I have created a small render farm (a group of computers which can render 3D scenes/animations together and thus faster than an individual machine), consisting of my laptop and a desktop pc using Farmerjoe. More information: [link] .

Howto (for Ubuntu and Debian)

It happened by simply creating a shared folder on the host (laptop) which contained the content of the ZIP folder.

First I edited the config file, setting values that matched my system. I left the port through which 'slaves' connect the default value. I think it would be wise to unblock it if there is a firewall on the host. the IP adress of the host goes after 'master = '. I assume 'jobs = jobs' and 'logs = logs' means that the folder to store jobs in is called jobs and same with logs. The *_root is the shared folder's location, the *_blender is the location of blender, which I assume has to be the same on all slaves. *Note: It would be wise to put Blender in the shared directory as well, so all 'slaves' can access it. Lastly, *_composite; the path to imagemagick, which puts all parts together in one image (in case of bucket rendering). I left the default value for 'appserver_port' too. Would probably be nice to unblock this one as well in case of a firewall.

I then ran the program with the option --master on the host so I was able to submit jobs to it. I also ran the program with the option --appserver so that I got a web interface by connecting to the host. I furthermore ran (still the same) program twice without flags, because the laptop/host is a dual core machine. That means 4 instances of the same program on the host. On the desktop I ran the program without flags. That adds up to 3 'slave' processes.

I added the render script to the scripts folder in .blender, which enabled me to open the script in Blender and submit the job to get rendered using both computers.
*Note: to render something, the .blend has to be saved, otherwise it can't render using Farmerjoe.
Both frame and bucket rendering worked fine.
*Note: bucket rendering turns out a .tga image, even though you select it to be something different.

End of Howto (start reading again )

I recommend the readme file that was included instead of this howto, but this might be supportive for people (like me) who find the readme hard to follow. I managed to figure out this much which seems to make it pretty useful, but this is my second attempt at it, since the first (obviously) didn't get it up and running.

I can now add as many computers as I have to the network for rendering, although there is the limit of 4 wired computers on my router...