Ted Hisokawa
Jul 18, 2025 06:05
GitHub experienced three significant incidents in June 2025, affecting performance across its services, including Actions and Copilot. The company outlines mitigation steps and future improvements.
GitHub, a leading platform for version control and collaboration, encountered several service disruptions in June 2025, impacting its wide array of services. The detailed availability report highlights three significant incidents that led to degraded performance and user experience, according to the GitHub Blog.
June 5 Incident: Actions Service Degradation
On June 5, between 17:47 UTC and 19:20 UTC, GitHub’s Actions service faced a severe degradation. Users reported delays in run starts and intermittent job failures. During this period, 47.2% of runs were delayed by an average of 14 minutes, and 21% of runs failed. The issue extended beyond Actions, affecting 60% of Copilot Coding Agent sessions and causing failures in deploying Pages sites using branch-based builds, although Pages serving remained unaffected. The root cause was identified as a spike in load that exposed a misconfiguration, triggering request throttling. The issue was resolved by correcting the service configuration and updating deployment processes to prevent future occurrences.
June 12 Incident: Copilot Service Outage
The second incident occurred on June 12, from 17:55 UTC to 21:07 UTC, impacting the GitHub Copilot service. Users experienced elevated error rates, slow responses, and chat functionality interruptions across various platforms, including VS Code and JetBrains IDEs. The disruption was traced to an outage affecting one of GitHub’s model providers. GitHub mitigated the impact by disabling the affected provider endpoints temporarily and is now working on improving incident response strategies and monitoring systems to enhance detection and mitigation of similar issues in the future.
June 17 Incident: Network Reachability Issues
On June 17, from 19:32 UTC to 20:03 UTC, GitHub faced network reachability issues due to an internal routing policy deployment affecting certain network address blocks within its datacenters. This led to error rates of 3-4% for authenticated users of the GitHub UI and 40% for API callers. Unauthenticated requests faced nearly 100% error rates. The deployment was rolled back at 19:54 UTC, restoring network availability, and normal operations resumed by 20:03 UTC. GitHub plans to expand its validation process for routing policy changes to prevent similar disruptions.
GitHub encourages users to visit their status page for real-time updates and further insights into ongoing improvements can be found on the GitHub Engineering Blog.
Image source: Shutterstock
Credit: Source link
