Ensure that Diagnostic Logs Are Enabled for All Services that Support it

Description:

Diagnostic 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, getting a secret from a Key Vault. Currently, 32 Azure resources support Diagnostic Logging (See the references 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. For example, Windows event system logs are a category of diagnostics logs
for VMs, and blob, table, and queue logs are categories of diagnostics logs for storage accounts.

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

Note: The CIS Benchmark covers some specific Diagnostic Logs
separately.

3.3 - Ensure Storage logging is enabled for Queue service for read, write, and delete requests

6.4 Ensure that Network Security Group Flow Log retention period is 'greater than 90 days'

Rationale:

A lack of Diagnostic Logs 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, Diagnostic Logs are not enabled by default. Specifically, without Diagnostic Logs it would be impossible to tell which entities had accessed a data store that
which was breached. In addition, alerts for failed attempts to access APIs for Web Services or Databases are only possible
when Diagnostic Logging is enabled.

Costs for Log Analytics Workspaces varies with Log Volume. Not every resource need to have logging enabled. It's 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.

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 logging at first, and then be more aggressive moving data to cold storage if the volume of data
becomes a cost concern.

From Azure Console
The specific steps for configuring resources within the Azure console vary depending on resource, but typically the steps are:

1. Go to the resource
2. Click on Diagnostic settings
3. In the blade that appears, click "Add diagnostic setting"
4. Configure the diagnostic settings
5. Click on Save

Using Azure Command Line Interface
Enable logging for all resources which support Diagnostic Logs to ensure interactions within the resource are logged and available. The skeleton command for creating
logs and metrics with unlimited retention on a
generic resource are shown below.

az monitor diagnostic-settings create --resource {ID} -n {name}
--event-hub-rule {eventHubRuleID} --storage-account {storageAccount}
--logs '[
{
"category": "WorkflowRuntime",
"enabled": true,
"retentionPolicy": {
"enabled": false,
"days": 0
}
}
]'
--metrics '[
{
"category": "WorkflowRuntime",
"enabled": true,
"retentionPolicy": {
"enabled": false,
"days": 0
}
}
]'
.