GitHub Engineering

Transit and Peering: How your requests reach GitHub

GitHub is at a scale that provides exposure to interesting aspects of running a major site and are working to mature and level-up many parts of our infrastructure as we grow. One of the areas where this is evident is in how your requests find their destination using DNS and make their way into our sites over transit and peering. Many organizations are either too small to need to tackle these sorts of problems or so large they have groups to maintain existing solutions for each portion of them. It is really compelling to be able to directly work on such projects and closely with great engineers who are solving others still.

Evolution of GitHub's data centers

Over the past 18 months we’ve made a significant investment in GitHub’s physical infrastructure. The goal of this work is to improve the redundancy and global availability of our system. In doing so we’ve solidified the foundation upon which we will expand our compute and storage footprint in support of our growing user base.

GitHub Debug

GitHub is proud to handle thousands of requests per second from our millions of users. The Internet, however, can be a fickle beast of cables and sparks, and sometimes those requests don’t happen very fast (or at all). While we’re happy to help you troubleshoot connection issues to us, we also know our users like swift answers and a hands-on approach.

Weak cryptographic standards deprecation update

Earlier this year, we announced the deprecation of several weak cryptographic standards. As noted during our initial announcement, the vast majority of HTTPS clients connect to GitHub using TLSv1.2 and won’t be affected by our disabling of TLSv1/TLSv1.1. Since the announcement, we have been focusing on the impact of disabling the diffie-hellman-group1-sha1 and diffie-hellman-group14-sha1 key exchanges for SSH. As of last week, we have enabled diffie-hellman-group-exchange-sha256. This key exchange method is widely supported and will allow most legacy clients to seamlessly transition away from diffie-hellman-group1-sha1 and diffie-hellman-group14-sha1.

Keeping an eye on our network

Visibility is essential to effectively operating complex systems. As our network has grown, we’ve had to improve the the way we collect data about it to keep up. Key to these improvements has been the ability to tag metrics. Tagging has allowed us to build dashboards that start at a high level and facilitate drilling down into interesting or problematic areas by filtering and aggregating on multiple dimensions. A simple example below shows data flowing across a set of our spine switches during a rolling update:

Older posts