This is a portable DOPA testing and development environment. It consists of a set of configuration scripts that automate the creation of a virtual machine that runs Stratosphere.
The virtual machine makes it easy to learn about using, modify, and improving the DOPA platform software Stratosphere.
You'll need to install recent versions of Vagrant and VirtualBox.
Next, you'll need a copy of the vm that you can download from this GIT repository.
Download the content of this repository and extract it to a directory of your choice.
Once you have done that, open up a terminal or a command-prompt, and change your
working directory to the location of the extracted (or git-cloned) files.
From there, run vagrant up to provision and boot the virtual machine.
You'll now have to wait a bit (15-20 min), as Vagrant needs to retrieve the base image from Canonical, retrieve some additional packages, and installs and configures each of them.
If it all worked, you should be able to browse to http://localhost:8090/ and see the main page of your DOPA test instance.
You can close the vagrant session by the command 'vagrant halt'.
To access a command shell on your virtual environment, run vagrant ssh from
the root the directory you downloaded the virtual machine to.
From Windows this might cause problems see: http://stackoverflow.com/questions/9885108/ssh-to-vagrant-box-in-windows
Follow the installation instructions.
Run vagrant list-roles to see the available roles.
Use vagrant enable-role stratodev to enable the developer role and
vagrant provision to apply the changes.
Vagrant will download the essential source directories from the remote GIT repository.
ssh-keygen && cat ~/.ssh/id-rsa.pubgit remote add myrepo git@github.com:USERNAME/stratosphere-sopremo.gitgit push myrepo master:featurenameThere are two main entries for new users. If you are an experienced JAVA developer, are familar with the MapReduce concept and you want to model your data flow graphs by yourself you probably want to work on the PACT level. Otherwise, it might be easier to start with a simple meteor program.
To access the PACT web interface navigate to
http://localhost:8081/launch.html
You should see a page titled Stratosphere Query Interface. Now you can start with the hello world analogon for MapReduce WordCount. The wordcount example is preshipped with your Stratosphere installation. Upload
stratosphere\examples\pact\pact-examples-${Version}-WordCount.jar
from the your local download copy of the DOPA-VM to the web interface.
Check the checkbox in the right frame of the new word count example and specify e.g.
4 file:///dopa-vm/data/opendata/wikienmath.xml file:///dopa-vm/data/output/wordcount1.csv
as arguments. Click run job and click run on the next page again.
cd /dopa-vm
sudo ./stratosphere/bin/meteor-client.sh script.meteor --wait
this should copy test.json to test_result.json
If you want to use the features that are accessible by DOPA members only you have to copy your private key for the git repositories to dopa-vm/puppet/files/dopa.ppk. In unix this can be done with the following command
cp ~/.ssh/id_rsa dopa-vm/puppet/files/dopa.ppk && chmod 700 dopa-vm/puppet/files/dopa.ppk
The dopa-vm is equipped with a standalone cloundera (cdh4) installation.
Access
The HBASE support is limit at the moment and needs some manual tuning. See this changeset.
If you have problem to access the internet from inside the vm (you can test that via ping google.com ) run the following command to reset the virtual nic:
sudo ifdown eth0 && sudo ifup eth0
Bugs can be reported at