Stand with Ukraine flag
Try it now Pricing
Community Edition
Community Edition Professional Edition Cloud Edge PE Edge IoT Gateway License Server Trendz Analytics Mobile Application PE Mobile Application MQTT Broker
Installation > On premise > Docker (Linux or Mac OS)
Getting Started Documentation Devices Library Guides
Architecture API FAQ
On this page

Installing ThingsBoard using Docker (Linux or Mac OS)

Doc info icon
ThingsBoard Cloud

We recommend to use ThingsBoard Cloud - fully managed, scalable and fault-tolerant platform for your IoT applications
ThingsBoard Cloud is for everyone who would like to use ThingsBoard but don’t want to host their own instance of the platform.

This guide will help you to install and start ThingsBoard using Docker on Linux or MacOS.

Prerequisites

Install Docker:

doc warn icon

Don’t forget to add your linux user to the docker group. See Manage Docker as a non-root user.

Running

Running this image requires a server with at least 4GB of RAM (8GB is recommended) and minimum load (few messages per second).

Choose ThingsBoard queue service

ThingsBoard platform currently supports two type of messaging brokers for storing the messages and communication between ThingsBoard services: In-memory and Kafka-based brokers.

  • In Memory queue implementation is built-in and default. It is useful for development(PoC) environments and is not suitable for production deployments or any sort of cluster deployments.

  • Kafka is recommended for production deployments. This queue is used on the most of ThingsBoard production environments now. It is useful for both on-prem and private cloud deployments. It is also useful if you like to stay independent from your cloud provider. However, some providers also have managed services for Kafka. See AWS MSK for example.

  • Confluent Cloud is a fully managed streaming platform based on Kafka. Useful for a cloud agnostic deployments.

See corresponding architecture page and rule engine page for more details.

In Memory (built-in and default)
Kafka (recommended for on-prem, production installations)
Confluent Cloud (Event Streaming Platform based on Kafka)

ThingsBoard includes In Memory Queue service and use it by default without extra settings.

Create docker compose file for ThingsBoard queue service:

1

nano docker-compose.yml

Add the following lines to the yml file:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35

services:
  postgres:
    restart: always
    image: "postgres:16"
    ports:
      - "5432"
    environment:
      POSTGRES_DB: thingsboard
      POSTGRES_PASSWORD: postgres
    volumes:
      - postgres-data:/var/lib/postgresql/data
  thingsboard-ce:
    restart: always
    image: "thingsboard/tb-node:4.0.1.1"
    ports:
      - "8080:8080"
      - "7070:7070"
      - "1883:1883"
      - "8883:8883"
      - "5683-5688:5683-5688/udp"
    logging:
      driver: "json-file"
      options:
        max-size: "100m"
        max-file: "10"
    environment:
      TB_SERVICE_ID: tb-ce-node
      SPRING_DATASOURCE_URL: jdbc:postgresql://postgres:5432/thingsboard
    depends_on:
      - postgres

volumes:
  postgres-data:
    name: tb-postgres-data
    driver: local

Apache Kafka is an open-source stream-processing software platform.

Create docker compose file for ThingsBoard queue service:

1

nano docker-compose.yml

Add the following lines to the yml file.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66

services:
  postgres:
    restart: always
    image: "postgres:16"
    ports:
      - "5432"
    environment:
      POSTGRES_DB: thingsboard
      POSTGRES_PASSWORD: postgres
    volumes:
      - postgres-data:/var/lib/postgresql/data
  kafka:
    restart: always
    image: bitnami/kafka:4.0
    ports:
      - 9092:9092 #to localhost:9092 from host machine
      - 9093 #for Kraft
    environment:
      ALLOW_PLAINTEXT_LISTENER: "yes"
      KAFKA_CFG_LISTENERS: "PLAINTEXT://:9092,CONTROLLER://:9093"
      KAFKA_CFG_ADVERTISED_LISTENERS: "PLAINTEXT://:9092"
      KAFKA_CFG_LISTENER_SECURITY_PROTOCOL_MAP: "CONTROLLER:PLAINTEXT,PLAINTEXT:PLAINTEXT"
      KAFKA_CFG_INTER_BROKER_LISTENER_NAME: "PLAINTEXT"
      KAFKA_CFG_AUTO_CREATE_TOPICS_ENABLE: "false"
      KAFKA_OFFSETS_TOPIC_REPLICATION_FACTOR: "1"
      KAFKA_TRANSACTION_STATE_LOG_MIN_ISR: "1"
      KAFKA_TRANSACTION_STATE_LOG_REPLICATION_FACTOR: "1"
      KAFKA_CFG_PROCESS_ROLES: "controller,broker" #KRaft
      KAFKA_CFG_NODE_ID: "0" #KRaft
      KAFKA_CFG_CONTROLLER_LISTENER_NAMES: "CONTROLLER" #KRaft
      KAFKA_CFG_CONTROLLER_QUORUM_VOTERS: "0@kafka:9093" #KRaft
      KAFKA_CFG_LOG_RETENTION_MS: "300000"
      KAFKA_CFG_SEGMENT_BYTES: "26214400"
    volumes:
      - kafka-data:/bitnami
  thingsboard-ce:
    restart: always
    image: "thingsboard/tb-node:4.0.1.1"
    ports:
      - "8080:8080"
      - "7070:7070"
      - "1883:1883"
      - "8883:8883"
      - "5683-5688:5683-5688/udp"
    logging:
      driver: "json-file"
      options:
        max-size: "100m"
        max-file: "10"
    environment:
      TB_SERVICE_ID: tb-ce-node
      SPRING_DATASOURCE_URL: jdbc:postgresql://postgres:5432/thingsboard
      TB_QUEUE_TYPE: kafka
      TB_KAFKA_SERVERS: kafka:9092
    depends_on:
      - postgres
      - kafka

volumes:
  postgres-data:
    name: tb-postgres-data
    driver: local
  kafka-data:
    name: tb-ce-kafka-data
    driver: local

Confluent Cloud Configuration

To access Confluent Cloud you should first create an account, then create a Kafka cluster and get your API Key.

Create docker compose file for ThingsBoard queue service:

1

nano docker-compose.yml

Add the following line to the yml file. Don’t forget to replace “CLUSTER_API_KEY”, “CLUSTER_API_SECRET” and “localhost:9092” with your real Confluent Cloud bootstrap servers:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64

services:
  postgres:
    restart: always
    image: "postgres:16"
    ports:
      - "5432"
    environment:
      POSTGRES_DB: thingsboard
      POSTGRES_PASSWORD: postgres
    volumes:
      - postgres-data:/var/lib/postgresql/data
  thingsboard-ce:
    restart: always
    image: "thingsboard/tb-node:4.0.1.1"
    ports:
      - "8080:8080"
      - "7070:7070"
      - "1883:1883"
      - "8883:8883"
      - "5683-5688:5683-5688/udp"
    logging:
      driver: "json-file"
      options:
        max-size: "100m"
        max-file: "10"
    environment:
      TB_SERVICE_ID: tb-ce-node
      SPRING_DATASOURCE_URL: jdbc:postgresql://postgres:5432/thingsboard
      TB_QUEUE_TYPE: kafka
      TB_KAFKA_SERVERS: localhost:9092
      TB_QUEUE_KAFKA_REPLICATION_FACTOR: 3
      TB_QUEUE_KAFKA_USE_CONFLUENT_CLOUD: true
      TB_QUEUE_KAFKA_CONFLUENT_SASL_JAAS_CONFIG: 'org.apache.kafka.common.security.plain.PlainLoginModule required username="CLUSTER_API_KEY" password="CLUSTER_API_SECRET";'
      # These params affect the number of requests per second from each partitions per each queue.
      # Number of requests to particular Message Queue is calculated based on the formula:
      # ((Number of Rule Engine and Core Queues) * (Number of partitions per Queue) + (Number of transport queues)
      #  + (Number of microservices) + (Number of JS executors)) * 1000 / POLL_INTERVAL_MS
      # For example, number of requests based on default parameters is:
      # Rule Engine queues:
      # Main 10 partitions + HighPriority 10 partitions + SequentialByOriginator 10 partitions = 30
      # Core queue 10 partitions
      # Transport request Queue + response Queue = 2
      # Rule Engine Transport notifications Queue + Core Transport notifications Queue = 2
      # Total = 44
      # Number of requests per second = 44 * 1000 / 25 = 1760 requests
      # 
      # Based on the use case, you can compromise latency and decrease number of partitions/requests to the queue, if the message load is low.
      # By UI set the parameters - interval (1000) and partitions (1) for Rule Engine queues.
      # Sample parameters to fit into 10 requests per second on a "monolith" deployment: 
      TB_QUEUE_CORE_POLL_INTERVAL_MS: 1000
      TB_QUEUE_CORE_PARTITIONS: 2
      TB_QUEUE_RULE_ENGINE_POLL_INTERVAL_MS: 1000
      TB_QUEUE_TRANSPORT_REQUEST_POLL_INTERVAL_MS: 1000
      TB_QUEUE_TRANSPORT_RESPONSE_POLL_INTERVAL_MS: 1000
      TB_QUEUE_TRANSPORT_NOTIFICATIONS_POLL_INTERVAL_MS: 1000
      TB_QUEUE_VC_INTERVAL_MS: 1000
      TB_QUEUE_VC_PARTITIONS: 1      
    depends_on:
      - postgres

volumes:
  postgres-data:
    name: tb-postgres-data
    driver: local

You can update default Rule Engine queues configuration using UI. More about ThingsBoard Rule Engine queues see in documentation.

Where:

  • 8080:8080 - connect local port 8080 to exposed internal HTTP port 8080
  • 1883:1883 - connect local port 1883 to exposed internal MQTT port 1883
  • 8883:8883 - connect local port 8883 to exposed internal MQTT over SSL port 8883
  • 7070:7070 - connect local port 7070 to exposed internal Edge RPC port 7070
  • 5683-5688:5683-5688/udp - connect local UDP ports 5683-5688 to exposed internal COAP and LwM2M ports
  • tb-postgres-data - name of the docker volume that stores the PostgreSQL’s data
  • thingsboard-ce - friendly local name of the ThingsBoard container
  • restart: always - automatically start ThingsBoard in case of system reboot and restart in case of failure.
  • image: "thingsboard/tb-node:4.0.1.1" - ThingsBoard docker image and version.

Initialize database schema & system assets

Before you start ThingsBoard, initialize the database schema and load built-in assets by running:

1

docker compose run --rm -e INSTALL_TB=true -e LOAD_DEMO=true thingsboard-ce

Environment variables:

  • INSTALL_TB=true - Installs the core database schema and system resources (widgets, images, rule chains, etc.).
  • LOAD_DEMO=true - Loads sample tenant account, dashboards and devices for evaluation and testing.

Start the platform & tail logs

Bring up all containers in detached mode, then follow the ThingsBoard logs:

1

docker compose up -d && docker compose logs -f thingsboard-ce

After executing this command you can open http://{your-host-ip}:8080 in you browser (for ex. http://localhost:8080). You should see ThingsBoard login page. Use the following default credentials:

You can always change passwords for each account in account profile page.

You can safely detach from the log stream (e.g. Ctrl+C); containers will continue running.

Inspect logs & control container lifecycle

If something goes wrong, you can stream the ThingsBoard container logs in real time:

1

docker compose logs -f thingsboard-ce

Bring down every container defined in your Compose file:

1

docker compose down

Launch all services in detached mode:

1

docker compose up -d

Upgrading

When a new CE release is available, follow these steps to update your installation without losing data:

  1. Change the version of the thingsboard/tb-node in the docker-compose.yml file to the new version (e.g. 4.0.1)

  2. Execute the following commands:

1
2
3
4

docker pull thingsboard/tb-node:4.0.1
docker compose stop thingsboard-ce
docker compose run --rm -e UPGRADE_TB=true thingsboard-ce
docker compose up -d

Troubleshooting

DNS issues

NOTE If you observe errors related to DNS issues, for example

1

127.0.1.1:53: cannot unmarshal DNS message

You may configure your system to use Google public DNS servers. See corresponding Linux and Mac OS instructions.

Next steps