Git Dumb HTTP Transport to be turned off in 90 days
UPDATE: Dumb HTTP has been disabled as of Sunday, June 12, 2011 at 2AM PST. If you’d like a little more time to transition, please contact support and we can…
UPDATE: Dumb HTTP has been disabled as of Sunday, June 12, 2011 at 2AM PST. If you’d like a little more time to transition, please contact support and we can temporarily enable support on a repository by repository basis.
UPDATE: We’ve extended the shutoff date to Sunday, June 12, 2011 to give people a bit more time.
Starting June 7th, 2011, GitHub will no longer support fetching over
Git’s “dumb” HTTP transport mechanism. This won’t effect you unless
you’re running a git version prior to v1.6.6 and fetch using a http
remote URL.
Based on our logs, less than 7% of all git HTTP clients are using the
dumb HTTP transport today.
Git v1.6.6 (released January, 2010) added a “smart” HTTP transport
that is superior to the old transport in every way. We added support
for the smart transport on April 22, 2010. The “dumb” transport is
overly chatty on the wire and requires exposing internal repository
data, restricting our ability to store things efficiently on disk in
some cases.
Please check that your git version is at least 1.6.6 and upgrade if
necessary before June 5th:
Most Linux distributions have newer packages available in their stable
repositories. Latest versions of Mac, Windows, and Java clients should
also support the smart transport.
If you rely on git’s dumb HTTP transport and feel you have no easy
upgrade path, we want to hear from you: please
open a support ticket.
Written by
Related posts
First Look: Exploring OpenAI o1 in GitHub Copilot
We’ve tested integrating OpenAI o1-preview with GitHub Copilot. Here’s a first look at where we think it can add value to your day to day.
GitHub Availability Report: August 2024
In August, we experienced one incident that resulted in degraded performance across GitHub services.
Fine-tuned models are now in limited public beta for GitHub Copilot Enterprise
Fine-tuned models empower organizations to receive code suggestions specifically tailored to their coding practices and internal languages.