Clustered Deployment

In this section, we provide a tutorial for running a three-node Kafka cluster and ZooKeeper ensemble. By the end of this tutorial, you will have successfully installed and run a simple deployment with Docker.

Note

If you’re looking for a simpler tutorial, please refer to our quick start guide, which is limited to a single node Kafka cluster.

It is worth noting that we will be configuring Kafka and ZooKeeper to store data locally in the Docker containers. For production deployments (or generally whenever you care about not losing data), you should use mounted volumes for persisting data in the event that a container stops running or is restarted. This is important when running a system like Kafka on Docker, as it relies heavily on the filesystem for storing and caching messages. Refer to our documentation on Docker external volumes for an example of how to add mounted volumes to the host machine.

Installing & Running Docker

For this tutorial, we’ll run Docker using the Docker client. If you are interested in information on using Docker Compose to run the images, skip to the bottom of this guide.

To get started, you’ll need to first install Docker and get it running. The Confluent Platform Docker Images require Docker version 1.11 or greater.

Docker Client: Setting Up a Three Node Kafka Cluster

If you’re running on Windows or Mac OS X, you’ll need to use Docker Machine to start the Docker host. Docker runs natively on Linux, so the Docker host will be your local machine if you go that route. If you are running on Mac or Windows, be sure to allocate at least 4 GB of ram to the Docker Machine.

Now that we have all of the Docker dependencies installed, we can create a Docker machine and begin starting up Confluent Platform.

Note

In the following steps we’ll be running each Docker container in detached mode. However, we’ll also demonstrate how access the logs for a running container. If you prefer to run the containers in the foreground, you can do so by replacing the -d flags with --it.

1. Create and configure the Docker machine.

docker-machine create --driver virtualbox --virtualbox-memory 6000 confluent

Next, configure your terminal window to attach it to your new Docker Machine:

eval $(docker-machine env confluent)

2. Start Up a 3-node ZooKeeper Ensemble

docker run -d \
   --net=host \
   --name=zk-1 \
   -e ZOOKEEPER_SERVER_ID=1 \
   -e ZOOKEEPER_CLIENT_PORT=22181 \
   -e ZOOKEEPER_TICK_TIME=2000 \
   -e ZOOKEEPER_INIT_LIMIT=5 \
   -e ZOOKEEPER_SYNC_LIMIT=2 \
   -e ZOOKEEPER_SERVERS="localhost:22888:23888;localhost:32888:33888;localhost:42888:43888" \
   confluentinc/cp-zookeeper:4.0.0

docker run -d \
   --net=host \
   --name=zk-2 \
   -e ZOOKEEPER_SERVER_ID=2 \
   -e ZOOKEEPER_CLIENT_PORT=32181 \
   -e ZOOKEEPER_TICK_TIME=2000 \
   -e ZOOKEEPER_INIT_LIMIT=5 \
   -e ZOOKEEPER_SYNC_LIMIT=2 \
   -e ZOOKEEPER_SERVERS="localhost:22888:23888;localhost:32888:33888;localhost:42888:43888" \
   confluentinc/cp-zookeeper:4.0.0

docker run -d \
   --net=host \
   --name=zk-3 \
   -e ZOOKEEPER_SERVER_ID=3 \
   -e ZOOKEEPER_CLIENT_PORT=42181 \
   -e ZOOKEEPER_TICK_TIME=2000 \
   -e ZOOKEEPER_INIT_LIMIT=5 \
   -e ZOOKEEPER_SYNC_LIMIT=2 \
   -e ZOOKEEPER_SERVERS="localhost:22888:23888;localhost:32888:33888;localhost:42888:43888" \
   confluentinc/cp-zookeeper:4.0.0

Before moving on, we’ll check the logs to see the broker has booted up successfully by running the following command:

docker logs zk-1

You should see messages like this at the end of the log output:

[2016-07-24 07:17:50,960] INFO Created server with tickTime 2000 minSessionTimeout 4000 maxSessionTimeout 40000 datadir /var/lib/zookeeper/log/version-2 snapdir /var/lib/zookeeper/data/version-2 (org.apache.zookeeper.server.ZooKeeperServer)
[2016-07-24 07:17:50,961] INFO FOLLOWING - LEADER ELECTION TOOK - 21823 (org.apache.zookeeper.server.quorum.Learner)
[2016-07-24 07:17:50,983] INFO Getting a diff from the leader 0x0 (org.apache.zookeeper.server.quorum.Learner)
[2016-07-24 07:17:50,986] INFO Snapshotting: 0x0 to /var/lib/zookeeper/data/version-2/snapshot.0 (org.apache.zookeeper.server.persistence.FileTxnSnapLog)
[2016-07-24 07:17:52,803] INFO Received connection request /127.0.0.1:50056 (org.apache.zookeeper.server.quorum.QuorumCnxManager)
[2016-07-24 07:17:52,806] INFO Notification: 1 (message format version), 3 (n.leader), 0x0 (n.zxid), 0x1 (n.round), LOOKING (n.state), 3 (n.sid), 0x0 (n.peerEpoch) FOLLOWING (my state) (org.apache.zookeeper.server.quorum.FastLeaderElection)

You can repeat the command for the two other ZooKeeper nodes. Next, you should verify that ZK ensemble is ready:

for i in 22181 32181 42181; do
  docker run --net=host --rm confluentinc/cp-zookeeper:4.0.0 bash -c "echo stat | nc localhost $i | grep Mode"
done

You should see one leader and two follower nodes. The output should look something like the following:

Mode: follower
Mode: leader
Mode: follower

3. Now that ZooKeeper is up and running, we can fire up a three node Kafka cluster.

docker run -d \
    --net=host \
    --name=kafka-1 \
    -e KAFKA_ZOOKEEPER_CONNECT=localhost:22181,localhost:32181,localhost:42181 \
    -e KAFKA_ADVERTISED_LISTENERS=PLAINTEXT://localhost:29092 \
    confluentinc/cp-kafka:4.0.0

docker run -d \
    --net=host \
    --name=kafka-2 \
    -e KAFKA_ZOOKEEPER_CONNECT=localhost:22181,localhost:32181,localhost:42181 \
    -e KAFKA_ADVERTISED_LISTENERS=PLAINTEXT://localhost:39092 \
    confluentinc/cp-kafka:4.0.0

 docker run -d \
     --net=host \
     --name=kafka-3 \
     -e KAFKA_ZOOKEEPER_CONNECT=localhost:22181,localhost:32181,localhost:42181 \
     -e KAFKA_ADVERTISED_LISTENERS=PLAINTEXT://localhost:49092 \
     confluentinc/cp-kafka:4.0.0

Check the logs to see the broker has booted up successfully

docker logs kafka-1
docker logs kafka-2
docker logs kafka-3

You should see start see bootup messages. For example, docker logs kafka-3 | grep started will show the following:

[2016-07-24 07:29:20,258] INFO [Kafka Server 1003], started (kafka.server.KafkaServer)
[2016-07-24 07:29:20,258] INFO [Kafka Server 1003], started (kafka.server.KafkaServer)

You should see the messages like the following on the broker acting as controller.

[2016-07-24 07:29:20,283] TRACE Controller 1001 epoch 1 received response {error_code=0} for a request sent to broker localhost:29092 (id: 1001 rack: null) (state.change.logger)
[2016-07-24 07:29:20,283] TRACE Controller 1001 epoch 1 received response {error_code=0} for a request sent to broker localhost:29092 (id: 1001 rack: null) (state.change.logger)
[2016-07-24 07:29:20,286] INFO [Controller-1001-to-broker-1003-send-thread], Starting  (kafka.controller.RequestSendThread)
[2016-07-24 07:29:20,286] INFO [Controller-1001-to-broker-1003-send-thread], Starting  (kafka.controller.RequestSendThread)
[2016-07-24 07:29:20,286] INFO [Controller-1001-to-broker-1003-send-thread], Starting  (kafka.controller.RequestSendThread)
[2016-07-24 07:29:20,287] INFO [Controller-1001-to-broker-1003-send-thread], Controller 1001 connected to localhost:49092 (id: 1003 rack: null) for sending state change requests (kafka.controller.RequestSendThread)

4. Test that the broker is working as expected.

Now that the brokers are up, we’ll test that they’re working as expected by creating a topic.

docker run \
  --net=host \
  --rm \
  confluentinc/cp-kafka:4.0.0 \
  kafka-topics --create --topic bar --partitions 3 --replication-factor 3 --if-not-exists --zookeeper localhost:32181

You should see the following output:

Created topic "bar".

Now verify that the topic is created successfully by describing the topic.

docker run \
    --net=host \
    --rm \
    confluentinc/cp-kafka:4.0.0 \
    kafka-topics --describe --topic bar --zookeeper localhost:32181

You should see the following message in your terminal window:

Topic:bar   PartitionCount:3    ReplicationFactor:3 Configs:
Topic: bar  Partition: 0    Leader: 1003    Replicas: 1003,1002,1001    Isr: 1003,1002,1001
Topic: bar  Partition: 1    Leader: 1001    Replicas: 1001,1003,1002    Isr: 1001,1003,1002
Topic: bar  Partition: 2    Leader: 1002    Replicas: 1002,1001,1003    Isr: 1002,1001,1003

Next, we’ll try generating some data to the bar topic we just created.

docker run \
  --net=host \
  --rm confluentinc/cp-kafka:4.0.0 \
  bash -c "seq 42 | kafka-console-producer --broker-list localhost:29092 --topic bar && echo 'Produced 42 messages.'"

The command above will pass 42 integers using the Console Producer that is shipped with Kafka. As a result, you should see something like this in your terminal:

Produced 42 messages.

It looked like things were successfully written, but let’s try reading the messages back using the Console Consumer and make sure they’re all accounted for.

docker run \
 --net=host \
 --rm \
 confluentinc/cp-kafka:4.0.0 \
 kafka-console-consumer --bootstrap-server localhost:29092 --topic bar --new-consumer --from-beginning --max-messages 42

You should see the following (it might take some time for this command to return data. Kafka has to create the __consumers_offset topic behind the scenes when you consume data for the first time and this may take some time):

1
4
7
10
13
16
....
41
Processed a total of 42 messages

Docker Compose: Setting Up a Three Node Kafka Cluster

Before you get started, you will first need to install Docker and Docker Compose. Once you’ve done that, you can follow the steps below to start up the Confluent Platform services.

1. Clone the Confluent Platform Docker Images Github Repository.

git clone https://github.com/confluentinc/cp-docker-images

We have provided an example Docker Compose file that will start up ZooKeeper and Kafka. Navigate to cp-docker-images/examples/kafka-cluster, where it is located:

2. Start ZooKeeper and Kafka using Docker Compose up command.

   docker-compose up
   

   In another terminal window, go to the same directory (kafka-cluster). Before we move on, let’s make sure the services are up and running:

   docker-compose ps
   

   You should see the following:

          Name                       Command            State   Ports
   ----------------------------------------------------------------------
   kafkacluster_kafka-1_1       /etc/confluent/docker/run   Up
   kafkacluster_kafka-2_1       /etc/confluent/docker/run   Up
   kafkacluster_kafka-3_1       /etc/confluent/docker/run   Up
   kafkacluster_zookeeper-1_1   /etc/confluent/docker/run   Up
   kafkacluster_zookeeper-2_1   /etc/confluent/docker/run   Up
   kafkacluster_zookeeper-3_1   /etc/confluent/docker/run   Up
   

   Check the ZooKeeper logs to verify that ZooKeeper is healthy. For example, for service zookeeper-1:

   docker-compose logs zookeeper-1
   

   You should see messages like the following:

   zookeeper-1_1  | [2016-07-25 04:58:12,901] INFO Created server with tickTime 2000 minSessionTimeout 4000 maxSessionTimeout 40000 datadir /var/lib/zookeeper/log/version-2 snapdir /var/lib/zookeeper/data/version-2 (org.apache.zookeeper.server.ZooKeeperServer)
   zookeeper-1_1  | [2016-07-25 04:58:12,902] INFO FOLLOWING - LEADER ELECTION TOOK - 235 (org.apache.zookeeper.server.quorum.Learner)
   

   Verify that ZK ensemble is ready:

   for i in 22181 32181 42181; do
      docker run --net=host --rm confluentinc/cp-zookeeper:4.0.0 bash -c "echo stat | nc localhost $i | grep Mode"
   done
   

   You should see one leader and two follower

   Mode: follower
   Mode: leader
   Mode: follower
   

   Check the logs to see the broker has booted up successfully.

   docker-compose logs kafka-1
   docker-compose logs kafka-2
   docker-compose logs kafka-3
   

   You should see start see bootup messages. For example, docker-compose logs kafka-3 | grep started shows the following

   kafka-3_1      | [2016-07-25 04:58:15,189] INFO [Kafka Server 3], started (kafka.server.KafkaServer)
   kafka-3_1      | [2016-07-25 04:58:15,189] INFO [Kafka Server 3], started (kafka.server.KafkaServer)
   

   You should see the messages like the following on the broker acting as controller.

   kafka-3_1      | [2016-07-25 04:58:15,369] INFO [Controller-3-to-broker-2-send-thread], Controller 3 connected to localhost:29092 (id: 2 rack: null) for sending state change requests (kafka.controller.RequestSendThread)
   kafka-3_1      | [2016-07-25 04:58:15,369] INFO [Controller-3-to-broker-2-send-thread], Controller 3 connected to localhost:29092 (id: 2 rack: null) for sending state change requests (kafka.controller.RequestSendThread)
   kafka-3_1      | [2016-07-25 04:58:15,369] INFO [Controller-3-to-broker-1-send-thread], Controller 3 connected to localhost:19092 (id: 1 rack: null) for sending state change requests (kafka.controller.RequestSendThread)
   kafka-3_1      | [2016-07-25 04:58:15,369] INFO [Controller-3-to-broker-1-send-thread], Controller 3 connected to localhost:19092 (id: 1 rack: null) for sending state change requests (kafka.controller.RequestSendThread)
   kafka-3_1      | [2016-07-25 04:58:15,369] INFO [Controller-3-to-broker-1-send-thread], Controller 3 connected to localhost:19092 (id: 1 rack: null) for sending state change requests (kafka.controller.RequestSendThread)
   

Note

Tip: docker-compose log | grep controller makes it easy to grep through logs for all services.

3. Follow section 4 in the “Docker Client” section above to test that your brokers are functioning as expected.