A major outage has led to the failure of several GitHub services. Pull requests and issue generation or viewing were not possible at all for a while; there are already signs of recovery, however.
GitHub reports the events on its own status page. Although a huge spike can be seen on Downdetector, the problem appears to have been of a relatively short duration. This is in contrast to a major outage in August 2024, when a database change led to a problem with greater consequences than today. At the time, GitHub’s own site was not addressable and Copilot, for example, was not working either.
Failover
At the time of writing, a failover of one of the primary caching hosts is taking place. With this, GitHub hopes to mitigate the problem. Still, most of the Microsoft subsidiary’s services are available as usual, limiting the impact.
Although an explanation of a GitHub outage sometimes comes up, that is not always the case. The services that are down also vary continuously. For example, the Git Operations service was down half a month ago. In most situations, organizations can guard against an outage, especially if it is short-lived. “Fortunately,” a GitHub incident is common enough that the problem occurs with some regularity, making a backup option obvious.
Also read: Fake stars undermine GitHub: 4.5 million fraudulent stars discovered