Name: single-node-hadoop
Owner: SpatialDev
Description: Pseudo-distributed Hadoop testing environment via Vagrant and Ansible
Forked from: AlanHohn/single-node-hadoop
Created: 2017-01-09 22:04:59.0
Updated: 2017-01-09 22:05:02.0
Pushed: 2016-03-05 23:35:43.0
Homepage: null
Size: 659
Language: Shell
GitHub Committers
| User | Most Recent Commit | # Commits |
|---|
Other Committers
| User | Most Recent Commit | # Commits |
|---|
This repository includes a Vagrant VM that uses Ansible to install and start Hadoop in psudo-distributed mode.
To get started, install Vagrant, Ansible, and VirtualBox.
Then run vagrant up from inside the repository. The
VM is configured to use an Ubuntu wily64 box. Hadoop will
be downloaded from a mirror and installed. At the moment
Hadoop 2.6.3 will be installed; you can configure the
version by setting hadoop_version before running the
first vagrant up or before running vagrant provision.
rt hadoop_version=2.7.1
After install, you can visit the [HDFS Name Node][nn] and [YARN Application Manager][yarn] pages to see the running services.
The wordcount directory has an example application from
the Hadoop docs. To use, first bring up the VM, and run
vagrant ssh to get a shell. Then:
vagrant/wordcount
ild.sh
.sh
n.sh
The application will show up in the application manager and will also print status and information on the console.