Skip to content

jslusher/storm

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

13 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Description

Installs Twitter's Storm distributed computation platform. Includes recipes for installing both the Nimbus / Web UI component and the Supervisor component.

Requirements

  • Ubuntu 10.04 / 12.04

  • May function on other distributions, but has not been tested

  • java cookbook

  • runit cookbook

Attributes

Usage

This recipe relies on two setup components that need to be noted as they are not used in many (or any) community cookbooks.

Role Based Cluster Setup: This cookbook relies on a cluster identification role to allow more than one storm cluster to run in a single Chef environment, while not breaking Chef search. Create a role with a name of your choosing. The role may be left empty or you may use it to apply the your application's topology and all necessary JARs within your topology. You will need to specify the name of this role using the node attribute ['storm']['cluster_role'], which is empty by default. You will need to apply this cluster role to both supervisor and the nimbus/UI node in your cluster

Deploy Flag: This cookbook uses a deploy flag to prevent the application from deploying unless desired and allows for an undeploy recipe to run prior to the deploy. The deploy recipe will also cleanup the state of storm and is sufficient to wipe clean any topology deploy, although it does not stop the actual topology (that's in the works). Once you've applied the supervisor or nimbus recipes to a node you need to have "deploy_build=true" set in your shell. "sudo deploy_build=true chef-client" can be used to set the environment variable and run Chef in a single command.

About

Chef cookbook for storm

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Ruby 100.0%