Software teams were stymied late Sunday and early Monday by a widespread GitHub outage that prevented developers from using the vital code repository for several hours.
No data was lost, according to a status update posted late Sunday by Jason Warner, senior vice president for technology at GitHub. While that was undoubtedly a huge relief to software teams trying to access GitHub when they reported for work Monday morning, developers were still unable to work on issues or examine pull requests on GitHub during the outage because of inconsistent results generated by database errors.
The incident only underscores the central role GitHub plays in the process of developing software around the world. Millions of developers, working on small projects as well as projects as massive as Windows, rely on GitHub to store their code and manage the process of updating that code.
It’s still not clear exactly what caused the issue as GitHub works to clean up the remainder of the mess, but the incident could be a argument for moving the service onto Microsoft’s Azure cloud computing service once the $7.5 billion acquisition clears the regulatory hurdles. LinkedIn, Microsoft’s most prominent (and most expensive) web acquisition still runs its services on its own infrastructure, but given GitHub’s utility-like in the world of software factories, uptime is an extremely important consideration.