Steps to install and configure classic ONA in a virtual environment
Clone the ONA repo and rename it
cd /opt
git clone https://github.com/onaio/onadata.git
mv onadata onadata-venv
cd onadata-venv
Checkout the classic ONA commit
git checkout 394f06e483c88cdfc7aab31673310e99742c5c3e
git checkout -b classic_ona
From here the configuration is more or less the steps as described in https://github.com/onaio/onadata/blob/master/install.md with a few modifications
Become root
sudo su -
cd /opt/onadata-venv
./script/install/ubuntu
Modify the username and password accordingly
su postgres -c "psql -c \"CREATE USER onadata_venv WITH PASSWORD 'onadata_venv';\""
su postgres -c "psql -c \"CREATE DATABASE onadata_venv OWNER onadata_venv;\""
su postgres -c "psql -d onadata_venv -c \"CREATE EXTENSION IF NOT EXISTS postgis;\""
su postgres -c "psql -d onadata_venv -c \"CREATE EXTENSION IF NOT EXISTS postgis;\""
su postgres -c "psql -d onadata_venv -c \"CREATE EXTENSION IF NOT EXISTS postgis_topology;\""
virtualenv .
source ./bin/activate
Copy the default settings and edit the file appropriately with the credentials used above
cp onadata/settings/default_settings.py onadata/settings/local_settings.py
syncdb option has been deprecated, so in the file Makefile change the line
python2 manage.py syncdb --noinput
to
python2 manage.py migrate --run-syncdb
Install the requirements and ran make
bin/pip install -r requirements/base.pip
make
You may at this point start core with bin/python manage.py runserver --nothreading or continue with the setup.
cd docs
make html
cd ..
Copy static files to static dir
bin/python manage.py collectstatic --noinput
Create the super user that will be used from the UI
bin/python manage.py createsuperuser
The configuration file is missing in the git repo. Download a sample and edit it appropriately
bin/pip install uwsgi
wget https://gist.githubusercontent.com/soloincc/43cd165b5c835eb6d6b7132e8ec933b4/raw/45de558b28b144f251491d1e3b785bcb2c2f1699/onadata.conf
# Edit the config file before moving it
mv onadata.conf /etc/init/onadata-venv.conf
# Start the service
start onadata-venv
# Ensure it is running well
cat /var/log/onadata-venv.log
The config is missing from the git repo, download a sample and edit it appropriately
apt-get install rabbitmq-server
wget https://gist.githubusercontent.com/soloincc/6bfe3791f73afdccad8a1ddf07404729/raw/d0f78668cb83c2c957c015e89aaa1231dca44e29/celeryd-ona
# Edit the config file before moving it
mv celeryd-ona /etc/default/celeryd-ona-venv
cp /etc/default/celeryd-ona-venv /etc/init.d/
chmod +x /etc/init.d/celeryd-ona-venv
# Create the startup scripts
update-rc.d celeryd-ona-venv defaults
# Start the celery service
service celeryd-ona-venv start
The nginx config file is also missing, download it and edit it appropriately
apt-get install nginx
wget https://gist.githubusercontent.com/soloincc/ce791bc0f4bd4b1c8b59f6c321711f6e/raw/32febd14088284dd5bc3a94ebedfd572fad56070/nginx-onadata
# Edit the config file appropriately
mv nginx-onadata /etc/nginx/sites-available/onadata-venv
ln -s /etc/nginx/sites-available/onadata-venv /etc/nginx/sites-enabled/onadata-venv
# update and test /etc/nginx/sites-available/onadata-venv
service nginx configtest
# remove default nginx server config
sudo unlink /etc/nginx/sites-enabled/default
sudo service nginx restart
Thats it. Classic ONA should be installed and configured well. To start it ran bin/python manage.py runserver --nothreading. If running it from a local machine, go to http://127.0.0.1:8000 to access it. If running it from a remote server, ensure that nginx is configured well for proxy pass and your firewall is not blocking incoming connections.
In order to ran the server using a different port other than the default 8000, edit and make the necessary changes in /opt/onadata-venv/onadata/settings/common.py and then start the server bound at that point bin/python manage.py runserver 127.0.0.1:<port> --nothreading