Creating CloudWatch alarms to monitor AWS IoT - AWS IoT Core
How can I be notified if my things do not connect successfully each day?How can I be notified if my things are not publishing data each day?How can I be notified if my thing's shadow updates are being rejected each day?How can I create a CloudWatch alarm for jobs?

Creating CloudWatch alarms to monitor AWS IoT

You can create a CloudWatch alarm that sends an Amazon SNS message when the alarm changes state. An alarm watches a single metric over a time period you specify. When the value of the metric exceeds a given threshold over a number of time periods, one or more actions are performed. The action can be a notification sent to an Amazon SNS topic or Auto Scaling policy. Alarms trigger actions for sustained state changes only. CloudWatch alarms do not trigger actions simply because they are in a particular state; the state must have changed and been maintained for a specified number of periods.

You can see all the metrics that CloudWatch alarms can monitor at AWS IoT metrics and dimensions.

How can I be notified if my things do not connect successfully each day?

  1. Create an Amazon SNS topic named things-not-connecting-successfully, and record its Amazon Resource Name (ARN). This procedure will refer to your topic's ARN as sns-topic-arn.

    For more information on how to create an Amazon SNS notification, see Getting Started with Amazon SNS.

  2. Create the alarm.

    aws cloudwatch put-metric-alarm \
    --alarm-name ConnectSuccessAlarm \
    --alarm-description "Alarm when my Things don't connect successfully" \
    --namespace AWS/IoT \
    --metric-name Connect.Success \
    --dimensions Name=Protocol,Value=MQTT \
    --statistic Sum \
    --threshold 10 \
    --comparison-operator LessThanThreshold \
    --period 86400 \
    --evaluation-periods 1 \
    --alarm-actions sns-topic-arn

  3. Test the alarm.

    aws cloudwatch set-alarm-state --alarm-name ConnectSuccessAlarm --state-reason "initializing" --state-value OK
    aws cloudwatch set-alarm-state --alarm-name ConnectSuccessAlarm --state-reason "initializing" --state-value ALARM

  4. Verify that the alarm appears in your CloudWatch console.

How can I be notified if my things are not publishing data each day?

  1. Create an Amazon SNS topic named things-not-publishing-data, and record its Amazon Resource Name (ARN). This procedure will refer to your topic's ARN as sns-topic-arn.

    For more information on how to create an Amazon SNS notification, see Getting Started with Amazon SNS.

  2. Create the alarm.

    aws cloudwatch put-metric-alarm \
    --alarm-name PublishInSuccessAlarm\
    --alarm-description "Alarm when my Things don't publish their data \
    --namespace AWS/IoT \
    --metric-name PublishIn.Success \
    --dimensions Name=Protocol,Value=MQTT \
    --statistic Sum \
    --threshold 10 \
    --comparison-operator LessThanThreshold \
    --period 86400 \
    --evaluation-periods 1 \
    --alarm-actions sns-topic-arn

  3. Test the alarm.

    aws cloudwatch set-alarm-state --alarm-name PublishInSuccessAlarm --state-reason "initializing" --state-value OK
    aws cloudwatch set-alarm-state --alarm-name PublishInSuccessAlarm --state-reason "initializing" --state-value ALARM

  4. Verify that the alarm appears in your CloudWatch console.

How can I be notified if my thing's shadow updates are being rejected each day?

  1. Create an Amazon SNS topic named things-shadow-updates-rejected, and record its Amazon Resource Name (ARN). This procedure will refer to your topic's ARN as sns-topic-arn.

    For more information on how to create an Amazon SNS notification, see Getting Started with Amazon SNS.

  2. Create the alarm.

    aws cloudwatch put-metric-alarm \
    --alarm-name UpdateThingShadowSuccessAlarm \
    --alarm-description "Alarm when my Things Shadow updates are getting rejected" \
    --namespace AWS/IoT \
    --metric-name UpdateThingShadow.Success \
    --dimensions Name=Protocol,Value=MQTT \
    --statistic Sum \
    --threshold 10 \
    --comparison-operator LessThanThreshold \
    --period 86400 \
    --unit Count \
    --evaluation-periods 1 \
    --alarm-actions sns-topic-arn

  3. Test the alarm.

    aws cloudwatch set-alarm-state --alarm-name UpdateThingShadowSuccessAlarm --state-reason "initializing" --state-value OK
    aws cloudwatch set-alarm-state --alarm-name UpdateThingShadowSuccessAlarm --state-reason "initializing" --state-value ALARM

  4. Verify that the alarm appears in your CloudWatch console.

How can I create a CloudWatch alarm for jobs?

The Jobs service provides CloudWatch metrics for you to monitor your jobs. You can create CloudWatch alarms to monitor any Jobs metrics.

The following command creates a CloudWatch alarm to monitor the total number of failed job executions for Job SampleOTAJob and notifies you when more than 20 job executions have failed. The alarm monitors the Jobs metric FailedJobExecutionTotalCount by checking the reported value every 300 seconds. It is activated when a single reported value is greater than 20, meaning there were more than 20 failed job executions since the job started. When the alarm goes off, it sends a notification to the provided Amazon SNS topic.

aws cloudwatch put-metric-alarm \
    --alarm-name TotalFailedJobExecution-SampleOTAJob \
    --alarm-description "Alarm when total number of failed job execution exceeds the threshold for SampleOTAJob" \
    --namespace AWS/IoT \
    --metric-name FailedJobExecutionTotalCount \
    --dimensions Name=JobId,Value=SampleOTAJob \
    --statistic Sum \
    --threshold 20 \
    --comparison-operator GreaterThanThreshold \
    --period 300 \
    --unit Count \
    --evaluation-periods 1 \
    --alarm-actions arn:aws:sns:<AWS_REGION>:<AWS_ACCOUNT_ID>:SampleOTAJob-has-too-many-failed-job-ececutions

The following command creates a CloudWatch alarm to monitor the number of failed job executions for Job SampleOTAJob in a given period. It then notifies you when more than five job executions have failed during that period. The alarm monitors the Jobs metric FailedJobExecutionCount by checking the reported value every 3600 seconds. It is activated when a single reported value is greater than 5, meaning there were more than 5 failed job executions in the past hour. When the alarm goes off, it sends a notification to the provided Amazon SNS topic.

aws cloudwatch put-metric-alarm \
    --alarm-name FailedJobExecution-SampleOTAJob \
    --alarm-description "Alarm when number of failed job execution per hour exceeds the threshold for SampleOTAJob" \
    --namespace AWS/IoT \
    --metric-name FailedJobExecutionCount \
    --dimensions Name=JobId,Value=SampleOTAJob \
    --statistic Sum \
    --threshold 5 \
    --comparison-operator GreaterThanThreshold \
    --period 3600 \
    --unit Count \
    --evaluation-periods 1 \
    --alarm-actions arn:aws:sns:<AWS_REGION>:<AWS_ACCOUNT_ID>:SampleOTAJob-has-too-many-failed-job-ececutions-per-hour