We're confident that the latest adjustment fixed the issue.

We will continue monitoring FF closely after resolution, but we don't see customer-facing impact anymore.

Thank you for the patience.

The service is stable as we are applying some optimizations.

We've resolved the immediate issue with the throttling and the service is back to normal.

We are still investigating further what the root cause of the congestion.

We identified the issue with our flags service being throttled on cpu allocations. The service has been scaled up and we have seen latency and error counts stablilise again.

We're investigating what could have led to this cpu exhaustion and will update soon.

We're experiencing an elevated level of latency and errors on our feature flag endpoints and are currently looking into the issue.