Naive/clueless git question: I used "git clone --bare [local-repo-path] repo.git" to produce a bare repository folder and uploaded that to my static file server. Then I tried "git clone https://my.server/repo.git" and it fails ("fatal: repository 'https://my.server/repo.git' not found"). Visiting that same url in the browser produces 403 (there is no index.html in that folder of course) which at least tells me the folder is there. Is my approach too naive or is there some way to make this work? :D
@freebliss I am not 100% sure, but I don't think one can clone a git repository from a static HTTP hosting.
You can try out something like `GIT_CURL_VERBOSE=1 git clone --verbose https://github.com/torvalds/linux.git` to see the exact HTTP requests git makes and I think it first POSTs some data to the server and the git server then responds with the right data.
@lukas hey cool appreciate the input! i was missing an extra step to enable the "Dumb HTTP" protocol which allows this to work: "git update-server-info" inside repo.git (see other reply on my original post for detiled link) either way thanks for the pointer!
@freebliss you're missing one step - to make bare repositories HTTP accessible see the "Dumb HTTP" section here: https://git-scm.com/book/en/v2/Git-on-the-Server-The-Protocols
(Brought to you by collaborating this way decades ago 😭)
@eaon awesome, works like a charm now (after "cd repo.git && git update-server-info" and reuploading) - many thanks!
Welcome to post.lurk.org, an instance for discussions around cultural freedom, experimental, new media art, net and computational culture, and things like that.