Here are the commands that I think Brubeck should have. Some work has been done and some of that will be carried over.

I think Brubeck environments should basically be virtual environments. These things are handy for managing lots of flexibility in environments, in addition to isolating our needs from the system.

$ bpm project create [-n project_name]

This will create a Project directory called brubeck_project, unless you provide a different name.

$ cd <project_name>
$ bpm env create 

In that directory will be settings, static, templates and a directory named after your project.

Two hidden directories are also used. One is .virtualenv where the python environment is stored. The other is .var, where we find the logs, sock and run.

$ cd <project_name>
$ source bin/bpmenv

At this point it is just virtualenv. Turn it off by typing deactivate. By default the virtualenv is installed in project_name/.virtualenv.

To start Brubeck, use the up command. I think this command should probably use procer to watch over the Python processes.

$ bpm up

To start n Brubeck instances, pass a number to up:

$ bpm up 4

This command will start any relevant web servers and probably also sit on top of supervisord or procer to keep the processes up until you tell them to stop.