Answer: Configure an Amazon Simple Queue Service (Amazon SQS) queue as the on-failure destination. Modify the Lambda function to process messages in the queue.

## Explanation **Correct Answer: D** **Why Option D is correct:** 1. **Dead Letter Queue (DLQ) Pattern**: Configuring an SQS queue as an on-failure destination (also known as a Dead Letter Queue or DLQ) is a standard AWS best practice for handling failed Lambda invocations. 2. **Message Persistence**: When the Lambda function fails due to network connectivity issues, the message (notification) is automatically sent to the SQS queue, which provides durable message storage. 3. **Retry Mechanism**: The Lambda function can be modified to periodically poll and process messages from the SQS queue, ensuring eventual processing of all notifications. 4. **Decoupling**: This approach decouples the notification reception (SNS) from the processing (Lambda), providing a buffer that can handle temporary failures. **Why other options are incorrect:** **Option A**: Deploying Lambda across multiple Availability Zones doesn't address the core issue. Lambda functions are already highly available and automatically run across multiple AZs in a region. The problem is with message processing failures, not Lambda availability. **Option B**: Increasing CPU and memory allocations won't solve network connectivity issues. Network problems are external to the Lambda function's compute resources. **Option C**: While SNS does have retry capabilities, increasing retries and wait times is not the most robust solution. SNS retries are limited in duration (up to 24 hours), and this approach doesn't provide persistent storage for failed messages. If the network issue persists beyond the retry period, messages will still be lost. **Key AWS Services Relationship:** - **SNS → Lambda**: Direct integration where SNS can invoke Lambda functions - **SNS → SQS → Lambda**: More resilient pattern where SNS publishes to SQS queue, and Lambda polls from the queue - **Lambda DLQ**: Lambda can be configured with SQS or SNS as a DLQ destination for failed invocations **Best Practice Implementation:** 1. Create an SQS queue as a DLQ for the Lambda function 2. Configure the Lambda function's `on-failure` destination to point to the SQS queue 3. Modify the Lambda function to include logic to process messages from the DLQ 4. Optionally, set up CloudWatch alarms to monitor the DLQ depth 5. Consider implementing exponential backoff retry logic in the Lambda function for processing DLQ messages

Author: LeetQuiz Editorial Team