Setting up test & development environment

• Developers contributing on the development of ODM's modules
• ODM Staff testing the actual development
• System administrators willing to understand better ODM'S architecture

• How to replicate the test and development environment using both Vagrant and Webfaction.

Vagrant is free and open-source software for creating and configuring virtual development environments. It can be seen as a wrapper around virtualization software such as VirtualBox, and around configuration management software such as Chef, Salt or Puppet.

Ansible is an open-source software platform for configuring and managing computers. It combines multi-node software deployment, ad hoc task execution, and configuration management. It manages nodes over SSH and requires Python (2.4 or later) to be installed on them. Modules work over JSON and standard output and can be written in any programming language. The system uses YAML to express reusable descriptions of systems.

ODM developers can choose Vagrant for managing Virtual Machines where testing and development tasks can be undertaken. For this, following software needs to be installed and configured:

• Vagrant
• Virtual Box

TBC

TBC

In order to start the development box, change directory to where the Vagrantfile is and type:

cd odm-deployment
vagrant up   

This will take some time because the virtual box will be provisioned with the instructions specified in the bootstrap.sh. This script basically install some needed dependencies such as git and ansible and runs the set of Ansible instructions available on deployment/ansible in order to provision the VM.

The provisioning process might take between 5 and 15 minutes, depending on the power of your machine and speed of your internet connection.

Aparently, the VM needs to be restarted once after the deployment is ready in order for the different services to start properly. Following command needs to be run:

vagrant reload

Sometimes, the deployment of the Development Environment will be stuck at some points that caused by wrong configuration or the speed of internet connection.

• Next vagrant destroy and up again.

Once the process is finished and no errors have occurred, it is time to access the machine via ssh. There are still some necessary steps to be done.

vagrant ssh  

Exit the ssh session and call vagrant halt, this way the box will be stopped, you can start it any time by calling vagrant up again.

exit
vagrant halt

In order to start the provisioning, we need to run the ansible-playbook command:

cd ansible
ansible-playbook site.yml -i stage 

The provisioning scripts installs CKAN and its dependencies plus some extensions, but there are still some actions to be taken in order to configure the development environment completely:

After activating the python environment with

. ~/.virtualenvs/ckan/bin/activate

, you can execute

paster --plugin=ckan sysadmin add USERNAME ~/etc/ckan/default/development.ini

OR

paster --plugin=ckan sysadmin add USERNAME -c ~/etc/ckan/default/development.ini

for creating a new user with sysadmin permissions, or attribute those permissions to an existing user. USERNAME is the account name, you can name whatever eg. admin or user1

Once this is done and password has been set, you can login into CKAN under http://192.168.33.10:8081/user/login.

Since the ckanext-odm_migration extension gets installed automatically while provisioning the vagrant machine. Once logged in, you can access http://192.168.33.10:8081/migration where you will find a series of utilities for inserting initial data and importing taxonomy terms and their translations and/or law records.

Sometimes it can happen that the during the “Update packages list” step of the common Ansible playbook, the following error is thrown, stoping the execution of the provisioning.

In order to solve this, a failsafe sources list can be selected. For that, it is just necessary to change the following line in the deployment/ansible/roles/common-ubuntu/tasks/main.yml file:

- name: add new sources.list that checks for mirror
  copy: src=sources.list dest=/etc/apt/sources.list owner=root group=root mode=0644
  sudo: yes

And replace src=sources.list with one of the available failsafe sources lists in deployment/ansible/roles/common/tasks/main.yml

If the machine is provisioned and started but CKAN and/or Tomcat are not accesible through their respective URLs, it can be that supervisor has not being able to start them for some reason. In order to manually start them:

1. Activate virtualenv:

   . ~/.virtualenvs/ckan/bin/activate

2. Run supervisorctl:

   supervisorctl

3. If following error message appears: unix:/home/vagrant/tmp/supervisor.sock no such file run: supervisord.