integration-resilience-engineer

Design resilient integration patterns for Kotlin Spring services by setting explicit timeout budgets, applying exponential backoff with jitter, enforcing idempotency for deduplication, and isolating fault domains with circuit breakers to tolerate partial failures across HTTP, messaging, and scheduled workflows.

Prevent thundering herds and cascading failures with unstable external systems by applying timeout budgets, exponential backoff, jitter, and circuit breaking to isolate fault domains and control retry storms in Kotlin Spring integrations.

Handle duplicate message deliveries safely in Spring messaging workflows by implementing idempotency and deduplication logic, ensuring that repeated deliveries across boundaries do not cause duplicate side effects or data corruption.

Design a dead letter queue for poison messages by capturing failed messages after exhausting retries, providing clear visibility into retry attempts and outcomes through observable metrics and traces for Spring integration failure analysis.

This resilience pattern works for HTTP calls, message queues, and scheduled jobs in Kotlin Spring services, providing reliability guarantees like timeout budgets, idempotency, and circuit breaking across all three integration types.

Avoid using retries without idempotency for external HTTP integrations when partial failures occur, because retrying non-idempotent operations risks duplicate side effects; always pair retry policies with deduplication to ensure safe delivery handling.