Skip to content

Getting Confluent Kafka components to work on Mac OS X

License

Notifications You must be signed in to change notification settings

santthosh/kafka-for-mac

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

9 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

kafka-for-mac

For those who use macOS and docker, it can be quite challenging to get several Confluent Kafka components up and running. Support for Docker on macOS / Windows is not yet available as per Confluent. This repo is an attempt to remove the uncertainty and let people experiment with Confluent Kafka faster. Here are some references for you to understand why this challenging

I can't connect to Kafka on OS X

Note: Important Caveat - Images Not Tested for Docker for Mac or Windows

Networking features in Docker for Mac

Pre-requisites

As of this writing, my local is running

Docker version 17.12.0-ce, build c97c6d6r

Getting Started

This docker-compose file has all the components necessary to bring up single node / single broker Kafka cluster,

  • Zookeeper
  • Kafka Broker
  • Schema Registry
  • Kafka REST Proxy

Bootstrap

docker-compose up -d

You should see some results like this

Creating kafkaformac_kafka_1           ... done
Creating kafkaformac_zookeeper_1       ...
Creating kafkaformac_schema-registry_1 ... done
Creating kafkaformac_schema-registry_1 ...
Creating kafkaformac_kafka-rest_1      ... done

Ensure the processes are running

docker-compose ps

You should see some results like this

            Name                         Command            State                           Ports
--------------------------------------------------------------------------------------------------------------------------
kafkaformac_kafka-rest_1        /etc/confluent/docker/run   Up      0.0.0.0:29080->29080/tcp, 8082/tcp
kafkaformac_kafka_1             /etc/confluent/docker/run   Up      0.0.0.0:29092->29092/tcp, 9092/tcp
kafkaformac_schema-registry_1   /etc/confluent/docker/run   Up      0.0.0.0:29081->29081/tcp, 8081/tcp
kafkaformac_zookeeper_1         /etc/confluent/docker/run   Up      2181/tcp, 2888/tcp, 0.0.0.0:32181->32181/tcp, 3888/tcp

Check the health of zookeeper by monitoring its logs through this command

docker-compose logs zookeeper | grep -i binding

Check if the kafka broker is healthy

docker-compose logs kafka | grep -i started

Similarly inspect logs for schema-registry and kafka-rest

docker-compose logs schema-registry

docker-compose logs kafka-rest

Consuming and Producing using the Kafka clients

  • Add a mapping with your machine ip to docker.for.mac.host.internal to your /etc/hosts file
    • e.g. 10.0.4.162 docker.for.mac.host.internal
    • find your machine ip with ifconfig | grep -e "inet "
    • (!) your machine ip changes if you change network, update the config and restart the container
  • Connect to the kafka cluster via localhost:29092 (bootstrap server)

Publishing and Subscribing using Kafka REST

confluentinc/kafka-rest has full documentation around this, but the validate.sh script allows you to run all the same quickly, go ahead and run it

./validate.sh

You should see the RESTful endpoints invoked in sequence for publishing and subscribing from your Kafka

# Get a list of topics
["_schemas","jsontest"]
# Produce a message with JSON data
{"offsets":[{"partition":0,"offset":6,"error_code":null,"error":null}],"key_schema_id":null,"value_schema_id":null}
# Get info about one topic
{"name":"jsontest","configs":{"file.delete.delay.ms":"60000","segment.ms":"604800000","min.compaction.lag.ms":"0","retention.bytes":"-1","segment.index.bytes":"10485760","cleanup.policy":"delete","follower.replication.throttled.replicas":"","message.timestamp.difference.max.ms":"9223372036854775807","segment.jitter.ms":"0","preallocate":"false","segment.bytes":"1073741824","message.timestamp.type":"CreateTime","message.format.version":"0.11.0-IV2","max.message.bytes":"1000012","unclean.leader.election.enable":"false","retention.ms":"604800000","flush.ms":"9223372036854775807","delete.retention.ms":"86400000","leader.replication.throttled.replicas":"","min.insync.replicas":"1","flush.messages":"9223372036854775807","compression.type":"producer","min.cleanable.dirty.ratio":"0.5","index.interval.bytes":"4096"},"partitions":[{"partition":0,"leader":1,"replicas":[{"broker":1,"leader":true,"in_sync":true}]}]}
# Create a consumer for JSON data, starting at the beginning of the topic's log. The consumer group is called "my_json_consumer" and the instance is "my_consumer_instance\  .
{"instance_id":"my_consumer_instance","base_uri":"http://kr-host:8082/consumers/my_json_consumer/instances/my_consumer_instance"}
# Subscribe the consumer to a topic
No content in response
# Then consume some data from a topic using the base URL in the first response.
[]
# Finally, close the consumer with a DELETE to make it leave the group and clean up its resources.
No content in response

Tear down

docker-compose down

Caveats

  • You really don't want bridge network on your docker compose, while its not well documented on why, its advised so. confluentinc/cp-docker-images

Contributing

Fork this repository, send in a pull request

Releases

No releases published

Packages

No packages published

Languages