in the dev container. Container won't fail to start if you don't provide a value to it.
  • OPENVIDU_SECRET: default to MY_SECRET in the dev container. Container won't fail to start if you don't provide a value to it.
  • HTTPS_PORT: default to 4443 in the dev container.

Below there is an example of a docker run command to launch the OpenVidu development container while setting multiple configuration properties:

docker run -p 4443:4443 --rm \
    -e OPENVIDU_SECRET=my_password \
    -e OPENVIDU_RECORDING=true \
    -e OPENVIDU_RECORDING_PATH=/home/openvidu/recordings \
    -v /var/run/docker.sock:/var/run/docker.sock \
    -v /home/openvidu/recordings:/home/openvidu/recordings \
    -e OPENVIDU_CDR=true \
    -e OPENVIDU_CDR_PATH=/home/openvidu/cdr \
    -v /home/openvidu/cdr:/home/openvidu/cdr \
    -e OPENVIDU_WEBHOOK=true \
    -e OPENVIDU_WEBHOOK_ENDPOINT=https://localhost:7777/my-endpoint \
openvidu/openvidu-dev:latest