rawmind/alpine-kafka repository overview

⁠alpine-kafka

This image is the kafka base. It comes from alpine-jvm8⁠.

⁠Build

docker build -t rawmind/alpine-kafka:<version> .

⁠Versions

• 1.1.0-1 (Dockerfile)⁠
• 1.0.0-3 (Dockerfile)⁠
• 0.11.0.1-0 (Dockerfile)⁠
• 0.11.0.0-3 (Dockerfile)⁠
• 0.10.2.1 (Dockerfile)⁠
• 0.10.2.0-2 (Dockerfile)⁠
• 0.10.0.1-1 (Dockerfile)⁠
• 0.10.0.0-7 (Dockerfile)⁠
• 0.9.0.1-2 (Dockerfile)⁠

⁠Configuration

This image runs kafka⁠ with monit. Kafka is started with user and group "kafka".

Besides, you can customize the configuration in several ways:

⁠Default Configuration

kafka is installed with the default configuration and some parameters can be overrided with env variables:

• KAFKA_HEAP_OPTS=${JVMFLAGS:-"-Xmx1G -Xms1G"} # Kafka memory value
• KAFKA_ADVERTISE_PORT=${KAFKA_ADVERTISE_PORT:-"9092"} # Port to advertise
• KAFKA_LISTENER=${KAFKA_LISTENER:-"PLAINTEXT://0.0.0.0:${KAFKA_ADVERTISE_PORT}"} # Listerner string
• KAFKA_DELETE_TOPICS=${KAFKA_DELETE_TOPICS:-"false"} # Enable kafka delete topics
• KAFKA_LOG_DIRS=${KAFKA_LOG_DIRS:-"${SERVICE_HOME}/logs"} # Log directories.
• KAFKA_LOG_RETENTION_HOURS=${KAFKA_LOG_RETENTION_HOURS:-"168"} # Log retention hours
• KAFKA_NUM_PARTITIONS=${KAFKA_NUM_PARTITIONS:-"1"} # Number of partitions
• KAFKA_ZK_HOST=${KAFKA_ZK_HOST:-"127.0.0.1"} # Zk host
• KAFKA_ZK_PORT=${KAFKA_ZK_PORT:-"2181"} # Zk port
• KAFKA_EXT_IP=${KAFKA_EXT_IP:-""} # Advertise external ip or name if value != ""
• KAFKA_AUTO_CREATE_TOPICS=${KAFKA_AUTO_CREATE_TOPICS:-"true"} # Auto create topics
• KAFKA_REPLICATION_FACTOR=${KAFKA_REPLICATION_FACTOR:-1} # Replication factor

⁠Custom Configuration

Kafka is installed under /opt/kafka and make use of /opt/kafka/config/server.properties.

You can edit this files in order customize configuration

You could also include FROM rawmind/alpine-kafka at the top of your Dockerfile, and add your custom config.

⁠Rancher

If you are running it in rancher, you could run rancher-kafka⁠ as a sidekick to get dynamic configuration.

⁠k8s

If you are running it in k8s, you could run k8s-kafka⁠ at the same pod to get dynamic configuration.

⁠Example

See rancher-example⁠, that run kafka in a rancher system with dynamic configuration.