Home

Server Access

SSH in using a .pem file provided by a developer on the project.

To connect to the MetPetDB.com server SSH:

1. Move .pem file (metpetdb_security.pem) to your ssh keys directory.
   cp metpetdb_security.pem ~/.ssh/metpetdb_security.pem

2. SSH in.
   It's important that you sign into the server with the username for what you want to access. The main deployment is on the username metpet, the test deployment is on the username test, and the wiki is on the username wiki.
   ssh -i ~/.ssh/metpetdb_security.pem **username**@metpetdb.com

Deployment

Development/Local

1. Install a few required packages:
   sudo apt install python-dev libpq-dev libxml2-dev libproj-dev libgeos-dev libgdal-dev
2. Install pip.
3. Install virtualenvwrapper:
   sudo pip install virtualenvwrapper
4. Create a directory where you're going to store your virtual Python environments. We suggest ~/.virtualenvs, but use what you please.
5. Add the following two lines to your ~/.bashrc file (or .zshrc if you're using zsh instead of bash) and change the WORKON_HOME directory to the one you created:

export WORKON_HOME=$HOME/.virtualenvs
source /usr/local/bin/virtualenvwrapper_lazy.sh

6. Restart your terminal or run source ~/.bashrc or source ~/.zshrc.
7. Create a virtual Python environment for MetPetDB:
   mkvirtualenv metpetdb
   You should notice your terminal prompt now begins with (metpetdb).
8. Exit the virtual environment:
   deactivate.
9. Re-enter the virtual environment by running (this and the last step are just a sanity check):
   workon metpetdb
10. Create a project folder for MetPetDB, and inside this project folder clone the interface repository and place a copy of the app_variables.env file that you grab from a developer on the project.
11. Enter the cloned repository and install Python dependency modules for the project:
    pip install -r requirements_new_ubuntu_14.04.txt
12. If an error occurs when editing/adding samples/subsamples/chemical analyses, install the latest requests module:
    pip install requests
13. Start the interface:
    python app.py
14. Go to http://localhost:5000/ to make sure the interface is running. You're done!

Production Deployment & Redeployment

Section under construction, deployment uses nginx and uwsgi.

Contribution workflow

Developers for the project should create a feature branch for work; these changes should then be merged into the project's testing branch for review. Non-developers should fork the testing branch and submit a PR for merge consideration by project developers.