On August 13th, 2026, at approximately 18:15 UTC, Akamai observed a brief service outage affecting api.linode.com. The total service interruption lasted for approximately 3 minutes, concluding at 18:18 UTC. Following the restoration of initial connectivity, elevated API response latency persisted through 19:06 UTC, causing slower response times and intermittent delays for customers interacting with API services.

To address the performance impact, Akamai engineering teams identified a configuration discrepancy on the secondary caching infrastructure node that prevented it from absorbing the full traffic load after the failover. Engineers completed a controlled migration and moved request caching traffic back over to the primary host. Following this change, API latency rapidly decreased to normal operational levels.

The initial condition was triggered by an unexpected reboot of the primary caching node's physical host. While redundant infrastructure was active, the secondary node was unable to process the failover traffic seamlessly, causing the extended performance degradation.

Our engineering teams are conducting a follow-up investigation into the failover mechanisms to optimize execution speeds and align configuration settings across redundant nodes, ensuring secondary systems can handle traffic seamlessly in future events.

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.