US west region outage due to facility cooling loss on August 13, 2026
Blog post from Blacksmith
On August 13, 2026, utility fluctuations and city grid outages caused a Phoenix data center to lose cooling, leading temperatures to rise and much of the region’s compute, storage, caching, sticky disk, and static IP infrastructure to fail. Storage clusters lost quorum first, causing cache-dependent jobs to hang until caching was manually disabled, while approximately 2,400 compute hosts became unavailable and remote management systems prevented direct recovery until early August 14. Workloads were shifted to other regions, preserving service availability but increasing queue times across the network, with tens of thousands of jobs waiting in Phoenix and Ashburn and some larger runners remaining unscheduled for hours. GitHub Actions job adoption in Phoenix returned near normal by 21:45 UTC on August 13, while storage-related services were restored by 18:09 UTC on August 14 after remote access enabled power cycling and recovery work. Planned improvements include distributing capacity across more facilities, automatically bypassing unavailable storage rather than allowing jobs to hang, and making incident-response configuration changes possible without service redeployments.
No tracked trend matches for this post yet.
Use this post, company, and trend context to find content marketing opportunities, perform competitive analysis, or address product feature gaps via the Plushcap MCP server or the Plushcap API.