Queue depth has fallen from 240k to 41k. Delivery latency is recovering. We will continue to watch for the next 30 minutes before resolving.
A misconfigured retry policy was producing duplicate sends and saturating the worker pool. Configuration rolled back; throughput recovering.
We are seeing increased error rates on the webhook delivery service. The dashboard and read API are not affected.
One email per incident. Unsubscribe in one click. Webhook and Slack also available.