Skip to main content
Version: 8.1

Docker

This page guides you through Camunda Platform 8 Docker images and how to run the platform in a developer setup using Docker Compose.

Docker images

We provide Docker images via Dockerhub. All these images are publicly accessible (except for Web Modeler Beta).

info

The provided Docker images are supported for production usage only on Linux systems. Windows or macOS are only supported for development environments.

ComponentDocker imageLink to configuration options
Zeebecamunda/zeebe:latestEnvironment variables
Operatecamunda/operate:latestOperate configuration
Tasklistcamunda/tasklist:latestTasklist configuration
Identitycamunda/identity:latestConfiguration variables
Optimizecamunda/optimize:latestEnvironment variables
Connectorscamunda/connectors:latestConnectors configuration
Connectors Bundlecamunda/connectors-bundle:latestConnectors configuration

Zeebe is the only component that is often run on its own as a standalone component. In this scenario, it does not need anything else, so a simple docker run is sufficient:

docker run --name zeebe -p 26500-26502:26500-26502 camunda/zeebe:latest

This will give you a single broker node with the following ports exposed:

  • 26500: Gateway API (this is the port clients need to use)
  • 26501: Command API (internal, gateway-to-broker)
  • 26502: Internal API (internal, broker-to-broker)

Web Modeler

Beta offering

Web Modeler Self-Managed is currently offered as a beta release with limited availability for enterprise customers only. It is not recommended for production use. Special terms & conditions apply.

The Docker images for Web Modeler are not publicly accessible, but available to enterprise customers only from Camunda's private Docker registry.

Web Modeler ComponentDocker image
Backend (restapi)registry.camunda.cloud/web-modeler-ee/modeler-restapi:0.1.0-beta
Frontend (webapp)registry.camunda.cloud/web-modeler-ee/modeler-webapp:0.1.0-beta
WebSocket serverregistry.camunda.cloud/web-modeler-ee/modeler-websockets:0.1.0-beta

To pull the images you first need to log in using the credentials you received from Camunda:

$ docker login registry.camunda.cloud
Username: your_username
Password: ******
Login Succeeded

See the instructions below on how to use Web Modeler images with Docker Compose.

Docker Compose

A Docker Compose configuration to run Zeebe, Operate, Tasklist, Optimize, Identity, and Connectors Bundle is available in the camunda-platform repository. Follow the instructions in the README.

danger

While the Docker images themselves are supported for production usage, the Docker Compose files are designed to be used by developers to run an environment locally; they are not designed to be used in production. We recommend to use Kubernetes in production.

This Docker Compose configuration serves two purposes:

  1. It can be used to start up a development environment locally.
  2. It documents how the various components need to be wired together.
note

We recommend to use Helm + KIND instead of Docker Compose for local environments, as the Helm configurations are battle-tested and much closer to production systems.

Web Modeler

Beta offering

Web Modeler Self-Managed is currently offered as a beta release with limited availability for enterprise customers only. It is not recommended for production use. Special terms & conditions apply.

An additional Docker Compose configuration to run Web Modeler is also available in the camunda-platform repository. Follow the instructions in the README to utilize this configuration.

Configuration hints

Zeebe

Volumes

The default data volume is under /usr/local/zeebe/data. It contains all data which should be persisted.

Configuration

The Zeebe configuration is located at /usr/local/zeebe/config/application.yaml. The logging configuration is located at /usr/local/zeebe/config/log4j2.xml.

The configuration of the Docker image can also be changed using environment variables. The configuration template file also contains information on the environment variables to use for each configuration setting.

Commonly used environment variables:

  • ZEEBE_LOG_LEVEL - Sets the log level of the Zeebe Logger (default: info).
  • ZEEBE_BROKER_NETWORK_HOST - Sets the host address to bind to instead of the IP of the container.
  • ZEEBE_BROKER_CLUSTER_INITIALCONTACTPOINTS - Sets the contact points of other brokers in a cluster setup.

Optimize

Some configuration properties are optional and have default values. See a description of these properties and their default values in the table below:

NameDescriptionDefault value
SPRING_PROFILES_ACTIVEStarts Optimize in Self-Managed mode.
CAMUNDA_OPTIMIZE_IDENTITY_ISSUER_URLThe URL at which Identity can be accessed by Optimize.
CAMUNDA_OPTIMIZE_IDENTITY_ISSUER_BACKEND_URLThe URL at which the Identity auth provider can be accessed by Optimize. This should match the configured provider in Identity and is to be used for container to container communication.
CAMUNDA_OPTIMIZE_IDENTITY_CLIENTIDThe Client ID used to register Optimize with Identity.
CAMUNDA_OPTIMIZE_IDENTITY_CLIENTSECRETThe secret used when registering Optimize with Identity.
CAMUNDA_OPTIMIZE_IDENTITY_AUDIENCEThe audience used when registering Optimize with Identity.
OPTIMIZE_ELASTICSEARCH_HOSTThe address/hostname under which the Elasticsearch node is available.localhost
OPTIMIZE_ELASTICSEARCH_HTTP_PORTThe port number used by Elasticsearch to accept HTTP connections.9200
CAMUNDA_OPTIMIZE_SECURITY_AUTH_COOKIE_SAME_SITE_ENABLEDDetermines if same-site is enabled for Optimize cookies. This must be set to false.true
CAMUNDA_OPTIMIZE_ENTERPRISEThis should only be set to true if an Enterprise License has been acquired.true
CAMUNDA_OPTIMIZE_ZEEBE_ENABLEDEnables import of Zeebe data in Optimize.false
CAMUNDA_OPTIMIZE_ZEEBE_NAMEThe record prefix for exported Zeebe records.zeebe-record
CAMUNDA_OPTIMIZE_ZEEBE_PARTITION_COUNTThe number of partitions configured in Zeebe.1
CAMUNDA_OPTIMIZE_SHARING_ENABLEDEnable/disable the possibility to share reports and dashboards.true
CAMUNDA_OPTIMIZE_UI_LOGOUT_HIDDENDisables the logout button (logout is handled by Identity).true
SPRING_SECURITY_OAUTH2_RESOURCESERVER_JWT_JWK_SET_URIAuthentication for the Public REST API using a resource server to validate the JWT token. Complete URI to get public keys for JWT validation.null
OPTIMIZE_API_ACCESS_TOKENAuthentication for the Public REST API using a static shared token. Will be ignored if SPRING_SECURITY_OAUTH2_RESOURCESERVER_JWT_JWK_SET_URI is also set.null

Like for example this docker-compose configuration:

optimize:
container_name: optimize
image: camunda/optimize:latest
ports:
- 8090:8090
environment:
- SPRING_PROFILES_ACTIVE=ccsm
- CAMUNDA_OPTIMIZE_IDENTITY_ISSUER_URL=http://localhost:9090
- CAMUNDA_OPTIMIZE_IDENTITY_ISSUER_BACKEND_URL=http://keycloak:8080/auth/realms/camunda-platform
- CAMUNDA_OPTIMIZE_IDENTITY_CLIENTID=optimize
- CAMUNDA_OPTIMIZE_IDENTITY_CLIENTSECRET=secret
- CAMUNDA_OPTIMIZE_IDENTITY_AUDIENCE=optimize-api
- OPTIMIZE_ELASTICSEARCH_HOST=localhost
- OPTIMIZE_ELASTICSEARCH_HTTP_PORT=9200
- CAMUNDA_OPTIMIZE_SECURITY_AUTH_COOKIE_SAME_SITE_ENABLED=false
- CAMUNDA_OPTIMIZE_ENTERPRISE=false
- CAMUNDA_OPTIMIZE_ZEEBE_ENABLED=true
- CAMUNDA_OPTIMIZE_ZEEBE_NAME=zeebe-record
- CAMUNDA_OPTIMIZE_ZEEBE_PARTITION_COUNT=1
- CAMUNDA_OPTIMIZE_SHARING_ENABLED=true
- CAMUNDA_OPTIMIZE_UI_LOGOUT_HIDDEN=true
- SPRING_SECURITY_OAUTH2_RESOURCESERVER_JWT_JWK_SET_URI=https://weblogin.cloud.company.com/.well-known/jwks.json
- OPTIMIZE_API_ACCESS_TOKEN=secret

Self-Managed Optimize must be able to connect to Elasticsearch to write and read data. In addition, Optimize needs to connect to Identity for authentication purposes. Both of these requirements can be configured with the options described above.

Optimize must also be configured as a client in Identity, and users will only be granted access to Optimize if they have a role that has write:* permission for Optimize.

For Optimize to import Zeebe data, Optimize must also be configured to be aware of the record prefix used when the records are exported to Elasticsearch. This can also be configured per the example above.

Connectors

Use the provided Docker Compose files to execute all out-of-the-box Connectors automatically. This uses the Connectors Bundle Docker image.

Note that some out-of-the-box Connectors are licensed under the Camunda Platform Self-Managed Free Edition license. Find an overview in the Connectors Bundle project.

Refer to the Connector installation guide for details on how to provide the Connector templates for modeling.

Custom set of Connectors

To add custom Connectors, you can build on top of our Connectors base image that includes the pre-packaged runtime environment without any Connector. To use the image, at least one Connector must be added to the classpath. We recommend providing JARs with all dependencies bundled.

caution

As all Connectors share a single classpath, different versions of the same dependency can be available and cause conflicts. To prevent this, common dependencies like jackson can be shaded and relocated inside the Connector's JAR.

You can add a Connector JAR by extending the base image with a JAR from a public URL:

FROM camunda/connectors:0.3.0

ADD https://repo1.maven.org/maven2/io/camunda/connector/connector-http-json/0.11.0/connector-http-json-0.11.0-with-dependencies.jar /opt/app/

You can also add a Connector JAR using volumes:

docker run --rm --name=connectors -d -v $PWD/connector.jar:/opt/app/ camunda/connectors:0.3.0