(This is Sputnik the easy way. For other ways to install Sputnik ,see Custom Installation).
Requirements
Sputnik can be easily installed on UNIX, including Linux and OSX. You don't need root access to install it. It should also be possible to install it on Windows.
For UNIX installation, you will need wget and make for the installation. If you don't have them:
- On Ubuntu run "sudo apt-get install build-essential"
- On OSX, install XCode, then install wget using fink (get fink, install it, then run "fink install wget").
Installation
First, create a directory where Sputnik will live and cd into it. For instance:
mkdir ~/sputnik
cd ~/sputnik
Second, install Kepler into it. On UNIX, install Kepler using the following commands:
wget http://luaforge.net/frs/download.php/3468/kepler-install-1.1-1
bash kepler-install-1.1-1 --prefix=${PWD} --without-readline
(For Windows, see Kepler's installation page.)
Third, install Sputnik:
./bin/luarocks --only-from=http://sputnik.freewisdom.org/rocks/earth install sputnik 8.08.13
./bin/lua -lluarocks.require -e 'require("sputnik").setup()'
(On Windows, adjust the paths to luarocks and lua.)
Finally, create a directory where Sputnik will be storing the data:
mkdir wiki-data && chmod -R a+rw wiki-data
(Or the equivalent for Windows - make sure that the directory is writable by the server.) Note that storing data in files is just the default, but you can go with other options.
Running Sputnik with Xavante
You can now run Sputnik with Xavante, a webserver that comes with Kepler:
./bin/xavante_start
This will start Xavante on port 8080 by default. You can access Sputnik at http://localhost:8080/sputnik.ws.
If you are curious as to what is happening here, Xavante looks for "sputnik.ws" in ~/sputnik/kepler/htdocs directory.
sputnik.ws is a WSAPI application that looks like this:
require('sputnik')
return sputnik.new_wsapi_run_fn{
VERSIUM_PARAMS = { '/tmp/sputnik3/wiki-data/' },
BASE_URL = '/sputnik.ws',
}
Note that this application just calls Sputnik setting two parameters: VERSIUM_PARAMS (which tells Sputnik where to store data) and BASE_URL (which tells Sputnik what its URL is).
Running Sputnik with CGI
You can also run Sputnik through CGI with a web server like Apache. The installation
process already created a sputnik.cgi file for you, so just copy it wherever your server keeps the CGI files. For instance:
cd ..
sudo cp sputnik.cgi /usr/lib/cgi-bin/
If your cgi files are available via http://localhost/cgi-bin/ then you are done: just go to http://localhost/cgi-bin/sputnik.cgi.
If not, you'll need to edit sputnik.cgi, changing /cgi-bin/sputnik.cgi to whatever it needs to be changed to.
Note: If you do not have your server configured to run CGI, please read the documentation for your server. Unfortunately, configuring CGI varies from one server to another, and even for the same server between different distributions of Linux. We can't help you here.
Running Sputnik with FastCGI
To use FastCGI we need to install an extra rock:
./bin/luarocks install wsapi-fcgi
Then copy sputnik.cgi to your CGI directory as "sputnik.fcgi" and change ".cgi" to ".fcgi" everywhere in that file.
After the Installation
After the installation, register an account "Admin". Note that some of Sputnik's pages, such as "sputnik/navigation" are configured (by default) to only be editable by "Admin".
After that, go on to Basic Configuration.