Skip to content

Latest commit

 

History

History
55 lines (34 loc) · 1.65 KB

ADVANCED.md

File metadata and controls

55 lines (34 loc) · 1.65 KB
Raw

Additional virtual machine details

Useful Vagrant commands

  1. Run vagrant ssh to get into your VM.
  2. Run ```vagrant halt`` to stop your VM.
  3. Run vagrant destroy when you want to destroy and get rid of the VM, or vagrant destroy && vagrant up to destroy and bring up a new VM.

Getting around the virtual machine

SSH into your VM:

vagrant ssh

This will log you in as the vagrant user. For root access run the following:

sudo su -

Hadoop and friends are all installed under /usr/local.

Changing the virtual machine memory

Edit Vagrantfile and look for the line that contains --memory - modify the value that you see there.

Speeding up the virtual machine provisioning time

The first time you run vagrant up it will download Hadoop and Spark and put their tarballs into the resources directory. When you reprovision your virtual machine with the vagrant destroy && vagrant up command, the deployment scripts will be smart enough to pick up the previously downloaded tarballs.

The only exception is Java - by default Java is installed via yum, so every time you reprovision your virtual machine it will download Java. To speed this up you can manually download the Oracle JDK into the resources directory. By default the script expects the file to be:

jdk-8u51-linux-x64.gz

You can change the default by editing scripts/common.sh.

Vagrant boxes

A list of available Vagrant boxes is shown at http://www.vagrantbox.es.

Vagrant box location

The Vagrant box is downloaded to the ~/.vagrant.d/boxes directory. On Windows, this is C:/Users/{your-username}/.vagrant.d/boxes.