Fly.io has resolved a major incident affecting its GRU region. The platform identified a bug during a distributed state database migration that prevented users from creating or updating machines in the region, while also causing connection failures for applications running in GRU and clients connecting from nearby areas.
Fly.io provides a platform for deploying and running applications globally across multiple regions, serving developers and organizations that rely on the service for application hosting and infrastructure.
The company began investigating the issue after identifying the bug, confirming that while existing apps continued to run, new machine deployments and updates were failing. Engineers implemented a fix and transitioned to monitoring the results to ensure stability was restored.
The incident is now fully resolved.