Prompt
What should I use for event-driven integrations with retries and logs?
Latest observation
Oct 5, 2026
OpenAI APIWeb search: off
For event-driven integrations with retries and logs, a solid choice is usually:
Best general-purpose options
- Apache Kafka — great for high-throughput event streams, replayability, and decoupled systems.
- RabbitMQ — better if you want simpler work queues, routing, and built-in retry/DLQ patterns.
- AWS SNS/SQS — if you’re on AWS and want managed infrastructure with easy retries and dead-letter queues.
- Azure Service Bus / Google Pub/Sub — similar managed options on those clouds.
What to use depends on your needs
- Need event replay, ordering, and many consumers? → Kafka
- Need simple task delivery + retries + dead-letter queues? → RabbitMQ or SQS
- Need fully managed with minimal ops? → Cloud-native queues/pub-sub
- Need durable workflow orchestration with retries and audit logs? → Temporal or AWS Step Functions
For logs + retries specifically
Make sure the platform supports:
- Dead-letter queues/topics
- Message acknowledgments
- Retry backoff
- Idempotency
- Correlation IDs / tracing
- Centralized logging/observability
If you want, I can recommend a specific stack based on your cloud, language, and integration volume.