From the source path, use the provided setup.py script.
python setup.py install
Or using the pip system:
pip install testfarm
If you want to deploy a server receiving client reports, documentation and configuration files for different deployment methods are available at the 'deploy' directory. The recommended setup is using mod_wsgi on Apache.
Either if you want to install a server or run a client locally
you would like to deploy a log file structure.
This is done with the testfarmserver script.
This will initialize the a log structure:
testfarmserver init ~/testfarmlogs
To create a new project:
testfarmserver addproject ~/testfarmlogs MyProject
# this sets the metadata
testfarmserver meta ~/testfarmlogs MyProject \
--set description 'My Project' \
--set othermeta 'other value' \
...
To allow clients to connect you have to add them first:
testfarmserver addclient ~/testfarmlogs MyProject MyClient
You can test it by doing a fake run. See the help for options.
testfarmserver fakerun ~/testfarmlogs MyProject MyClient
To generate the websites for all the projects:
testfarmserver generate ~/testfarmlogs ~/testfarmweb
- create a client key
- configure your client
- configure your task specific id
- recommend having a sandbox
- recommend having a dedicated user
- recommend using local installs
- launching by hand
- cron jobs