ghtorrent-vagrant

A Vagrant box with Puppet provisioning for testing GHTorrent

This will create a Debian 8 based virtual machine with all the required infrastructure (MongoDB, MySQL, RabbitMQ) to test GHTorrent and run it locally for smaller projects.

Installation

• Install VirtualBox.

• Install Vagrant

  • On Debian 8, you can use apt-get install vagrant
  • On the Mac, you can use brew cask install vagrant
  • For other platforms, you can try this link

• You need to install the Vagrant VirtualBox Additions plug-in: vagrant plugin install vagrant-vbguest

• Clone the repository (git clone https://github.com/ghtorrent/ghtorrent-vagrant.git)

• You will need a GitHub API key. You can create one using the process described here (Fair Use)

Running

• Switch to the checked out directory and do: vagrant up
• When the provisioning process finishes, do: vagrant ssh

This will get you to the command line of the the vagrant VM. You will notice that the file config.yaml exists. Open it and replace the token field (default value: abcd) with the GitHub API key you created before.

Retrieving info for a single project

To retrieve information for a single repository do

$ ght-retrieve-repo -c config.yaml owner repo

where owner and repo are the owner and repo of the GitHub repository you want to retrieve info for.

ght-retrieve-repo accepts lots of command line arguments to restrict the amount of data in downloads, which can be found using the --help switch.

Listening to GitHub's event stream

You need two steps:

• Run the event collector: ght-mirror-events -c config.yaml. This will query the event stream API end point and then will push events to RabbitMQ.

• Run the data collector: ght-data-retrieval -c config.yamlThis step will create the necessary queues on RabbitMQ and start consuming events from them, in parallel. You can see the contents of the queues by running sudo /usr/sbin/rabbitmqctl list_queues or by connecting your browser to port 15672 of the virtual machine public IP address (find it with /sbin/ifconfig eth1).

This step effectively replicates the main GHTorrent system, as run at ghtorrent.org.

Viewing the retrieval results

Data is being collected into two sources

MongoDB database

(no username/password)

$ mongo github
MongoDB shell version: 2.6.11
connecting to: github
> show collections
commit_comments
commits
events
followers
forks
issue_comments
issue_events
issues
org_members
pull_request_comments
pull_requests
repo_collaborators
repo_labels
repos
users
watchers
> db.commits.count()
43
> db.events.find({'actor.login':'gousiosg'}).count()
10

MySQL database

(username/password/database: ghtorrent)

[email protected]:~$ mysql -u ghtorrent -p'ghtorrent' ghtorrent
[...]
mysql> select count(*) from commits;
+----------+
| count(*) |
+----------+
|       10 |
+----------+
1 row in set (0.00 sec)

mysql> select * from projects;
+----+----------------------------------------------------------+----------+-------------------+----------------------------------------------------------------------+----------+---------------------+-------------+---------+
| id | url                                                      | owner_id | name              | description                                                          | language | created_at          | forked_from | deleted |
+----+----------------------------------------------------------+----------+-------------------+----------------------------------------------------------------------+----------+---------------------+-------------+---------+
|  1 | https://api.github.com/repos/gousiosg/github-mirror      |        1 | github-mirror     | Scripts to mirror Github in a cloudy fashion                         | Ruby     | 2011-11-26 13:02:32 |        NULL |       0 |
|  2 | https://api.github.com/repos/ghtorrent/ghtorrent-vagrant |        2 | ghtorrent-vagrant | A Vagrant box with Puppet provisioning for running GHTorrent locally | Puppet   | 2015-09-25 19:56:52 |        NULL |       0 |
+----+----------------------------------------------------------+----------+-------------------+----------------------------------------------------------------------+----------+---------------------+-------------+---------+
2 rows in set (0.00 sec)