Cloning Fossil with self-hosted cgi script yields “unknown repository”

Question

I've been trying for two weeks to set up fossil to work and if I did not see the advantage of having a repository I would have given up by now. I am really at my last wits end with this.

I have set up fossil on both my server and my computer. My server is linux with virtualmin taking care of most of the hosting. My computer is windows 7.

My fossil binary are both in my paths.

I chose to have fossil host itself using a self-hosted cgi script. To be really specific the exact same script as the one on this page under the title heading of: "Another solution to automatically serve multiple repositories" - http://fossil-scm.org/fossil/wiki?name=Cookbook

It seems to work, if I go to my server's ip to the /cgi-bin/p file I can see a list of my repositories which I've greated using:

  fossil init ider.fsl

I can see the wiki, and the general web gui of the fossil repository however....

From my windows machine when I try to clone the source by using the following:

  fossil clone  192.168.1.200:81/cgi-bin/p  ider

I only hear the constant echo of:

      unknown repository: 192.168.1.200:81/cgi-bin/p

Could it be the permissions set on the ider.fsl file? Obviously I am new at SCM but there is something very wrong since I can't find anything in the documentation or any references from google describing this problem, unless I am trying to find out if I want to clone a dinosaur... :/

Answer 1

Just guessing here, but from eyeballing the script, it looks like it requires the repo name to be in the CGI variable PATH_INFO, i.e. the bit of the URL after the script name.

If this is the case, you need to clone the repo using:

fossil clone http://192.168.1.200:81/cgi-bin/p/ider ider
#                                             ^^^^^

Fossil also has a pretty active mailing list that you should be able to get help from.