/cassandra-docker

Cassandra Docker images published by DataStax.

Primary LanguageShellOtherNOASSERTION

Cassandra Docker

Allowing for multi-platform images for dse-server via GitHub actions.

./docker-build.sh

Contents

The DataStax base image now uses OpenJDK. Previously we were building with Oracle JDK. Starting with the below image versions and moving forward, prebuilt images on Docker Hub will include OpenJDK. If you would like to use OpenJDK with a version that was built with Oracle JDK we have built new images including OpenJDK with a tag of version-openjdk8

  • DSE 6.0.2 and 5.1.11
  • OpsCenter 6.5.1 and 6.1.7
  • Studio 6.0.1

Quick Reference

Where to get help:

DataStax Academy , DataStax Community, Github

Full documentation and advanced tutorials are located within DataStax Academy . Docker Compose examples for deploying DataStax Enterprise with Opscenter and Studio are available on our Github page

Featured Tutorial - DataStax Enterprise 6 Guided Tour

Where to file issues:

Ask questions and get help at the DataStax Community forum.

Maintained by

DataStax

What is DataStax Enterprise

Built on the best distribution of Apache Cassandra™, DataStax Enterprise is the always-on database designed to allow you to effortlessly build and scale your apps, integrating graph, search, analytics, administration, developer tooling, and monitoring into a single unified platform. We power your apps' real-time moments so you can create instant insights and powerful customer experiences.

Prerequisites

  • Basic understanding of Docker images and containers.

  • Docker installed on your local system, see Docker Installation Instructions.

  • When using Docker for Mac or Docker for Windows, the default resources allocated to the linux VM running docker are 2GB RAM and 2 CPU's. Make sure to adjust these resources to meet the resource requirements for the containers you will be running. More information can be found here on adjusting the resources allocated to docker.

Docker for mac

Docker for windows

Creating a DataStax Enterprise container

Use the options described in this section to create DataStax Enterprise server containers.

By default, the DSE server image runs in Cassandra-only mode.. To run with advanced DSE functionality, add the option that enables any combination of search, analytics, and graph to the end of the docker run command.

Option Description
-s Enables and starts DSE Search.
-k Enables and starts Analytics.
-g Enables and starts a DSE Graph.

Combine startup options to run more than one feature. For more examples, see Starting DataStax Enterprise as a stand-alone process .

Examples

Create a DSE database container

docker run -e DS_LICENSE=accept --name my-dse -d datastax/dse-server

Create a DSE container with Graph enabled

docker run -e DS_LICENSE=accept --name my-dse -d datastax/dse-server -g

Create a DSE container with Analytics (Spark) enabled

docker run -e DS_LICENSE=accept --name my-dse -d datastax/dse-server -k

Create a DSE container with Search enabled

docker run -e DS_LICENSE=accept --name my-dse -d datastax/dse-server -s

Create a DSE container with Search, Analytics, and Graph enabled

docker run -e DS_LICENSE=accept --name my-dse -d datastax/dse-server -s -k -g

Managing the configuration

Manage the DSE configuration using one of the following options:

  • DSE configuration volume For configuration management, we’re providing a simple mechanism to let you provide custom configuration file(s) without customizing the containers or over using host volumes. You can add any of the approved config files to a single mounted host volume and we’ll handle the hard work of mapping them within the container.

  • DSE environment variables that change the configuration at runtime.

    • Environment variables will trump any other settings meaning that if you set a custom snitch in cassandra.yaml and then a different one with an environment variable, the environment variable setting will be used.
    • DSE uses the default values defined for environment variables unless they are explicitly set at run time or unless overridden with the DSE_AUTO_CONF_OFF environment variable.

  • Docker file/directory volume mounts

  • Docker overlay file system

  • DSE uses the default values defined for the environment variables unless explicitly set at run time.

  • NOTE When using memory resource constraints, you must must set JVM heap size using the environment variable JVM_EXTRA_OPTS or custom cassandra-env.sh or DSE running inside the container due to java not honoring resource limits set for the container. Java utilizes the resources (memory and CPU) of the host. Otherwise DSE will set the heap to 1/4 of the physical ram of the docker host.

Using the DSE conf volume

To use this feature:

  1. Create a directory on your local host.
  2. Download and customize the configuration files you want to use from the config-templates page.
  3. Add the custom configuration files to the host directory you created.
  • The file name must match a corresponding configuration file in the image and include all the required values, for example cassandra.yaml, dse.yaml.
  1. Mount the exposed Volume /config to the local directory.
  2. Start the container. For example to start a database node:
docker run -e DS_LICENSE=accept --name my-dse  -v /dse/config:/config -d datastax/dse-server

Note When you make changes to or add config files to the /config volume, you will need to restart your container with docker restart container_name for DSE to pickup the changes. Restarting the container will restart DSE.

Using environment variables

Configure the DSE image by setting environment variables when the container is created using the docker run command -e flag.

Variable Setting Description
DS_LICENSE accept **
Required**. Set to accept to acknowledge that you agree with the terms of the DataStax license. To show the license, set the variable DS_LICENSE to the value accept. *
The image only starts if the variable set to
accept.*
LISTEN_ADDRESS *
IP_address* The IP address to listen for connections from other nodes. Defaults to the container's IP address.
BROADCAST_ADDRESS *
IP_address* The IP address to advertise to other nodes. Defaults to the same value as the LISTEN_ADDRESS.
NATIVE_TRANSPORT_ADDRESS *
IP_address* The IP address to listen for client/driver connections. Default: 0.0.0.0.
NATIVE_TRANSPORT_BROADCAST_ADDRESS *
IP_address* The IP address to advertise to clients/drivers. Defaults to the same value as the BROADCAST_ADDRESS.
SEEDS *
IP_address* The comma-delimited list of seed nodes for the cluster. Defaults to this node's BROADCAST_ADDRESS.
START_RPC true`false` Whether or not to start the Thrift RPC server. Will leave the default in the cassandra.yaml file if not set.
CLUSTER_NAME *
string* The name of the cluster. Default: Test Cluster.
NUM_TOKENS *
int* The number of tokens randomly assigned to the node. Default: not set .
DC *
string* Datacenter name. Default: Cassandra.
RACK *
string* Rack name. Default: rack1.
OPSCENTER_IP IP_address \ *
string* Address of OpsCenter instance to use for DSE management; it can be specified via linking the OpsCenter container using opscenter as the name.
JVM_EXTRA_OPTS *
string* Allows setting custom Heap using -Xmx and -Xms.
LANG *
string* Allows setting custom Locale
SNITCH *
string* This variable sets the snitch implementation this node will use. It will set the endpoint_snitch option of cassandra.yaml. Default: GossipingPropertyFileSnitch
DSE_AUTO_CONF_OFF *
string* Sometimes users want to set all variables in the config files. For these situations one must prevent default environment variables from overriding those values. This setting lets you provide a comma-separated list of filenames (options are cassandra.yaml and cassandra-rackdc.properties) that will not accept the Environmental variables or can be set to 'all' to disable default environment variables being set within either file.

Volumes and data

To persist data, pre-create directories on the local host and map the directory to the corresponding volume using the docker run -v flag.

NOTE: If the volumes are not mounted from the local host, all data is lost when the container is removed.

DSE images expose the following volumes.

  • For DataStax Enterprise Transactional, Search, Graph, and Analytics workloads:

    • /var/lib/cassandra: Data from Cassandra
    • /var/lib/spark: Data from DSE Analytics w/ Spark
    • /var/lib/dsefs: Data from DSEFS
    • /var/log/cassandra: Logs from Cassandra
    • /var/log/spark: Logs from Spark
    • /config: Directory to add custom config files for the container to pickup.

  • For OpsCenter: /var/lib/opscenter

  • For Studio: /var/lib/datastax-studio

docker run -v <local_directory>:<container_volume>

See Docker docs > Use volumes for more information.

Running DSE commands and viewing logs

Use the docker exec -it <container_name> command to specific commands.

docker exec -it my-dse nodetool status

Opening an interactive bash shell

If the container is running in the background (using the -d), use the following command to open an interactive bash shell to run DSE commands.

docker exec -it <container_name> bash

To exit the shell without stopping the container type exit.

Opening an interactive CQL shell (cqlsh)

Use the following command to open cqlsh.

docker exec -it <container_name> cqlsh

Viewing logs

You can view the DSE logs using the Docker log command. For example:

docker logs my-dse

Creating an Opscenter Container

Follow these steps to create an Opscenter container and a connected DataStax Enterprise server container on the same Docker host.

To create and connect the containers:

  1. First create an OpsCenter container.
  • docker run -e DS_LICENSE=accept -d -p 8888:8888 --name my-opscenter datastax/dse-opscenter
  • See OpsCenter Docker run options for additional options that persist data or manage configuration.
  1. Create a DataStax Enterprise (DSE) server container that is linked to the OpsCenter container.
  • docker run -e DS_LICENSE=accept --link my-opscenter:opscenter --name my-dse -d datastax/dse-server
  1. Get the DSE container IP address:
  • On the host running the DSE container run
  • docker inspect my-dse | grep '"IPAddress":'
  1. Open a browser and go to http://DOCKER_HOST_IP:8888.
  2. Click Manage existing cluster.
  3. In host name, enter the DSE IP address.
  4. Click Install agents manually. Note that the agent is already installed on the DSE image; no installation is required.

OpsCenter is ready to use with DSE. See the OpsCenter User Guide for detailed usage and configuration instructions.

Creating a Studio Container

Follow these steps to create a DataStax Studio container that is connected to a DataStax Enterprise (DSE) server container on the same Docker host.

To create and connect the containers:

  1. Create a DataStax Studio container: 
    docker run -e DS_LICENSE=accept --link my-dse --name my-studio -p 9091:9091 -d datastax/dse-studio
  2. Open a browser and go to http://DOCKER_HOST_IP:9091
  3. Create the new connection using my-dse as the hostname, see DataStax Studio User Guide > Creating a new connection for further instructions.

Studio is ready to use with DSE. See DataStax Studio User Guide for detailed usage and configuration instructions.

Building

The code in this repository builds the DSE, Opscenter, DDAC and Studio Docker images. To get started, clone this repo and modify for your requirements.

This repo uses Gradle to build the images.

Requirements:

  • JDK8
  • Docker engine locally or point DOCKER_HOST env variable to the Docker engine you would like to build against (version 17.05+)

By default, Gradle will download DataStax tarballs from DataStax Academy.

DataStax uses a common base image for all products. If you would like to customize the OS, install additional packages etc, you would modify the base/Dockerfile.

In order to support multiple product versions, but without duplicating files, Docker build contexts are generated out of source folders that might contain FreeMarker templates (the files with .ftl extensions). The following conventions are important to remember:

  • Docker build context for each product is generated out of folder having the product name (e.g. server, opscenter, or studio)
  • The files NOT ending with .ftl extension are copied to the build context AS IS
  • The files ending with .ftl are processed as FreeMarker templates
    • Template directives are written using angle bracket syntax
    • Square bracket syntax is used for interpolations
    • Resulting content is copied to the build context under the filename with .ftl extension stripped out (e.g. Dockerfile.ftl becomes Dockerfile)
  • Each FreeMarker template has access to version variable
    • The following properties are available
      • version.major returns product version major number
      • version.minor returns product version minor number
      • version.bugfix returns product version bugfix number
    • The following functions are available
      • version.lowerThan('x.y.z') return true if version is semantically lower than x.y.z
      • version.greaterEqualThan('x.y.z') return true if version is semantically greater or equal than x.y.z

If you would like to customize DSE, Opscenter, Studio or DDAC you would need to modify templates in their corresponding folder.

Building a product image for a given version requires invoking a Gradle task that follows the pattern:

./gradlew build<Product><Version>Image

For example: if you want to build DSE 5.1.10, you need to say

./gradlew buildServer5.1.0Image

on the command line.

You can build many images at once:

./gradlew buildServer5.1.0Image buildServer6.7.0Image buildOpscenter6.7.0Image

Publishing an image to a registry (default: Docker Hub) requires invoking a Gradle task that follows the pattern:

./gradlew push<Product><Version>Image

Pushing to a private registry requires providing the registry url and the credentials either through project properties on the command line or via gradle.properties file:

dockerRegistry=https://your.private.registry.org
registryUsername=<USERNAME>
registryPassword=<PASSWORD>

Run ./gradlew tasks to get the list of all available tasks.

Next Steps

Head over to DataStax Academy for advanced documentation including

  • Apache Cassandra™/Datastax configuration management
  • Using environment variables
  • Persisting data
  • Exposing public ports
  • Volumes and data directories
  • Docker Compose examples to spin up connected clusters of DataStax Enterprise, Studio, and Opscenter (also on github)
  • Step-by-step tutorials and examples
  • How to build applications using Apache Cassandra™/ DataStax

Known Limitations

  • CFS is not supported.
  • LCM is not supported.
  • Changing any file not included in the list of approved configuration files will require an additional host volume or customization of the image. An example is SSL key management.
  • The JVM heap size must be set for DataStax Enterprise (DSE) running inside the container using the JVM_EXTRA_OPTS variable or custom cassandra-env.sh. If not set, Java does not honor resource limits set for the container, and will peer through the container to use resources (memory and CPU) of the host. See the JVM_EXTRA_OPTS variable in Using environment variables for more information.

License