GitHub: what the August and September outage report shows

GitHub published its availability report on October 14, 2026, covering two months of work. Jakub Oleksy writes that the previous update stopped at August, and no separate post appeared in September while other public content covered the August 17 outage.

The short version: GitHub says it reduced shared-database pressure, added overload controls, created headroom for GitHub Actions, expanded failure containment for pull requests and moved more production workloads to Azure. The report also states plainly that significant incidents, a capacity pause and a missed migration target show the work is not complete.

As of October 7, GitHub had 120 of its 160 production services running in Central US. Thirteen of 51 primary MySQL databases had been moved. Azure carried 65.4% of read traffic for the core application, 48.3% of Git reads and 80.2% of stream-processing traffic. GitHub warns these figures cover different workloads and cannot be summed.

Central US deployments and traffic increases were paused on September 11 because capacity was uneven across availability zones. Permanent ramps resumed September 18 after temporary relief. GitHub missed its September 30 goal to deploy every production service there, and aims to unlock all available capacity by the end of October.

Outside the migration, 24 authentication-core tables moved off GitHub's oldest shared database, removing roughly one million queries per second from its replicas. The cluster behind issues and pull requests removed about 980 million unnecessary edit-history rows, an estimated 1.8 terabytes of recoverable storage. Weekly deadlocks fell around 77% from their June peak.

For Actions, production changes cut measured Redis operations by about 28% and relative CPU by about 30%. A separate infrastructure cache reduced Redis bandwidth by 84% and queued requests by 95%. GitHub still lists the August 6 Actions outage, the August 17 github.com outage, the September failover disruption and the controlled capacity pause as reminders that improvements matter only when they reduce customer risk.