in an effort to better understand the workings of, well… everything I’ve been trying to get ruby 1.8.6 going with rubygems 1.1.1 and rails 2.1.0.
First and foremost I noticed that rails 2.1.0 does not create .htaccess, so I used the one that 2.0.2 made. Not sure if that’s bad.
What I’ve set up in my .bashrc :
Readline was installed --prefix=$HOME, ruby was installed --prefix=$HOME --with-readline-dir=$HOME. Install went good. which ruby reports the one in home/bin and ruby -v is 1.8.6
rubygems was installed using ruby setup.rb --prefix=$HOME
once again which gem is the one in home/bin and gem-v is 1.1.1
I then used a local copy of all the gems required for rails (as well as fcgi and mysql, not sure if I need those?) and did gem install -l rails-2.1.0.gem (for example)
rails -v is 2.1.0 and which rails is the one in the appropriate spot (/home/myname/.gems/bin/rails).
In my error.log of my server dispatch.cgi is saying that the rails gem could not be found.
I also edited the shebang lines of my dispatch.cgi (and dispatch.fcgi when I switched .htaccess over - as well as dispatch.rb just for the sake of it) to !/home/myname/bin/ruby
Now that I have switched to fcgi I am getting two errors per each connection attempt to anything other than index.html (anything that sould invoke rails?) which are:
FastCGI: comm with (dynamic) server [my url] aborted: (first read) idle timeout (60 sec)
FastCGI: incomplete headers (0 bytes) received from server [my url]
Can anyone help shed some light on what I’ve done wrong or what I should be looking for?