GitHub Actions Outage and Accelerated Migration to Azure
GitHub is speeding up its move to Azure to prevent future service disruptions.
1GitHub Actions Outage Incident
The GitHub Actions outage on August 6, 2026, significantly impacted user productivity and trust in the platform. The prolonged downtime highlighted GitHub's failure to meet customer commitments. The company is deepening its root cause analysis (RCA) and will update the public once the investigation is complete. The primary cause of the outage was GitHub Actions still running in data centers, leading to capacity shortages.
2Accelerated Migration to Azure
GitHub is accelerating the migration of its Actions service to Azure. This move will provide the platform with greater capacity and resilience against outages. While most Actions already run on Azure, the launch service bridging the monolith to Actions had not yet been prioritized due to its asynchronous nature and ability to queue work. However, the recent outage necessitated speeding up this transition.
3Traffic Management Improvements
GitHub has implemented stronger stability controls for moving production traffic to Azure. In July, these controls helped reduce shared dependencies and allowed work to resume more confidently. GitHub is becoming less reliant on shared infrastructure and single data center locations, improving its ability to absorb growth and isolate failures.
4Monolithic Traffic Migration to Azure
In July, 52.75% of monolithic read traffic was served from Azure Central US, marking the first time over half of traffic came from Azure. Git traffic on Azure increased to 47%, and 29% of repositories now have a second replica in Central US, reducing disruption during regional issues. Critical customer workflows have also seen fewer shared failure points.
5Authentication and Permission Checks
GitHub has moved the first authentication tables from its oldest shared database to dedicated infrastructure, setting a migration pattern for remaining work. Authentication and permission checks now exert much less pressure on shared paths. The dedicated user service handles over one million queries per second, and 80% of major authorization lookups have shifted to isolated service paths.
6Repository Content and Pull Requests
Repository content traffic now runs entirely from dedicated infrastructure in Central US. The dedicated pull request service, which already serves anonymous traffic, has reached 99.87% parity with the monolith for authenticated reads. Gradual traffic shifting to this service will reduce performance degradation from shared infrastructure. One production change halved total query time on the artifacts table.
7Search Service Enhancements
Search services are now less exposed to capacity constraints that contributed to past outages. All production search workloads are served from Central US with additional headroom during demand spikes. GitHub has also begun measuring the health of critical customer workflows like pull requests, enabling earlier detection of performance issues. High-risk manual production activities are being replaced with automation.
8Reliability Metrics and Automation
GitHub is changing how it measures and manages reliability. Beyond infrastructure health, it now tracks the health of key customer workflows, allowing teams to spot degradation sooner. High-risk manual production tasks are being automated, with review controls and operational safeguards reducing dependency on flawless human execution.
9Targets and Next Steps
GitHub aims to serve 70% of read traffic and 30% of write traffic from Central US this quarter. Moving database primaries will enable more write traffic, while a second Azure region will bolster regional resilience. The company plans to fully migrate dotcom production traffic out of its data centers by the end of 2026, prioritizing availability, then capacity, then features.
10Outage Reports and Lessons Learned
GitHub has published reports on eight recent outages, detailing system strengths, weaknesses, and subsequent changes. These incidents prompted thorough reviews of infrastructure and processes. The company continues to implement new measures to enhance customer experience, with these lessons informing future planning.
11GitHub's Long-Term Strategy
GitHub's long-term goal is complete Azure migration and regional resilience. Reducing shared dependencies and expanding dedicated services are top priorities. The company is also developing better methods to measure and manage customer workflow health. These strategies aim to significantly improve overall platform reliability.
12Conclusion and Significance
GitHub's outage and accelerated Azure migration serve as a critical lesson in cloud service reliability and scalability. Its infrastructure improvements and automation efforts set an example for other tech firms. GitHub's customer-centric approach and rapid response to outages are raising industry reliability standards.
Original article: The GitHub Blog — GitHub Eylemlerinde Yaşanan Kesinti ve Alınan Önlemler