CloudWatch log metric filter and alarm for usage of root account should be configured¶
Description¶
A CloudWatch metric filter and alarm should be established for root login attempts. Monitoring for root account logins provides visibility into the use of a fully privileged account and the opportunity to reduce it.
Console Remediation Steps¶
This is a two part process. First, you create a Metric Filter for specific CloudTrail log events. Next, you create a CloudWatch alarm for the filter. See Creating CloudWatch Alarms for CloudTrail Events: Examples for more information.
Step 1: To create the Metric Filter:
Navigate to CloudWatch.
In the left navigation, click Log Groups and select the desired log group. The log group must be assigned to a multi-region CloudTrail trail that has logging enabled (i.e., is in Logging status).
Select Metric filters > Create Metric Filter.
In Filter pattern, enter the following:
{ $.userIdentity.type = "Root" && $.userIdentity.invokedBy NOT EXISTS && $.eventType != "AwsServiceEvent" }Click Next.
For Filter name, type RootAccountUsage.
For Metric namespace, type CloudTrailMetrics.
For Metric name, type RootAccountUsageCount.
For Metric value, type 1.
Click Next > Create metric filter.
Step 2: To create an Alarm:
Check the box next to the newly created metric filter and click Create alarm.
Select the Threshold type.
Define the alarm condition and threshold value.
Click Next.
In Alarm state trigger, select In alarm.
Select an existing SNS topic, create new topic, or use topic ARN. Note that the SNS topic must have at least one subscriber.
If you selected to create a new topic, enter a name in Create a new topic.
Enter an email address in Email endpoints that will receive the notification.
Click Create topic.
Click Next.
Enter an Alarm name.
Optionally, enter an alarm description.
Click Next > Create alarm.
CLI Remediation Steps¶
This is a two part process. First, you create the Metric Filter and next you create the Metric Alarm.
Create the Metric Filter.
Create the Metric Alarm.