10.6 Git Internals - Transfer Protocols

If you’re setting up a repository to be served read-only over HTTP, the dumb protocol is likely what will be used. This protocol is called “dumb” because it requires no Git-specific code on the server side during the transport process; the fetch process is a series of HTTP GET requests, where the client can assume the layout of the Git repository on the server.

Let’s follow the http-fetch process for the simplegit library:

The first thing this command does is pull down the info/refs file. This file is written by the update-server-info command, which is why you need to enable that as a post-receive hook in order for the HTTP transport to work properly:

Now you have a list of the remote references and SHA-1s. Next, you look for what the HEAD reference is so you know what to check out when you’re finished:

You need to check out the master branch when you’ve completed the process. At this point, you’re ready to start the walking process. Because your starting point is the ca82a6 commit object you saw in the info/refs file, you start by fetching that:

You get an object back – that object is in loose format on the server, and you fetched it over a static HTTP GET request. You can zlib-uncompress it, strip off the header, and look at the commit content:

Next, you have two more objects to retrieve – cfda3b, which is the tree of content that the commit we just retrieved points to; and 085bb3, which is the parent commit:

That gives you your next commit object. Grab the tree object:

Oops – it looks like that tree object isn’t in loose format on the server, so you get a 404 response back. There are a couple of reasons for this – the object could be in an alternate repository, or it could be in a packfile in this repository. Git checks for any listed alternates first:

If this comes back with a list of alternate URLs, Git checks for loose files and packfiles there – this is a nice mechanism for projects that are forks of one another to share objects on disk. However, because no alternates are listed in this case, your object must be in a packfile. To see what packfiles are available on this server, you need to get the objects/info/packs file, which contains a listing of them (also generated by update-server-info):

There is only one packfile on the server, so your object is obviously in there, but you’ll check the index file to make sure. This is also useful if you have multiple packfiles on the server, so you can see which packfile contains the object you need:

Now that you have the packfile index, you can see if your object is in it – because the index lists the SHA-1s of the objects contained in the packfile and the offsets to those objects. Your object is there, so go ahead and get the whole packfile:

You have your tree object, so you continue walking your commits. They’re all also within the packfile you just downloaded, so you don’t have to do any more requests to your server. Git checks out a working copy of the master branch that was pointed to by the HEAD reference you downloaded at the beginning.

The dumb protocol is simple but a bit inefficient, and it can’t handle writing of data from the client to the server. The smart protocol is a more common method of transferring data, but it requires a process on the remote end that is intelligent about Git – it can read local data, figure out what the client has and needs, and generate a custom packfile for it. There are two sets of processes for transferring data: a pair for uploading data and a pair for downloading data.

This section contains a very basic overview of the transfer protocols. The protocol includes many other features, such as multi_ack or side-band capabilities, but covering them is outside the scope of this book. We’ve tried to give you a sense of the general back-and-forth between client and server; if you need more knowledge than this, you’ll probably want to take a look at the Git source code.