The incident has been resolved. It was triggered by an Ashby rollout that caused unexpected application updates and overloaded our EU infrastructure.

We have disabled this cascading behavior and added safeguards to prevent recurrence. We apologize for any inconvenience.

We've successfully rolled out a mitigation and are observing much improved latencies across affected endpoints. We're continuing to monitor the situation and will provide a postmortem once we've wrapped our internal post-incident workflows.

We have merged a possible mitigation and are monitoring its rollout.

We have identified the likely root cause of the incident and are working on a mitigation. Latency remains elevated on EU endpoints, particularly for ATS API surface.