v2.9 · Production runbook

Retry Policy Engineering

Design retry schedules that respect provider signals and operational objectives.

Principles

Use timestamps, queue IDs, remote responses, source IP, VMTA, sender domain and message class to establish the failure boundary. Avoid reacting to one isolated delivery attempt.

Policy design

  • Fast early retries for transient network conditions
  • Progressive spacing for provider throttling
  • Finite lifetime matched to message value
  • Separate policies for transactional and bulk traffic

Validation

Review the distribution of successful deliveries by retry attempt before shortening or extending queue lifetime.

Operational rule: collect evidence before changing policy, make one controlled change, verify the result, and retain a rollback path.
Search Trushilla Documentation