Results of security checks

When it runs checks against the enabled controls for the enabled security standards, AWS Security Hub generates findings. These findings use the AWS Security Finding Format (ASFF). Note that if the finding size exceeds the maximum of 240 KB, then the Resource.Details object is removed. For findings for controls that use AWS Config rules, you can view the resource details on the AWS Config console.

Security Hub normally charges for each security check for a control. However, if multiple controls use the same AWS Config rule, then Security Hub only charges once for each check against the AWS Config rule. It generates separate findings for each control based on the check.

For example, the AWS Config rule iam-password-policy is used by multiple controls in the CIS AWS Foundations Benchmark standard and by IAM.7 in the Foundational Security Best Practices standard. Each time Security Hub runs a check against that AWS Config rule, it generates a separate finding for each related control, but only charges once for the check.

Control-related information in the ASFF

In addition to the required information about the finding and its associated resource, findings generated from security checks contain the following control-specific information.

Compliance

For findings generated by security checks, the Compliance field in the ASFF contains the control-related findings details. The Compliance field includes the following information.

RelatedRequirements

The list of related requirements for the control. The requirements are from the third-party security framework for the control, such as the Payment Card Industry Data Security Standard.

Status

The result of the most recent check that Security Hub ran for a given control. The results of the previous checks are kept in an archived state for 90 days.

StatusReasons

Contains a list of reasons for the value of Compliance.Status. For each reason, StatusReasons includes the reason code and a description.

The following table lists the available status reason codes and descriptions.

Reason code	Compliance.Status	Description
CLOUDTRAIL_METRIC_FILTER_NOT_VALID	FAILED	The multi-Region CloudTrail trail does not have a valid metric filter.
CLOUDTRAIL_METRIC_FILTERS_NOT_PRESENT	FAILED	Metric filters are not present for the multi-Region CloudTrail trail.
CLOUDTRAIL_MULTI_REGION_NOT_PRESENT	FAILED	The account does not have a multi-Region CloudTrail trail with the required configuration.
CLOUDTRAIL_REGION_INVAILD	WARNING	Multi-Region CloudTrail trails are not in the current Region.
CLOUDWATCH_ALARM_ACTIONS_NOT_VALID	FAILED	No valid alarm actions are present.
CLOUDWATCH_ALARMS_NOT_PRESENT	FAILED	CloudWatch alarms do not exist in the account.
CONFIG_ACCESS_DENIED	WARNING AWS Config status is ConfigError	AWS Config access denied. Verify that AWS Config is enabled and has been granted sufficient permissions.
CONFIG_EVALUATIONS_EMPTY	PASSED	AWS Config evaluated your resources based on the rule. The rule did not apply to the AWS resources in its scope, the specified resources were deleted, or the evaluation results were deleted.
CONFIG_RETURNS_NOT_APPLICABLE	NOT_AVAILABLE	The compliance status is NOT_AVAILABLE because AWS Config returned a status of Not Applicable. AWS Config does not provide the reason for the status. Here are some possible reasons for the Not Applicable status: The resource was removed from the scope of the AWS Config rule. The AWS Config rule was deleted. The resource was deleted. The AWS Config rule logic can produce a Not Applicable status.
CONFIG_RULE_EVALUATION_ERROR	WARNING AWS Config status is ConfigError	This reason code is used for several different types of evaluation errors. The description provides the specific reason information. The type of error can be one of the following. An inability to perform the evaluation because of a lack of permissions. The description provides the specific permission that is missing. A missing or invalid value for a parameter. The description provides the parameter and the requirements for the parameter value. An error reading from an S3 bucket. The description identifies the bucket and provides the specific error. A missing AWS subscription. A general timeout on the evaluation. A suspended account.
CONFIG_RULE_NOT_FOUND	WARNING AWS Config status is ConfigError	Unable to find the supporting AWS Config rule. Verify that you have enabled AWS Config.
S3_BUCKET_CROSS_ACCOUNT_CROSS_REGION	WARNING	The finding is in a WARNING state, because the S3 bucket associated with this rule is in a different Region or account. This rule does not support cross-Region or cross-account checks. It is recommended that you disable this control in this Region or account. Only run it in the Region or account where the resource is located.
SNS_SUBSCRIPTION_NOT_PRESENT	FAILED	The CloudWatch Logs metric filters do not have a valid Amazon SNS subscription.
SNS_TOPIC_CROSS_ACCOUNT_CROSS_REGION	WARNING	The finding is in a WARNING state because the SNS topic associated with this rule is in a different Region or account. This rule does not support cross-Region or cross-account checks. It is recommended that you disable this control in this Region or account. Only run it in the Region or account where the resource is located.
THROTTLING_ERROR	NOT_AVAILABLE	The relevant API operation exceeded the allowed rate.

ProductFields

For findings generated by security checks, ProductFields includes the following fields:

StandardsGuideArn or StandardsArn

The ARN of the standard associated with the control.

For the CIS AWS Foundations Benchmark standard, the field is StandardsGuideArn.

For PCI DSS and AWS Foundational Security Best Practices standards, the field is StandardsArn.

StandardsGuideSubscriptionArn or StandardsSubscriptionArn

The ARN of the account's subscription to the standard.

For the CIS AWS Foundations Benchmark standard, the field is StandardsGuideSubscriptionArn.

For the PCI DSS and AWS Foundational Security Best Practices standards, the field is StandardsSubscriptionArn.

RuleId or ControlId

The identifier of the control.

For the CIS AWS Foundations Benchmark standard, the field is RuleId.

For the PCI DSS and AWS Foundational Security Best Practices standards, the field is ControlId.

RecommendationUrl

The URL to the remediation information for the control.

RelatedAWSResources:0/name

The name of the resource associated with the finding.

RelatedAWSResource:0/type

The type of resource associated with the control.

StandardsControlArn

The ARN of the control.

aws/securityhub/ProductName

For control-based findings, the product name is Security Hub.

aws/securityhub/CompanyName

For control-based findings, the company name is AWS.

aws/securityhub/annotation

A description of the issue uncovered by the control.

aws/securityhub/FindingId

The identifier of the finding.

Determining the severity of controls and control findings

The severity assigned to a Security Hub control identifies the importance of the control. It determines the severity label assigned to the control findings.

Severity criteria

The severity of a control is determined based on an assessment of the following criteria:

• How difficult is it for a threat actor to take advantage of the configuration weakness associated with the control?

  The difficulty is determined by the amount of sophistication or complexity required to use the weakness to carry out a threat scenario.

• How likely is it that the weakness will lead to a compromise of your AWS accounts or resources?

  A compromise of your AWS accounts or resources means that confidentiality, integrity, or availability of your data or AWS infrastructure is damaged in some way.

  The likelihood of compromise indicates how likely it is that the threat scenario will result in a disruption or breach of your AWS services or resources.

As an example, consider the following configuration weaknesses:

• IAM user access keys are not rotated every 90 days

• IAM root access key exists

Both weaknesses are equally difficult for an adversary to take advantage of. In both cases, the adversary can use credential theft or some other method to acquire a user key. They can then use it to access your resources in an unauthorized way.

However, the likelihood of a compromise is much higher if the threat actor acquires the root user access key, because the root key gives them greater access. As a result, the root user key weakness has a higher severity.

The severity does not take into account the criticality of the underlying resource. Criticality is the level of importance of the resources that are associated with the finding. For example, a resource associated with a mission critical application is more critical than one associated with nonproduction testing. To capture resource criticality information, use the Criticality field of the AWS Security Finding Format (ASFF).

The following table maps the difficulty to exploit and the likelihood of compromise to the security labels.

	Compromise highly likely	Compromise likely	Compromise unlikely	Compromise highly unlikely
Very easy to exploit	Critical	Critical	High	Medium
Somewhat easy to exploit	Critical	High	Medium	Medium
Somewhat difficult to exploit	High	Medium	Medium	Low
Very difficult to exploit	Medium	Medium	Low	Low

Severity definitions

The severity labels are defined as follows.

Critical – The issue should be remediated immediately to avoid it escalating.

For example, an open S3 bucket is considered a critical severity finding. Because so many actors scan for open S3 buckets, data in exposed S3 buckets is likely to be discovered and accessed by others.

In general, resources that are publicly accessible are considered critical security issues. You should treat critical findings with the utmost urgency. You also should consider the criticality of the resource.

High – The issue must be addressed as a near-term priority.

For example, if a default VPC security group is open to inbound and outbound traffic, it is considered high severity. It is somewhat easy for a threat actor to compromise a VPC using this method. It is also likely that the threat actor will be able to disrupt or exfiltrate resources once they are in the VPC.

Security Hub recommends that you treat a high severity finding as a near-term priority. You should take immediate remediation steps. You also should consider the criticality of the resource.

Medium – The issue should be addressed as a mid-term priority.

For example, lack of encryption for data in transit is considered a medium severity finding. It requires a sophisticated man-in-the-middle attack to take advantage of this weakness. In other words, it is somewhat difficult. It is likely that some data will be compromised if the threat scenario is successful.

Security Hub recommends that you investigate the implicated resource at your earliest convenience. You also should consider the criticality of the resource.

Low – The issue does not require action on its own.

For example, failure to collect forensics information is considered low severity. This control can help to prevent future compromises, but the absence of forensics does not lead directly to a compromise.

You do not need to take immediate action on low severity findings, but they can provide context when you correlate them with other issues.

Informational – No configuration weakness was found.

In other words, the status is PASSED.

There is no recommended action. Informational findings help customers to demonstrate that they are in a compliant state.

Rules for updating control findings

If a subsequent check against a given rule generates a new result (for example, the status of "Avoid the use of the root account" changes from FAILED to PASSED), a new finding is generated that contains the most recent result.

If a subsequent check against a given rule generates a result that is identical to the current result, the existing finding is updated. No new finding is generated.

Security Hub automatically archives findings from controls if the associated resource is deleted, the resource does not exist, or the control is disabled. A resource might no longer exist because the associated service is not currently used. The findings are archived automatically based on one of the following criteria:

• The finding was not updated in three days.

• The associated AWS Config evaluation returned NOT_APPLICABLE.

Determining the overall status of a control from its findings

In the findings generated by security checks, the Compliance.Status field is assigned one of the following values.

• PASSED. If Compliance.Status is PASSED, then Security Hub automatically sets Workflow.Status to RESOLVED.

  If Compliance.Status for a finding changes from PASSED to either FAILED, WARNING, or NOT_AVAILABLE, and Workflow.Status was either NOTIFIED or RESOLVED, then Security Hub automatically sets Workflow.Status to NEW.

• FAILED

• WARNING – Indicates that the check was completed, but Security Hub cannot determine whether the resource is in a PASSED or FAILED state.

• NOT_AVAILABLE – Indicates that the check cannot be completed because a server failed, the resource was deleted, or the result of the AWS Config evaluation was NOT_APPLICABLE.

  If the AWS Config evaluation result was NOT_APPLICABLE, then Security Hub automatically archives the finding.

The overall status for each control is based on the value of Compliance.Status for the active findings for that control. For the master account, the overall status includes the active findings in the master account and the member accounts.

The overall status ignores findings that have a Workflow.Status of SUPPRESSED.

The available values for the overall status are as follows:

• Passed – Indicates that all findings have a Compliance.Status of PASSED.

• Failed – Indicates that at least one finding has a Compliance.Status of FAILED.

• Unknown – Indicates that at least one finding has a Compliance.Status of WARNING or NOT_AVAILABLE. No findings are FAILED.

• No data – Indicates that there are no findings for the control. For example, a new control has this status until it begins to generate findings. A control also has this status if all of the findings are SUPPRESSED.

Determining the security score for a security standard

On the Security standards page, each enabled standard displays a security score, which is between 0% and 100%. The standard details page also displays the overall security score.

How the security score is calculated

The security score represents the proportion of Passed controls to enabled controls. The score is displayed as a percentage. For example, if 10 controls are enabled for a standard, and seven of those controls are in a Passed state, then the security score is 70%.

The security score calculation does not include enabled controls that do not have any findings (overall status is No data). For example, a standard has 12 controls enabled. Six of those controls are in a Passed state. Two controls have no data. Because the calculation omits the controls without data, the security score is 60%.

Security score calculation for master accounts

For the master account, the score for a standard reflects both the master account and all of the member accounts. The calculation process for a master account is different from the process for a member account. This is to ensure that the score for the master account includes controls that are enabled for member accounts but disabled for the master account.

The process is as follows:

1. For each control, Security Hub gets all of the active findings across all of the accounts.

   Security Hub does not check whether the control is enabled or disabled in each account. If a control has active findings, then Security Hub treats it as enabled.

   If an account disabled a control within the previous 24 hours, then the score might include findings for that control.

2. Security Hub uses those findings to determine the overall status of each control across all of the accounts.

3. Security Hub uses those control statuses to determine the overall security score.

Security scores on the Summary page

On the Summary page, the Security standards card displays the security scores for each enabled standard. It also displays a consolidated security score that represents the proportion of passed controls to enabled controls across all of the enabled standards.