The trigger for the outage is gone and you have restarted the service, and it collapses again within a minute of taking traffic. Why will it not recover on its own?

Short answer

The system is being held down by effects it now generates itself: accumulated client retries, an empty cache multiplying database load, and a queue of requests whose callers have already given up.

This is load shedding and admission control during recovery. Good load shedding protects the service from work it cannot finish, while admission control decides which requests are allowed in before queues explode. Weak load shedding starts only after latency is already catastrophic. Strong admission control lets the system warm caches, rebuild pools, and serve the highest-value traffic instead of accepting everything and failing everything.

Load shedding should begin before queues and thread pools are already saturated. Load shedding is a recovery control, not only a last-ditch defence.