Resolved
Our system completed an emergency scale-out, we restarted all stuck pods, and confirmed backlog processing was back to nominal levels.
Investigating
We experienced an unprecedented surge in incoming requests that saturated our ingest cluster.
