I’ve read numerous posts about ignoring the Rails log directory but often they only mention ignoring the log contents of the directory. This is because if the log directory does not exist Rails will refuse to start until it is created. That may be a preferred way to work until one starts using Capistrano. Capistrano will create a symbolic link in place of the log directory and that will screw up subversion which will give an error like, “svn: Expected 'log' to be a directory but found a file.” So the solution to make Subversion work with Capistrano is to remove the log directory and ignore it and then create it when necessary after a checkout. Here are the specifics of how I did this for one project.

After installing a new heatsink in a machine I realized that although I had munin monitoring the machine I did not have the sensor data necessary to compare the performance on the new heatsink to the old one. So I decided to add that monitoring data as well as monitoring data for the hard drive temperature and SMART statistics. Here’s what I did on a 64-bit Ubuntu 7.10 (Gusty Gibbon) system.