Starting 08:15 UTC on September 24, 2025 customers began to experience a global degradation of service provisioning new Object Storage buckets, autoscaling of LKE resources, attaching Block Storage volumes to Linodes, and any other action which relies upon the Compute API for processing. Akamai was able to restore functionality at approximately 10:00 UTC on September 24, 2025.

The investigation revealed that the issue started as a result of unstable database performance, which began after a change release at 07:45 UTC on September 24, 2025. As the change propagated, database response times have increased and service became degraded.

To mitigate the issue, we rolled back the change at 08:41 UTC on September 24, 2025 and introduced database performance improvement actions so that our primary database can recover quicker. 

At this time our systems are operating as expected. A reoccurrence of this issue is unlikely, however, Akamai will continue to investigate the root cause and will take appropriate preventive actions. We apologize for the inconvenience that this incident may have caused.

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.