Configurações Jitsi Docker
Procedimentos de configurações e customizações.

Jitsi Meet Load Balancing
Link: https://meetrix.io/blog/webrtc/jitsi/jitsi-meet-load-balancing.html 

 Setup multi server, load balanced videobriges 

 

 One of the amazing features in Jitsi Meet is the inbuilt horizontal scalability. When you want to cater large number of concurrent users, you can spin up multiple video bridges to handle the load. If we setup mulple video bridges and connect them to the same shard, Jicofo, the conference manager selects the least loaded Videobridge for the next new conference. 

 Run Prosody on all interfaces 

 By default prosody runs only on local interface (127.0.0.0), to allow the xmpp connections from external servers, we have to run prosody on all interfaces. To do that, add the following line at the beginning of  /etc/prosody/prosody.cfg.lua 

 component_interface = "0.0.0.0" 

 Allow inbound traffic for port 5222 

 Open inbound traffic from JVB server on port  5222  (TCP) of Prosody server. But DO NOT open this publicly. 

 Install JVB on a seperate server 

 Install a Jitsi Video Bridge on a different server. 

 Copy JVB configurations from Jitsi Meet server 

 Replace  /etc/jitsi/videbridge/config  and  /etc/jitsi/video-bridge/sip-communicator.properties  of JVB server with the same files from the original Jitsi Meet server 

 Update JVB config file 

 In  /etc/jitsi/videbridge/config  set the  XMPP_HOST  to the ip address/domain of the prosody server 

 # Jitsi Videobridge settings

# sets the XMPP domain (default: none)

JVB_HOSTNAME=example.com

# sets the hostname of the XMPP server (default: domain if set, localhost otherwise)

JVB_HOST=<XMPP_HOST>

# sets the port of the XMPP server (default: 5275)

JVB_PORT=5347

# sets the shared secret used to authenticate to the XMPP server

JVB_SECRET=<JVB_SECRET>

# extra options to pass to the JVB daemon

JVB_OPTS="--apis=rest,xmpp"

# adds java system props that are passed to jvb (default are for home and logging config file)

JAVA_SYS_PROPS="-Dnet.java.sip.communicator.SC_HOME_DIR_LOCATION=/etc/jitsi -Dnet.java.sip.communicator.SC_HOME_DIR_NAME=videobridge -Dnet.java.sip.communicator.SC_LOG_DIR_LOCATION=/var/log/jitsi -Djava.util.logging.config.file=/etc/jitsi/videobridge/logging.properties" 

 In  /etc/jitsi/video-bridge/sip-communicator.properties  file update the following properties 

 

 <XMPP_HOST> : The ip address of the prosody server. Better if you can use the private IP address if that can be accessed from JVB server. 

 <JVB_NICKNAME> : This should be a unique string used by Jicofo to identify each JVB 

 

 org.ice4j.ice.harvest.DISABLE_AWS_HARVESTER=true

org.ice4j.ice.harvest.STUN_MAPPING_HARVESTER_ADDRESSES=meet-jit-si-turnrelay.jitsi.net:443

org.jitsi.videobridge.ENABLE_STATISTICS=true

org.jitsi.videobridge.STATISTICS_TRANSPORT=muc

org.jitsi.videobridge.xmpp.user.shard.HOSTNAME=<XMPP_HOST>

org.jitsi.videobridge.xmpp.user.shard.DOMAIN=auth.example.com

org.jitsi.videobridge.xmpp.user.shard.USERNAME=jvb

org.jitsi.videobridge.xmpp.user.shard.PASSWORD=<JVB_SECRET>

org.jitsi.videobridge.xmpp.user.shard.MUC_JIDS=JvbBrewery@internal.auth.example.com

org.jitsi.videobridge.xmpp.user.shard.MUC_NICKNAME=<JVB_NICKNAME>

org.jitsi.videobridge.xmpp.user.shard.DISABLE_CERTIFICATE_VERIFICATION=true 

 You can add the following line at the beginning of  /usr/share/jitsi/jvb/jvb.sh  to generate a unique nickname for the JVB at each startup. This might be useful if you are using an auto-scaling mechanism. 

 sed -i "s/org.jitsi.videobridge.xmpp.user.shard.MUC_NICKNAME=.*/org.jitsi.videobridge.xmpp.user.shard.MUC_NICKNAME=$(cat /proc/sys/kernel/random/uuid)/g" /etc/jitsi/videobridge/sip-communicator.properties 

 Looking for commercial support for Jitsi Meet ?  Please contact us via  hello@meetrix.io 

 Updated:   October 19, 2020

Load Testing Jitsi Meet
Link: https://meetrix.io/blog/webrtc/jitsi/jitsi-meet-load-testing.html 

 Make sure your Jitsi Meet infrastructure is ready for production 

 

 By deploying a horizontally scalable Jitsi Meet setup, you can scale your Jitsi Meet conferencing infrastructure to cater thousands of concurrent users. But, before you go live you might want to make sure that your infrastructure is capable of handling the desired number of concurrent users. Jitsi Meet Torture  is a tool that can be used to run tests against a Jitsi Instance and it is capable of load testing Jitsi Meet. 

 

 Jitsi Meet Load Testing with Torture 

 

 Selenium Grid 

 To simulate hundrads of concurrent users, we need to deploy Jitsi Meet Torture agains a selenium grid setup. There would be a  selenium hub  which accepts commands from Jitsi Meet Torture and pass them to  selenium nodes . You can have hundrads of  selenium nodes  to simulate users. 

 

 Jitsi Meet Load Testing with Torture 

 

 

 Overwhelmed with managing Jitsi Infrastructure? 

 Outsource full-time, high-cost Jitsi infrastructure management and maintenance 

 Talk to an expert 

 Minimal setup with docker-compose 

 

 

 Install Docker. You can use following scripts on Ubuntu 18.04 

 sudo apt-get update

sudo apt-get -y install apt-transport-https ca-certificates curl software-properties-common

curl -fsSL https://download.docker.com/linux/ubuntu/gpg | sudo apt-key add -

sudo add-apt-repository -y \

"deb [arch=amd64] https://download.docker.com/linux/ubuntu \

$(lsb_release -cs) \

stable"

sudo apt-get -y update

sudo apt-get -y install docker-ce

sudo usermod -a -G docker $(whoami)

 

 

 

 Install the latest version of  Docker Compose 

 

 

 Create a  docker-compose.yml  file with following content. 

 version: "3.3"

services:

 torture:

 image: meetrix/jitsi-meet-torture

 hub:

 image: selenium/hub:3.141.59

 node:

 build: ./node

 image: meetrix/jitsi-meet-torture-selenium-node

 volumes:

 - /dev/shm:/dev/shm

 depends_on:

 - hub

 environment:

 HUB_HOST: hub

 

 

 

 Please note that  node  are configured to run only one chrome instance at a time. Run the setup with  docker-compose up -d --scale node=2  to run two nodes. 

 

 

 docker-compose exec torture /bin/bash 

 

 

 Change  <YOUR_JITSI_MEET_INSTANCE_URL>  to your Jitsi Meet installation (for example  https://meet.example.com ) in following script and run to perform the test. 

 

 

 ./scripts/malleus.sh --conferences=1 --participants=2 --senders=2 --audio-senders=2 --duration=300 --room-name-prefix=test --hub-url=http://hub:4444/wd/hub --instance-url=<YOUR_JITSI_MEET_INSTANCE_URL>

 

 Running the test with large number of nodes 

 You can scale the number of nodes with  docker-compose up --scale node=<NUMBER>  or you can create a VM using same docker images for  selenium hub  and  selenium-node . Then you can use an autoscaling mechanism (Such as Autoscaling groups on AWS) to scale the number of nodes. 

 Looking for commercial support for Jitsi Meet ?  Please contact us via  hello@meetrix.io 

 Updated:   June 19, 2020

Self-Hosting Guide - Log Analyser
Link: https://jitsi.github.io/handbook/docs/devops-guide/devops-guide-log-analyser       Last updated on Mar 7, 2025 

 Welcome to the Jitsi Meet Log Analyser project! This integration leverages Grafana Loki and OpenTelemetry to enable effective log management and analysis for Jitsi Meet components. 

 Overview ​ 

 This project offers a streamlined setup for collecting, parsing, and visualizing log data generated by Jitsi Meet components. The integration includes: 

 

 

 A Docker Compose setup file ( log-analyser.yml ) for configuring Loki and OpenTelemetry Collector. 

 A Docker Compose setup file ( grafana.yml ) for configuring Grafana. 

 A unified Docker Compose command that initiates all services concurrently. 

 Instructions on configuring Grafana with Loki as a log data source for enhanced visualization and analysis. 

 

 

 Getting Started ​ 

 Prerequisites ​ 

 Before you begin, ensure you have the following installed on your system: 

 

 

 Docker 

 Docker Compose 

 

 

 Setup ​ 

 To set up the project, follow these steps: 

 

 

 

 Clone the repository: 

 Clone the Jitsi Meet repository that contains the necessary Docker Compose files. 

 git clone https://github.com/jitsi/docker-jitsi-meet.git

cd docker-jitsi-meet 

 

 

 

 

 

 

 

 Update Jitsi Meet Docker Compose Configuration: 

 To enable log collection and analysis, you need to modify the  docker-compose.yml  file for Jitsi Meet components. Add the following configuration to each Jitsi service within the  docker-compose.yml  file: 

 logging:

 driver: "json-file"

 options:

 labels: "service" 

 

 

 

 

 

 This configuration ensures that logs are collected in JSON format and tagged with service labels, which is essential for Loki to properly ingest and index the logs. 

 

 

 Start the Docker containers: 

 Use the following command to start all required services, including the Jitsi Meet components, Grafana, Loki, and OpenTelemetry: 

 docker-compose -f docker-compose.yml -f log-analyser.yml -f grafana.yml up -d 

 

 

 

 

 

 

 

 Explanation: 

 

 The command combines multiple Docker Compose files to launch the entire stack:

 

 docker-compose.yml  launches the Jitsi Meet components. 

 log-analyser.yml  sets up the log analysis tools, including Loki and OpenTelemetry. 

 grafana.yml  initializes Grafana for log visualization. 

 

 

 Logs generated by the Jitsi Meet components are automatically forwarded to Loki via the OpenTelemetry Collector, making them available for analysis in Grafana. 

 

 

 

 Note:  If you only want to use Grafana for log visualization without the log analysis tools, you can run  grafana.yml  independently. However, for the complete log analysis setup, both  log-analyser.yml  and  grafana.yml  are required. 

 

 

 

 

 

 Access Grafana ​ 

 After starting the services, follow these steps to access Grafana: 

 

 

 

 Open Grafana in your web browser: 

 Navigate to  http://localhost:3000  to access the Grafana interface. 

 

 

 Log in to Grafana: 

 Use the default login credentials provided below: 

 

 Username:   admin 

 Password:   admin 

 

 (It is recommended to change these credentials after the first login for security purposes.) 

 

 

 

 Pre-configured Dashboards ​ 

 Grafana comes with several pre-configured dashboards specifically designed for monitoring different Jitsi Meet components. These dashboards will be automatically available once you log in to Grafana. 

 Important❗️:  For data to appear in these dashboards, logs need to be generated by the Jitsi Meet components. Here are the available dashboards: 

 

 

 Jicofo Dashboard:  Visualizes logs related to Jitsi Conference Focus (Jicofo), which handles media and signaling in Jitsi Meet. 

 JVB Dashboard:  Focuses on Jitsi Videobridge (JVB) logs, showing details on video streaming and performance metrics. 

 Prosody Dashboard:  Monitors Prosody, the XMPP server used by Jitsi Meet for signaling. 

 Web Dashboard:  Displays logs and metrics related to the web frontend of Jitsi Meet. 

 Jitsi All Dashboard:  A comprehensive dashboard that aggregates logs from all Jitsi Meet components. 

 

 

 

 Filtering Logs with LogQL ​ 

 Beyond the pre-configured dashboards, you can explore and filter logs in Grafana's "Explore" section using LogQL, a powerful query language designed for Loki. Here’s how you can filter and analyze logs: 

 

 

 

 Access the Explore section: 

 In Grafana, navigate to the  Explore  tab from the left sidebar. This section allows you to run queries on your logs in real-time. 

 

 

 Select the Loki Data Source: 

 In the Explore section, ensure that the  Data Source  is set to  Loki . This is essential because Loki is the backend that stores and manages the log data collected from Jitsi Meet components. 

 

 

 Using LogQL for Filtering: 

 LogQL enables you to create complex queries to filter specific logs. For example, the following LogQL query filters logs for the  jitsi-jicofo  component and parses them: 

 {exporter="OTLP"} | json | attributes_attrs_service="jitsi-jicofo" 

 

 

 

 

 

 

 Explanation: 

 

 {exporter="OTLP"} : Selects logs that are exported via OpenTelemetry. 

 | json : Parses the log data as JSON, making attributes accessible for filtering. 

 attributes_attrs_service="jitsi-jicofo" : Filters logs where the  attributes_attrs_service  field equals  "jitsi-jicofo" . 

 

 

 

 You can modify the query to filter logs from other components or adjust the criteria according to your needs. 

 For more details and advanced usage of LogQL, you can refer to the official  LogQL documentation  from Grafana. 

 

 

 

 

 Usage ​ 

 Once the setup is complete, you can start exploring your log data in Grafana. The pre-configured dashboards provide an insightful visualization of the logs collected from Jitsi Meet components. Use these dashboards to: 

 

 

 Parse Logs:  View detailed logs collected from various components. 

 Visualize Logs:  Analyze log data through various charts, graphs, and panels to gain insights into system performance and issues. 

 

 

 Customizing Dashboards ​ 

 While the pre-configured dashboards provide a solid starting point, you may want to customize them or create new dashboards to suit your specific needs. Here's how you can do that: 

 

 

 

 Create a New Dashboard: 

 

 Go to the Grafana home page, click on the "+" icon on the left sidebar, and select "Dashboard." 

 Add panels to visualize different metrics or logs by selecting the appropriate data source (Loki) and using LogQL queries. 

 

 

 

 Customize Existing Dashboards: 

 

 Navigate to an existing dashboard and click on the "Edit" button (pencil icon) on any panel. 

 Adjust the LogQL query, visualization type, and panel settings to match your requirements. 

 

 

 

 Save and Share Dashboards: 

 

 After customizing, save the dashboard. You can also export it as a JSON file to share with others or for backup. 

 To export the dashboard:

 

 Click on the dashboard title to open the options menu. 

 Select  "Dashboard settings"  >  "JSON Model" . 

 Click  "Download JSON"  to save the file locally. 

 

 

 

 

 

 Contribute to Jitsi Meet Dashboards: 

 

 If you've created or customized a dashboard that could benefit the wider Jitsi community, you can contribute by updating the relevant dashboard JSON file. 

 Here’s how to do it:

 

 Export the JSON file of your customized dashboard as described above. 

 Locate the corresponding Jitsi component dashboard JSON file in the repository (e.g.,  jicofo-dashboard.json ,  jvb-dashboard.json ). 

 Update the existing JSON file with your changes. 

 Submit a pull request to the repository with the updated JSON file and a brief description of the changes you made. 

 

 

 This helps improve the pre-configured dashboards and makes your contributions available to all users. 

 

 

 

 Grafana Provisioning: 

 

 The Jitsi Meet Log Analyser uses Grafana provisioning to manage dashboards. When you update a dashboard JSON file in the repository, it will be automatically provisioned in Grafana when the stack is deployed. 

 This ensures that everyone using the repository gets the latest version of the dashboards without needing to manually import them. 

 

 

 

 

 By following these steps, you can not only customize your own monitoring setup but also contribute improvements back to the Jitsi community. 

 Troubleshooting ​ 

 If you encounter issues while setting up or using the Jitsi Meet Log Analyser, here are some common problems and their solutions: 

 

 

 

 Grafana Not Starting: 

 

 Check if the Grafana container is running with  docker ps  and inspect logs using  docker logs grafana  for any errors. 

 

 

 

 No Logs in Grafana Dashboards: 

 

 Ensure that Jitsi Meet components are generating logs. Clear browser cache, reload Grafana. Ensure OpenTelemetry, Loki, and Grafana containers are all running with  docker ps , and inspect each container's logs for issues using  docker logs <container_name> . 

 

 

 

 OpenTelemetry Collector Not Forwarding Logs: 

 

 Check OpenTelemetry's logs with  docker logs otel , ensure it's connected to the correct endpoints, and verify the log format is correct. 

 

 

 

 Authentication Failures in Grafana: 

 

 Restart Grafana with  docker restart grafana otel . If still unsuccessful, delete the data volume with  docker-compose down -v  and restart to reset to default credentials (admin/admin). 

 

 

 

 Slow Queries: 

 

 If LogQL queries are slow, try optimizing the query in Grafana. 

 

 

 

 Permission Issues: 

 

 If you encounter permission issues, make sure that Docker has the necessary access rights to the directories where logs are stored. 

 

 

 

 Docker Network Issues: 

 

 Verify Docker network connections, IP range, and restart the network if necessary. 

 

 

 

 OpenTelemetry Collector Not Forwarding Logs: 

 

 Check OpenTelemetry logs, verify configuration, and ensure log format compatibility. 

 

 

 

 Docker Containers Failing to Start: 

 

 Use  docker-compose logs  to view detailed startup errors, and check for common issues like incorrect configurations. 

 

 

 

 

 Acknowledgements ​ 

 We appreciate your interest in the Jitsi Meet Log Analyser project! If you encounter any issues or have questions, feel free to reach out to the project maintainers or contribute to the repository. 

 Last updated on Mar 7, 2025