Using dead-letter queues in Amazon SQS

This Python example shows you how to use a queue to receive and hold messages from other queues that the queues can't process.

The scenario

A dead letter queue is one that other (source) queues can target for messages that can't be processed successfully. You can set aside and isolate these messages in the dead letter queue to determine why their processing did not succeed. You must individually configure each source queue that sends messages to a dead letter queue. Multiple queues can target a single dead letter queue.

In this example, Python code is used to route messages to a dead letter queue. The code uses the SDK for Python to use dead letter queues using this method of the AWS.SQS client class:

• set_queue_attributes.

For more information about Amazon SQS dead letter queues, see Using Amazon SQS Dead Letter Queues in the Amazon Simple Queue Service Developer Guide.

Prerequisite tasks

To set up and run this example, you must first complete these tasks:

• Create an Amazon SQS queue to serve as a dead letter queue. For an example of creating an Amazon SQS queue, see Create a queue.

Configure source queues

After you create a queue to act as a dead letter queue, you must configure the other queues that route unprocessed messages to the dead letter queue. To do this, specify a redrive policy that identifies the queue to use as a dead letter queue and the maximum number of receives by individual messages before they are routed to the dead letter queue.

The example below shows how to:

• Configure a source queue using set_queue_attributes.

Example

import json

import boto3

# Create SQS client
sqs = boto3.client('sqs')

queue_url = 'SOURCE_QUEUE_URL'
dead_letter_queue_arn = 'DEAD_LETTER_QUEUE_ARN'

redrive_policy = {
    'deadLetterTargetArn': dead_letter_queue_arn,
    'maxReceiveCount': '10'
}


# Configure queue to send messages to dead letter queue
sqs.set_queue_attributes(
    QueueUrl=queue_url,
    Attributes={
        'RedrivePolicy': json.dumps(redrive_policy)
    }
)