removed matplotlib global Agg config. Updated vagrant virtual box settings with hostname and vm name. upated python example to set Agg

scripts/vagrant/zeppelin-dev/README.md

@@ -120,6 +120,7 @@ To Test plotting using matplotlib into a rendered %html SVG image, try
 %pyspark
 
 import matplotlib
+matplotlib.use('Agg')   # turn off interactive charting so this works for server side SVG rendering
 import matplotlib.pyplot as plt
 import numpy as np
 import StringIO

scripts/vagrant/zeppelin-dev/Vagrantfile

@@ -36,8 +36,11 @@ Vagrant.configure(VAGRANTFILE_API_VERSION) do |config|
   # within the machine from a port on the host machine.
 
   # Forward local host port 8080 to zeppelin port 8080
+  # Forward local host port 4040 to spark UI port 4040
   # comment out the following line out to use a fixed IP
   config.vm.network "forwarded_port", guest: 8080, host: 8080
+  config.vm.network "forwarded_port", guest: 4040, host: 4040
+
 
   # Uncomment to create a private network, which allows access to the machine
   # using a specific IP.  Port forwarding not required in this case.
@@ -45,10 +48,20 @@ Vagrant.configure(VAGRANTFILE_API_VERSION) do |config|
   # in your network environment.  192.168.x.x is an example
   #config.vm.network "private_network", ip: "192.168.51.52"
 
+  # Set the hostname.  Comment this out if you don't want the hostname managed by the VM
+  config.vm.hostname = "zeppelinvm"
 
- config.vm.provider "virtualbox" do |vb|
+  # Modify the number of CPUs and memory used by the VM
+  # by adjusting
+  #   --memory
+  # and uncommenting and/or updating
+  #   --cpus
+
+  config.vm.provider "virtualbox" do |vb|
      vb.customize ["modifyvm", :id, "--memory", "2048"]
- end
+     #vb.customize ["modifyvm", :id, "--cpus", "2"]
+     vb.name = "zeppelinvm"
+  end
 
   config.vm.provision "ansible" do |ansible|
      ansible.playbook = "ansible-roles.yml"

scripts/vagrant/zeppelin-dev/roles/python-addons/tasks/main.yml

@@ -26,10 +26,3 @@
     - python-scipy
     - python-numpy
     - python-pandas
-
-# Change python backend that pyspark can render 
-# matplotlib plots without an interactive window
-# Fully quoted because of the ': ' on the line.
-- name: use Agg as pythons default window instead of tkAgg
-  lineinfile: "dest=/etc/matplotlibrc regexp='^backend' line='backend : Agg'"
-