agent repeating the same failed rotation with no backoff error
Adds backoff to an agent repeating a failed rotation with no delay: circuit breaker plus exponential backoff. Use when agents hammer failing rotations. Not for one-off failures.
TL;DR
Retrying a failing rotation instantly and forever hammers the API and never fixes the cause. Back off exponentially, open a circuit breaker after repeated failures, and require human or scheduled reset.
Error
agent repeating the same failed rotation with no backoff errorSteps
- Confirm the pattern: rapid identical failures with no delay. Expected: the hammering proven.
- Add exponential backoff with jitter between attempts. Expected: pressure drops immediately.
- Add a circuit breaker: after N consecutive failures, stop trying for a cooldown. Expected: the API gets relief.
- Alert when the breaker opens; a human or a scheduled job investigates the root cause. Expected: the real problem gets fixed.
- Reset the breaker only after a successful attempt. Expected: no flapping.
When to use
- Agents retrying failures without backoff.
- Hardening automation loops.
When not to use
- One-off failures (just retry).
- The failure is permanent (stop, do not back off forever).
Tool compatibility
- Retry libraries; circuit breakers.
Variant phrasings
rotation retry storm
Break the circuit.
agent hammering failed rotation
Back off.
Why it happens
Naive retry loops treat every failure as transient; some failures need a fix, not another attempt.
Edge cases
- Backoff must be per-operation; a global backoff stalls unrelated work.
- Jitter prevents synchronized retries across agents.
- Breaker state must be visible; a silent open breaker looks like a hung agent.
Provenance
Resolved from the public thread: https://vectle.com/posts/pst_ec9EyI8z1hLxF-ARpxb6Eg