Dead letter queue

In message queueing the dead letter queue is a service implementation to store messages that meet one or more of the following criteria:

1. Message that is sent to a queue that does not exist.[1][2]
2. Queue length limit exceeded.
3. Message length limit exceeded.
4. Message is rejected by another queue exchange.[3]
5. Message reaches a threshold read counter number, because it is not consumed. Sometimes this is called a "back out queue".
6. The message expires due to per-message TTL (time to live)[4]

Dead letter queue storing of these messages allows developers to look for common patterns and potential software problems.[5]

Queueing systems that incorporate dead letter queues include Amazon EventBridge,[6] Amazon Simple Queue Service,[5] Apache ActiveMQ, Google Cloud Pub/Sub,[7] HornetQ, Microsoft Message Queuing,[1] Microsoft Azure Event Grid and Azure Service Bus,[8] WebSphere MQ,[9] Solace PubSub+,[10] Rabbit MQ[3] and Apache Pulsar,[11][12]

References

1. Redkar, Arohi (2004). Pro MSMQ: Microsoft Message Queue Programming. Apress. p. 148. ISBN 1430207329.
2. "Dead-letter queues". IBM. Retrieved 23 February 2014.
3. RabbitMQ dead letter queue "Dead Letter Exchanges".
4. "Dead Letter Exchanges — RabbitMQ". www.rabbitmq.com. Retrieved 2020-12-06.
5. "Using Amazon SQS Dead Letter Queues". Amazon. Retrieved 23 February 2014.
6. "Amazon EventBridge announces support for Dead Letter Queues". Amazon.
7. "Forwarding to dead-letter topics | Cloud Pub/Sub". Google Cloud. Retrieved 2020-12-26.
8. spelluru. "Compare Azure messaging services". docs.microsoft.com. Retrieved 2020-01-17.
9. Böhm-Mäder, Johannes. WebSphere MQ Security: Tales of Scowling Wolves Among Unglamorous Sheep. BoD. p. 68. ISBN 3842381506.
10. "Solace Dead Message Queues".
11. "Apache Pulsar documentation".
12. "Apache Pulsar PIP-22:Dead Letter Topic".