Due to an issue in which client-side events were unable to be ingested by our system, a thundering herd issue occurred as a result of aggressive client-side retry policies.

We have both fixed the underlying issue as well as hardened our client SDK in @interfere/next v4.0.0, and improved our alerting and rollback systems to be more resilient to potential failures like this in the future.

An issue with our ingestion pipeline has been identified. We're on it.