Azure ATP Architecture
Azure ATP monitors your domain controllers by capturing and parsing network traffic and leveraging Windows events directly from your domain controllers, then analyzes the data for attacks and threats. Utilizing profiling, deterministic detection, machine learning, and behavioral algorithms Azure ATP learns about your network, enables detection of anomalies, and warns you of suspicious activities.
Azure Advanced Threat Protection architecture:
This section describes how the flow of Azure ATP's network and event capturing works, and drills down to describe the functionality of the main components: the Azure ATP portal, Azure ATP sensor, and Azure ATP cloud service.
Installed directly on your domain controllers, the Azure ATP sensor accesses the event logs it requires directly from the domain controller. After the logs and network traffic are parsed by the sensor, Azure ATP sends only the parsed information to the Azure ATP cloud service (only a percentage of the logs are sent).
Azure ATP Components
Azure ATP consists of the following components:
Azure ATP portal
The Azure ATP portal allows creation of your Azure ATP instance, displays the data received from Azure ATP sensors, and enables you to monitor, manage, and investigate threats in your network environment.
Azure ATP sensor
Azure ATP sensors are installed directly on your domain controllers. The sensor directly monitors domain controller traffic, without the need for a dedicated server, or configuration of port mirroring.
Azure ATP cloud service
Azure ATP cloud service runs on Azure infrastructure and is currently deployed in the US, Europe, and Asia. Azure ATP cloud service is connected to Microsoft's intelligent security graph.
Azure ATP portal
Use the Azure ATP portal to:
- Create your Azure ATP instance
- Integrate with other Microsoft security services
- Manage Azure ATP sensor configuration settings
- View data received from Azure ATP sensors
- Monitor detected suspicious activities and suspected attacks based on the attack kill chain model
- Optional: the portal can also be configured to send emails and events when security alerts or health issues are detected
- If no sensor is installed on your Azure ATP instance within 60 days, the instance may be deleted and you’ll need to recreate it.
Azure ATP sensor
The Azure ATP sensor has the following core functionality:
- Capture and inspect domain controller network traffic (local traffic of the domain controller)
- Receive Windows Events directly from the domain controllers
- Receive RADIUS accounting information from your VPN provider
- Retrieve data about users and computers from the Active Directory domain
- Perform resolution of network entities (users, groups, and computers)
- Transfer relevant data to the Azure ATP cloud service
Azure ATP Sensor features
Azure ATP sensor reads events locally, without the need to purchase and maintain additional hardware or configurations. The Azure ATP sensor also supports Event Tracing for Windows (ETW) which provides the log information for multiple detections. ETW-based detections include Suspected DCShadow attacks attempted using domain controller replication requests and domain controller promotion.
Domain synchronizer process
The domain synchronizer process is responsible for synchronizing all entities from a specific Active Directory domain proactively (similar to the mechanism used by the domain controllers themselves for replication). One sensor is automatically chosen at random from all of your eligible sensors to serve as the domain synchronizer.
If the domain synchronizer is offline for more than 30 minutes, another sensor is automatically chosen instead.
The Azure ATP sensor includes a monitoring component that evaluates the available compute and memory capacity on the domain controller on which it's running. The monitoring process runs every 10 seconds and dynamically updates the CPU and memory utilization quota on the Azure ATP sensor process. The monitoring process makes sure the domain controller always has at least 15% of free compute and memory resources available.
No matter what occurs on the domain controller, the monitoring process continually frees up resources to make sure the domain controller's core functionality is never affected.
If the monitoring process causes the Azure ATP sensor to run out of resources, only partial traffic is monitored and the monitoring alert "Dropped port mirrored network traffic" appears in the Azure ATP portal Health page.
To enhance Azure ATP detection coverage of suspected identity theft (pass-the-hash), suspicious authentication failures, modifications to sensitive groups, creation of suspicious services, and Honeytoken activity types of attack, Azure ATP needs to analyze the logs of the following Windows events: 4776,4732,4733,4728,4729,4756,4757, and 7045. These events are read automatically by Azure ATP sensors with correct advanced audit policy settings.