Setup and Run Apache Kafka on Windows

To get started with Apache Kafka on Windows, complete the instructions given in this quickstart tutorial.

Note: Your computer must have Java 8+ installed to run Kafka.

Downloading Apache Kafka

Go to Apache Kafka official download page at https://kafka.apache.org/downloads and download the binary version 2.8.1.

To keep the length of the path short by not placing the Kafka file in folders with long name. Doing so will cause error with message "The input line is too long. The syntax of the command is incorrect" while running the Kafka server. Move the downloaded .tgz file to the root of C: drive and extract it there.

Change Default Configurations

1. Go to Apache Kafka extracted files and change the zookeeper data directory path value in config/zookeeper.properties file:
//from
dataDir=/tmp/zookeeper

//to
dataDir=D:/kafka/data/zookeeper
2. Next, open and update Apache Kafka log file path value in config/server.properties file as shown in the example below:
//from
log.dirs=/tmp/kafka-logs

//to
log.dirs=D:/kafka/data/kafka-logs

Creating Data folder for Zookeeper and Apache Kafka

Create kafka folder first and inside this kafka folder, create data folder. Now, inside the data folder, create zookeeper and kafka-logs folders.

Starting Zookeeper and Apache Kafka

1. Now, open a command prompt, change your directory to kafka_2.13-2.8.1\bin\windows and start Zookeeper by executing the zookeeper-server-start.bat command with config\zookeeper.properties file:
> cd kafka_2.13-2.8.1\bin\windows
> .\zookeeper-server-start.bat ..\..\config\zookeeper.properties

You need to make sure that Zookeeper started successfully.

2. Next, open a new command prompt and change the directory to kafka_2.13-2.8.1\bin\windows. Finally, start Apache Kafka by executing the kafka-server-start.bat command with config\server.properties file:
> cd kafka_2.13-2.8.1\bin\windows
> .\kafka-server-start.bat ..\..\config\server.properties

You also need to make sure that Apache Kafka started successfully.

Creating a Topic

A topic is required to store events. A topic in Kafka, is like a table in a database where data are stored. You can have multiple topics for different events. To create a new topic, run the following command by replacing my-topic-name with the name of your topic:

You can also view details of the new topic by running the following command:

Writing Events to the Topic

A Kafka client communicates with the Kafka broker to write events. After the events are written, the broker will store the events for as long as you want.

There are Kafka clients libraries, supported for different languages, using which we can send events to Kafka topics. An application that sends data to the Kafka topic is called a producer application.

You can also use the console Producer to write events to the topic. By default, each entered line is treated as a separated event:

The console Producer client can be stopped by pressing Crtl+C at any time.

Reading Events from the Topic

A Kafka client communicates with the Kafka broker to read events.

You can use Kafka clients libraries in your application to read events from Kafka topics. An application that reads data from the Kafka topics is called a consumer application.

You can also use console Consumer client to read events that you created. Following is the command to read events:

You can stop the Consumer client by pressing Ctrl+C at any time, followed by entering Y (Yes).

Stopping the Kafka Services

1. Stop the Kafka console Producer and Consumer clients.
2. Next, stop the Kafka broker by pressing Ctrl+C, followed by entering Y (Yes).
3. Lastly, stop the Kafka ZooKeeper by pressing Ctrl+C, followed by entering Y (Yes).

Import/Export Data as Streams of Events into Kafka

Sometimes, you may need to collect data from any existing relational databases or messaging system into Kafka. To achieve this, you can use Kafka Connect. Kafka connect is a tool that helps to stream data reliably and durably between Kafka and other external systems. Kafka Connect can collect data continuously from any external systems into Kafka and vice versa.