On April 16, 2025, at approximately 20:10 UTC, we noticed numerous internal alerts fired across multiple data centers, indicating that various servers were unreachable for an extended period. At the same time, customers started reporting issues with multiple products.
During the impact window, customers may have experienced issues with all Compute Products, which prevented them from deploying new Linodes, booting Linodes, and performing other host-level jobs. There was intermittent and/or total loss of network connectivity, as well as an inability to access services such as Object Storage or Linode Kubernetes Engine (LKE).
The investigation has indicated that the issue can be attributed to a configuration anomaly on BGP route reflector infrastructure providing routes for Compute services; however, it has been determined that only those route servers that had undergone a restart at some point were experiencing this impact.
Akamai implemented the correct new configuration across all affected route servers at approximately 22:15 UTC to alleviate the residual impact. Nevertheless, a degree of impact reemerged in specific locations during the mitigation efforts.
This process was undertaken in phases, commencing with the regions most adversely affected, and was concluded at approximately 03:45 UTC on April 17, 2025. Following an extensive monitoring period of our systems, we verified that the issue has been resolved.
This summary provides an overview of our current understanding of the incident, given the information available. Our investigation is ongoing, and any information herein is subject to change.