This alert is specifically designed to monitor and identify potential issues with message delivery, such as endpoint problems, network issues, or misconfigurations. This ensures that the majority of messages are being delivered successfully. This alert triggers when the percentage of successfully delivered notifications drops below 95%. Customization Guidance: - Threshold: The default threshold is set at a delivery rate of less than 95% over the last 10 minutes. Depending on the individual topic and expected load, this threshold may be adjusted to better reflect operational norms and service requirements. - Monitoring Period: The monitoring period can be adjusted to shorter or longer than 10 minutes based on the traffic pattern and criticality of the service. Shorter periods may be used for high-traffic, critical functions to catch issues more rapidly. - Notification Frequency: Consider the frequency of this alert to optimize the balance between responsiveness and noise. Adjust according to the criticality of the service's uninterrupted operation. Action: Should this alert trigger, check the reasons for message delivery failures, which could include invalid endpoints, permission issues, or network problems. Ensure that all subscribing endpoints are correctly configured and available to receive messages. Verify the health and performance of the SNS service to ensure there are no ongoing issues or outages. Look into any recent changes in your SNS configuration or the publishing application that might have affected message delivery.

This alert is designed to monitor and identify a high rate of failed notifications within Amazon SNS. A high rate of failed notifications can indicate issues with endpoint configurations, network problems, or other underlying issues that prevent successful message delivery. This alert trigger alert if the number of failed notifications exceeds 100 in the last 10 minutes. Customization Guidance: - Threshold: The default threshold is set at more than 100 failed notifications in the last 10 minutes. Depending on the individual topic and expected load, this threshold may be adjusted to better reflect operational norms and service requirements. - Monitoring Period: The monitoring period can be adjusted to shorter or longer than 10 minutes based on the traffic pattern and criticality of the service. Shorter periods may be used for high-traffic, critical functions to catch issues more rapidly. - Notification Frequency: Consider the frequency of this alert to optimize the balance between responsiveness and noise. Adjust according to the criticality of the service’s uninterrupted operation. Action: Should this alert trigger, check the reasons for message delivery failures, which could include invalid endpoints, permission issues, or network problems. Ensure that all subscribing endpoints are correctly configured and available to receive messages. Verify the health and performance of the SNS service to ensure there are no ongoing issues or outages. Look into any recent changes in your SNS configuration or the publishing application that might have affected message delivery.

This alert is designed to monitor and manage the rate of messages that are filtered out by subscription filter policies in Amazon Simple Notification Service (SNS). It aims to ensure that filter policies are appropriately calibrated, allowing for the efficient delivery of relevant messages while avoiding the exclusion of necessary communications. The alert triggers when more than 1000 messages are filtered out within the last 10 minutes. A high rate of filtered messages may indicate that filter policies are too restrictive, possibly due to overly stringent criteria or misalignment with the intended message content. This can prevent important messages from reaching subscribers, impacting service effectiveness and user satisfaction. Customization Guidance: - Threshold: The default threshold is set at more than 1000 filtered messages in a 10-minute period. Adjust this threshold based on the volume and expected variability of your message traffic, and the criticality of messages for your operation. - Monitoring Period: The standard monitoring period is 10 minutes, but this can be adjusted shorter for high-traffic services or extended for services with lower message frequencies to better capture relevant data. - Notification Frequency: Consider the frequency of this alert to optimize the balance between responsiveness and noise. Adjust according to the criticality of the service’s uninterrupted operation. Action: Should this alert trigger, ensure that subscription filter policies are correctly configured and not too restrictive. Verify that the published messages meet the criteria set by the filter policies. Look for patterns or changes in the rate of filtered out messages to detect potential issues or misconfigurations. If necessary, adjust the threshold to better reflect the operational norms and requirements of your applications.