In one of my Rails projects I’m using Sphinx to provide full-text search capabilities. To integrate both worlds I chose Thinking Sphinx, which is just great and so far has met all of my expectations.
Also, as I previously mentioned, I’m using RVM to manage my ruby installations on both my development and production machines and this setup is what motivated this post.
I use Monit to monitor the services running on my production server - nginx, mysql, php - and as of the first deploy of this application, it only made sense to also monitor Sphinx.
In order to create an initialization script, I would need at least a way to start and stop sphinx from the command line, which, using Thinking Sphinx, can be done using these rake tasks:
$ rake thinking_sphinx:stop
$ rake thinking_sphinx:start
It’s worth mentioning now that I don’t run sphinx as root. I run it with the same user my rails application uses. For the purpose of this post, let’s call it deploy.
When I tried using my script I got errors such as these:
Missing the Rails 2.3.5 gem. Please `gem install -v=2.3.5 rails`, update your RAILS_GEM_VERSION setting in config/environment.rb for the Rails version you do have installed, or comment out RAILS_GEM_VERSION to use the latest version installed.
After some digging around I found that the GEM_HOME environment variable for my deploy user wasn’t being set correctly - something to do with rvm, but not quite sure - and to fix this, well, I just had to set it, leaving the final working version of my script somewhat like this - simplified :
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 | |
After that, monit was able not only to monitor my sphinx instance, but also (re)start/stop it with the correct user.
I’m no Linux wizard so if you wanna suggest improvements to this script, feel free to do so!