GitHub February 2025 Availability Report
GitHub recorded two service degradation incidents during February 2025, both stemming from internal infrastructure and deployment issues rather than external factors. The incidents affected notifications and migration tooling, with no security concerns reported.
Migration Tooling Outage (February 3)
The first incident was declared on February 3 at 18:01 UTC, when GitHub's Migration Tools experienced a total failure. A deployment of a system component introduced missing Docker images, causing a 100% outage for any user attempting a migration during the event. Engineering teams rolled back to the previous stable version, restoring full service within approximately 30 minutes. To prevent recurrence, GitHub has expanded test coverage and enhanced workflows to validate critical dependencies.
Notification Delivery Delays (February 25)
The second incident spanned February 25 from 14:25 UTC to 16:44 UTC, lasting 2 hours and 19 minutes. During this window, email and web notifications were subject to delivery delays. At the peak, approximately 10% of notifications took more than 10 minutes to arrive, while the remaining 90% were delivered within 5–10 minutes. The slowdown tied to worker pools operating too close to their capacity limit at peak periods, which blocked queue processing.
GitHub mitigated the problem by scaling out the notification service to handle demand. The company has since raised baseline capacity and improved its capacity-planning procedures to proactively manage pool sizing going forward.
For minute-by-minute status updates and post-incident reports, refer to the status page. Engineering deep-dives and related work can be found on the GitHub Engineering Blog.
