Archive · 21 Aug 2026
GitHub Blames Seven-Hour Outage on Critical Infrastructure Failure
GitHub, a service that helps people store and manage computer code, had a major outage.
The outage lasted seven hours and 47 minutes on August 17.
A key part of GitHub’s infrastructure could not handle a new peak in traffic.
This created pressure that spread to other parts of the platform.
Users had trouble signing in and using tools such as GitHub Actions, pull requests, issues, APIs, and Copilot.
GitHub said the problem was not caused by a code or configuration change.
The company is adding computer capacity and changing how its systems connect to reduce the chance of another large outage.
It is also moving more of its platform to Microsoft Azure.
GitHub said a critical infrastructure component failed to scale when traffic reached a new peak on August 17.
The resulting capacity pressure caused authentication failures and disrupted GitHub’s website, APIs, Actions, pull requests, issues, and Copilot.
The outage lasted seven hours and 47 minutes, with some Copilot services recovering later than others.
GitHub said monthly commits had increased from 1.4 billion to 2.9 billion since April, intensifying pressure on its systems.
The company plans to add capacity, improve efficiency, remove architectural bottlenecks, and accelerate migration to Microsoft Azure.
- Who
- GitHub and users of its code-hosting platform.
- What
- A seven-hour-and-47-minute outage caused by a critical infrastructure component failing to scale under peak traffic.
- When
- Monday, August 17; the article does not specify a year.
- Where
- A GitHub data centre in the United States, with effects across the platform worldwide.
- Why
- Traffic reached a new peak, creating capacity pressure that spread through GitHub’s systems and caused service failures.
- Outage duration
- Seven hours and 47 minutes
- Primary failure
- A critical infrastructure component did not scale when traffic reached a new peak
- Affected services
- github.com, authentication, GitHub Actions, APIs, pull requests, issues, and Copilot
- Commit growth
- Monthly commits increased from 1.4 billion to 2.9 billion since April
- Recent incidents
- GitHub reported two major outages in August, on August 6 and August 17
- Azure platform load
- Microsoft Azure serves roughly 58% of GitHub’s platform load and half of all Git operations
- Added capacity
- GitHub said it added more than 3 million CPU cores and 120 petabytes of high-speed storage