During a rolling upgrade of our network acceleration nodes, an incorrect configuration by our operations team caused part of the traffic to fall back to the origin, resulting in a service disruption.
The configuration has now been rolled back and the upgrade process has been stopped.
Impact period: March 25, 19:00–19:20 GMT+8
Affected traffic: Approximately 30% of API requests
We apologize for any inconvenience caused and are reviewing the process to prevent similar issues in the future.
