thinking about moving to Trac

Question

We're starting development and are at a point where we need a bug tracking site. Our code is in SVN and we're thinking of using Bugzilla, but setup is tricky. I'm considering going for Trac instead. From what I understand, Trac would need to be on the same machine as SVN.

I'm really having a hard time parsing through the Trac documentation. What kind of things should I look for when I'll be migrating?

Answer 1

You are correct that Trac must be running on the same machine as the svn repository; it requires local access to the repo.

To use Trac, you need to: install the Trac software, create a Trac environment, configure your webserver, configure Trac authentication and users, customize Trac as needed for your purposes.

To install Trac, you can use the egg, rpm, or your distro's packaging system, etc.

To create your Trac environment, you will run

trac /path/to/new/trac/environment initenv

and follow the prompts. Since you have an existing svn repo, you will specify the path to it at that prompt.

To configure your webserver, you'll need to decide among several options. I tend to use mod_python though others will direct you to mod_wsgi.

I would recommend installing the AccountManagerPlugin and using its "form based login" instead of http auth. Using it with its htpasswd backend works well; you'd point it to an htpasswd file you create with an admin user and a good password. Then you would grant that admin account full access to the Trac environment like this:

trac /path/to/new/trac/environment permission add admin TRAC_ADMIN

You should now be able to log into Trac with your web browser as your admin user and customize Trac as desired.

Also, #trac on freenode is a good resource for getting help.

_{Disclosure: I'm one of the Trac devs.}