5.4 Ensure that Azure Monitor Resource Logging is Enabled for All Services that Support it

Information

Resource Logs capture activity to the data access plane while the Activity log is a subscription-level log for the control plane. Resource-level diagnostic logs provide insight into operations that were performed within that resource itself; for example, reading or updating a secret from a Key Vault. Currently, 95 Azure resources support Azure Monitoring (See the more information section for a complete list), including Network Security Groups, Load Balancers, Key Vault, AD, Logic Apps, and CosmosDB. The content of these logs varies by resource type.

A number of back-end services were not configured to log and store Resource Logs for certain activities or for a sufficient length. It is crucial that monitoring is correctly configured to log all relevant activities and retain those logs for a sufficient length of time. Given that the mean time to detection in an enterprise is 240 days, a minimum retention period of two years is recommended.

Rationale:

A lack of monitoring reduces the visibility into the data plane, and therefore an organization's ability to detect reconnaissance, authorization attempts or other malicious activity. Unlike Activity Logs, Resource Logs are not enabled by default. Specifically, without monitoring it would be impossible to tell which entities had accessed a data store that was breached. In addition, alerts for failed attempts to access APIs for Web Services or Databases are only possible when logging is enabled.

Impact:

Costs for monitoring varies with Log Volume. Not every resource needs to have logging enabled. It is important to determine the security classification of the data being processed by the given resource and adjust the logging based on which events need to be tracked. This is typically determined by governance and compliance requirements.

NOTE: Nessus has not performed this check. Please review the benchmark to ensure target compliance.

Solution

Azure Subscriptions should log every access and operation for all resources.
Logs should be sent to Storage and a Log Analytics Workspace or equivalent third-party system. Logs should be kept in readily-accessible storage for a minimum of one year, and then moved to inexpensive cold storage for a duration of time as necessary. If retention policies are set but storing logs in a Storage Account is disabled (for example, if only Event Hubs or Log Analytics options are selected), the retention policies have no effect. Enable all monitoring at first, and then be more aggressive moving data to cold storage if the volume of data becomes a cost concern.
From Azure Portal
The specific steps for configuring resources within the Azure console vary depending on resource, but typically the steps are:

Go to the resource

Click on Diagnostic settings

In the blade that appears, click 'Add diagnostic setting'

Configure the diagnostic settings

Click on Save

From Azure CLI
For each resource, run the following making sure to use a resource appropriate JSON encoded category for the --logs option.

az monitor diagnostic-settings create --name <diagnostic settings name> --resource <resource ID> --logs '[{category:<resource specific category>,enabled:true,rentention-policy:{enabled:true,days:180}}]' --metrics '[{category:AllMetrics,enabled:true,retention-policy:{enabled:true,days:180}}]' <[--event-hub <event hub ID> --event-hub-rule <event hub auth rule ID> | --storage-account <storage account ID> |--workspace <log analytics workspace ID> | --marketplace-partner-id <full resource ID of third-party solution>]>

From PowerShell
Create the log settings object

$logSettings = @()
$logSettings += New-AzDiagnosticSettingLogSettingsObject -Enabled $true -RetentionPolicyDay 180 -RetentionPolicyEnabled $true -Category <resource specific category>
$logSettings += New-AzDiagnosticSettingLogSettingsObject -Enabled $true -RetentionPolicyDay 180 -RetentionPolicyEnabled $true -Category <resource specific category number 2>

Create the metric settings object

$metricSettings = @()
$metricSettings += New-AzDiagnosticSettingMetricSettingsObject -Enabled $true -RetentionPolicyDay 180 -RetentionPolicyEnabled $true -Category AllMetrics

Create the diagnostic setting for a specific resource

New-AzDiagnosticSetting -Name '<diagnostic settings name>' -ResourceId <resource ID> -Log $logSettings -Metric $metricSettings

Default Value:

By default, Azure Monitor Resource Logs are 'Disabled' for all resources.

See Also

https://workbench.cisecurity.org/benchmarks/12346