Adoptable Cookbooks List

Looking for a cookbook to adopt? You can now see a list of cookbooks available for adoption!
List of Adoptable Cookbooks

Supermarket Belongs to the Community

Supermarket belongs to the community. While Chef has the responsibility to keep it running and be stewards of its functionality, what it does and how it works is driven by the community. The chef/supermarket repository will continue to be where development of the Supermarket application takes place. Come be part of shaping the direction of Supermarket by opening issues and pull requests or by joining us on the Chef Mailing List.

Select Badges

Select Supported Platforms


confluent-platform (16) Versions 2.1.0

Install and configure Confluent platform (Kafka)

cookbook 'confluent-platform', '= 2.1.0'
cookbook 'confluent-platform', '= 2.1.0', :supermarket
knife supermarket install confluent-platform
knife supermarket download confluent-platform
Quality 44%

Confluent Platform


Apache Kafka, an open source technology created and maintained by the founders of Confluent, acts as a real-time, fault tolerant, highly scalable messaging system. It is widely adopted for use cases ranging from collecting user activity data, logs, application metrics, stock ticker data, and device instrumentation. Its key strength is its ability to make high volume data available as a real-time stream for consumption in systems with very different requirements from batch systems like Hadoop, to real-time systems that require low-latency access, to stream processing engines that transform the data streams as they arrive.

This infrastructure lets you build around a single central nervous system transmitting messages to all the different systems and applications within your company. Learn more on

This cookbook focuses on deploying Confluent Platform elements on your clusters via Chef on systemd managed distributions. At the moment, this includes Kafka, Schema Registry and Kafka Rest.


Cookbooks and gems

Declared in [metadata.rb](metadata.rb) and in [Gemfile](Gemfile).


A systemd managed distribution: - RHEL Family 7, tested on Centos

Note: it should work quite fine on Debian 8 (with some attributes tuning) but the official docker image does not allow systemd to work easily, so it could not be tested.


Easy Setup

Default recipe does nothing. Each service Kafka, Schema Registry or Kafka Rest will be installed by calling respectively recipe [install_kafka](recipes/install_kafka.rb), [install_registry](recipes/install_registry.rb) and [install_rest](recipes/install_rest.rb).

By default, this cookbook installs openjdk from the official repositories (openjdk 8 on centos 7) in services recipe, just before launching the service. You can deactivate this behavior by setting node['confluent-platform']['java'] to "", or choose your package by setting the package name in node['confluent-platform']['java'][node[:platform]].

The recommended way to use this cookbook is through the creation of a different role per cluster, that is a role for Kafka, Schema Registry and Kafka Rest. This enables the search by role feature, allowing a simple service discovery.

In fact, there are two ways to configure the search: 1. with a static configuration through a list of hostnames (attributes hosts like in ['confluent-platform']['kafka']['hosts']) 2. with a real search, performed on a role (attributes role and size like in ['confluent-platform']['kafka']['role']). The role should be in the run-list of all nodes of the cluster. The size is a safety and should be the number of nodes in the cluster.

If hosts is configured, role and size are ignored.

See [roles](test/integration/roles) for some examples and Cluster Search documentation for more information.

Zookeeper Cluster

To install properly a Kafka cluster, you need a Zookeeper cluster. This is not in the scope of this cookbook but if you need one, you should consider using Zookeeper Platform.

The configuration of Zookeeper hosts use search and is done similarly as for Kafka, Schema Registry and Kafka Rest hosts, ie with a static list of hostnames or by using a search on a role.


This cookbook is fully tested through the installation of the full platform in docker hosts. This uses kitchen, docker and some monkey-patching.

If you run kitchen list, you will see 6 suites: - zookeeper-centos-7 - kafka-01-centos-7 - kafka-02-centos-7 - kafka-03-centos-7 - registry-01-centos-7 - rest-01-centos-7

Each corresponds to a different node in the cluster. They are connected through a bridge network named kitchen, which is created if necessary.

For more information, see [.kitchen.yml](.kitchen.yml) and [test](test) directory.

Local cluster

Of course, the cluster you install by running kitchen converge is fully working so you can use it as a local cluster to test your development (like a new Kafka client). Moreover, compared to a single node cluster usually installed on workstations, you can detected partition/timing/fault-tolerance issues you could not because of the simplicity of a single-node system.

You can access it by using internal DNS of the docker network named kitchen or by declaring each node in your hosts file. You can get each IP by running:

docker inspect --format \
  '{{}}' container_name

Then to produce some messages: \
  --broker-list \
  --topic my_topic

And to read them: \
  --zookeeper \
  --topic my_topic \

Or you can use Rest API with and full Schema Registry support, located at


Configuration is done by overriding default attributes. All configuration keys have a default defined in [attributes/default.rb](attributes/default.rb). Please read it to have a comprehensive view of what and how you can configure this cookbook behavior.



Does nothing.


Configure confluent repository.


Install and fully configure a given service by running repository and its 4 dedicated recipes: package, user, config and service, in that order.


Install given service from confluent repository.


Create given service system user and group.


Generate service configuration. May search for dependencies (like Zookeeper or other nodes of the same cluster) with the help of cluster-search cookbook.


Install systemd unit for the given service, then enable and start it.

Note: install java package (OpenJDK 8 on centos 7) by default, can be disabled by setting node['confluent-platform']['java'] to "". A platform specific configuration for the package to install is also possible.






Please read carefully []( before making a merge request.

License and Author

Copyright (c) 2015-2016 Sam4Mobile

Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at

Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
See the License for the specific language governing permissions and
limitations under the License.

Dependent cookbooks

yum >= 0.0.0
cluster-search >= 0.0.0

Contingent cookbooks

There are no cookbooks that are contingent upon this one.




  • Default confluent version to install is set to 3.0
    • Scala version to install is set to 2.11
    • Mandatory option ssl.client.auth is added to registry config
  • Make Systemd unit path configurable


  • Start Continuous Integration with gitlab-ci
  • Add security opts for docker, add package retries
  • Remove sleep in recipes, wait to strengthen tests



  • Switch to confluent 2.0
  • Rename recipes to respect rubocop rules (breaking change)


  • Switch to docker_cli, use prepared docker image
    • Switch kitchen driver from docker to docker_cli
    • Use sbernard/centos-systemd-kitchen image instead of bare centos
    • Remove privileged mode :)
    • Remove some now useless monkey patching
    • Remove dnsdock, use docker DNS (docker >= 1.10)
    • Use "kitchen" network, create it if needed


  • Fix all rubocop offenses
  • Use specific name for resources to avoid cloning
  • Add more details on configuration in README



  • Clarify and fix JVM options for services
  • Use to_hash instead of dup to work on node values
  • Improve readibility of default system user names


  • Fix and clean the creation of Kafka work directories
  • Fix zookeeper.connect chroot path


  • Rationalize docker provision to limit images
  • Fix typo in roles/rest-kitchen.json name
  • Wait 15s after registry start to strengthen tests


  • Reorganize README:
    • Move changelog from README to CHANGELOG
    • Move contribution guide to
    • Reorder README, fix Gemfile missing
  • Add Apache 2 license file
  • Add missing chefignore
  • Fix long lines in rest and registry templates


  • Cleaning, use only dependencies from supermarket


  • Set java-1.8.0-openjdk-headless as default java package


  • Initial version with Centos 7 support

Collaborator Number Metric

2.1.0 failed this metric

Failure: Cookbook has 1 collaborators. A cookbook must have at least 2 collaborators to pass this metric.

Contributing File Metric

2.1.0 failed this metric

Failure: To pass this metric, your cookbook metadata must include a source url, the source url must be in the form of, and your repo must contain a file

Foodcritic Metric

2.1.0 failed this metric

FC066: Ensure chef_version is set in metadata: confluent-platform/metadata.rb:1
FC069: Ensure standardized license defined in metadata: confluent-platform/metadata.rb:1
Run with Foodcritic Version 10.3.1 with tags metadata,correctness ~FC031 ~FC045 and failure tags any

License Metric

2.1.0 passed this metric

No Binaries Metric

2.1.0 passed this metric

Publish Metric

2.1.0 passed this metric

Supported Platforms Metric

2.1.0 passed this metric

Testing File Metric

2.1.0 failed this metric

Failure: To pass this metric, your cookbook metadata must include a source url, the source url must be in the form of, and your repo must contain a file

Version Tag Metric

2.1.0 failed this metric

Failure: To pass this metric, your cookbook metadata must include a source url, the source url must be in the form of, and your repo must include a tag that matches this cookbook version number