How to set up Cassandra test cluster using Docker

In this post I’ll walk you through set up of the test Apache Cassandra cluster in a Docker environment. This is useful if you want to experiment with this NoSql database or need it as part of your development environment.

First of all, you need to have Docker installed on your system.

If you use it from Mac OS X, like me, then please make sure your VirtualBox VM instance has enough memory for running Cassandra cluster.

  1. Open “VirtualBox” app
  2. Find the VM, in my case it was “default”
  3. Open “Settings”, then “System”
  4. Make sure it has at least 8GB available

Now we are ready to begin the fun part.

Single instance

Let’s run single Cassandra instance and then check that it is up using docker ps command:

$ docker run --name cassandra1 -m 2g -d cassandra:3.0.4 

$ docker ps
CONTAINER ID        IMAGE               COMMAND                  CREATED             STATUS              PORTS                                         NAMES
c8d489f00f17        cassandra:3.0.4     "/docker-entrypoint.s"   10 seconds ago      Up 10 seconds       7000-7001/tcp, 7199/tcp, 9042/tcp, 9160/tcp   cassandra1

Note that there is -m 2g option for docker run command, it gives 2GB RAM to this container. If you are curious run docker logs cassandra1 to see log message produced during Cassandra start up.

Connection to Cassandra instance with cqlsh is not difficult. Just need find out current container IP address.

$ docker inspect --format='{{ .NetworkSettings.IPAddress }}' cassandra1

$ docker run -it --link cassandra1 --rm cassandra:3.0.4 \
  sh -c 'exec cqlsh'
Connected to Test Cluster at
[cqlsh 5.0.1 | Cassandra 3.0.4 | CQL spec 3.4.0 | Native protocol v4]
Use HELP for help.


To form the cluster we need to run second Cassandra container cassandra2 and connect it to cassandra1.

$ docker run --name cassandra2 -m 2g -d \
  -e CASSANDRA_SEEDS="$(docker inspect --format='{{ .NetworkSettings.IPAddress }}' cassandra1)" \ 

Ok, now we have a cluster with two instances. Let’s verify that using nodetool status command which need to be executed inside the container.

$ docker exec -i -t cassandra1 sh -c 'nodetool status'
Datacenter: datacenter1
|/ State=Normal/Leaving/Joining/Moving
--  Address     Load       Tokens       Owns (effective)  Host ID                               Rack
UN  101.84 KB  256          100.0%            fb3d0aef-d70e-4637-8bdf-7e861acfcea6  rack1
UN  107.36 KB  256          100.0%            ccd1000e-1b9a-422a-a2b4-d291da9c0585  rack1

More instances can be added to this cluster using the same approach. Before running other instances make sure that “nodetool” returns “Status=Up” for all existing cluster nodes.

I hope it was easy!