GitLab – How to Block Crawlers Like Google from Indexing

I just tested this and it looks like the GitLab API response is using pagination. According to the documentation (http://api.gitlab.org), the default number of results per page is set to 20 and the starting page is 1.

To adjust the maximum results per page, you need to use the per_page variable in the HTTP request line. You can change the page number by using page as well, if you have more repositories than the maximum value of per_page. You can specify a maximum per_page value of 100.

For example, you request may look like:

https://git.example.com/api/v3/projects/all?page=1&per_page=100&private_token=abc123

The page and per_page variables are not required as they have default values, so you do not need to include either if you don't want to.

Hopefully this solves your problem.

Did you try the obvious naive solution?

server {

 [...ssl and servername stuff...]  

 location / {
    # fake the hostname to the hostname that gitlab expects
    #
    proxy_set_header Host hostname-for-gitlabhost;             
    proxy_pass          https://internal-gitlab-instance;
    proxy_read_timeout  90;
 }  
}

Additionally, you might set the location to something that allows only the

https://...../*.git

This should work:

Instead of

location /

location ~ ^/(.*\.git) {
   proxy_set_header Host hostname-for-gitlabhost;             
   proxy_pass          https://internal-gitlab-instance/$1;
   proxy_read_timeout  90;
}

This captures the request-uri and adds it to the proxy call withing the location statement.

I am not really sure if this works, just typed it from the top of my head.