Storm Cluster

Description

Apache Storm is a free and open source distributed
realtime computation system. Storm makes it easy to reliably process unbounded
streams of data, doing for realtime processing what Hadoop did for batch
processing. Storm is simple, can be used with any programming language, and is
a lot of fun to use!

Requirements

Cookbooks and gems

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

Platforms

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

Note: it should work fine on Debian 8 but the official docker image does not
allow systemd to work easily, so it could not be tested.

Usage

Easy Setup

This cookbook creates the number of nodes specified by attribute
node['storm-platform']['size'] (3 by default). The first node to be created
will be setup with a Nimbus, a LogViewer and a UI. All other nodes will host a
Supervisor and a LogViewer. To get this cookbook to setup the nth node to hold
the Nimbus, Supervisor and LogViewer instead of the first one, change attribute
node['storm-platform']['nimbus_id'].

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

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

Test

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 4 suites:

• dnsdock-centos-7
• zookeeper-centos-7
• storm-nimbus-centos-7
• storm-supervisor-centos-7

Each corresponds to a different node in the cluster.

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

Local cluster

The cluster installed with kitchen converge is fully working and can thus be
used as a local cluster for developments tests.

One can access the cluster by adding the dnsdock used in the cluster as main
DNS resolver: add
docker inspect --format '{{.NetworkSettings.IPAddress}}' dnsdock-storm
in /etc/resolv.conf.

Attributes

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.

Recipes

default

Installs a three node cluster as follow :

• Nimbus + LogViewer + UI
• Supervisor + LogViewer
• Supervisor + LogViewer

install

Extract a Storm tarball as per attributes.

create-user

Create system user for Storm binaries to run with.

config

Apply settings from attribute configuration to Storm configuration.

systemd-services

Create Systemd service files and set them up.

Resources/Providers

None.

Changelog

Available in [CHANGELOG](CHANGELOG).

Contributing

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

License and Author

• Author:: Guillaume Alaux (guillaume.alaux@s4m.io)
• Author:: Samuel Bernard (samuel.bernard@s4m.io)

Copyright:: 2015, 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

    http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.