Device-ID Overview

Learn about Device-ID.
Whether or not your environment supports a “Bring Your Own Device” (BYOD) policy, you likely already have a large number of devices in your network; maybe even more than you realize. Combined with the need for scalability as the number of users and their accompanying devices on your network increases, not to mention the growing infrastructure of the Internet of Things (IoT), this presents a constantly growing area of risk with many possibilities for exploitation by malicious users. Additionally, once you identify these devices, how do you secure them from vulnerabilities such as outdated operating software? Using Device-ID™ on your firewall or to push policy from Panorama, you can get device context for events on your network, obtain policy rule recommendations for those devices, write policies based on devices, and enforce Security policy based on the recommendations.
Similar to how User-ID provides user-based policy and App-ID provides app-based policy, Device-ID provides policy rules that are based on a device, regardless of changes to its IP address or location. By providing traceability for devices and associating network events with specific devices, Device-ID allows you to gain context for how events relate to devices and write policies that are associated with devices, instead of users, locations, or IP addresses, which can change over time. You can use Device-ID in Security, Decryption, Quality of Service (QoS) and Authentication policies.
Device-ID requires an IoT Security license, a Cortex Data Lake (CDL) license, and the device certificate.
If you use PAN-OS version 8.1.0 through PAN-OS 9.1.x on a firewall, the IoT Security license provides device classification, behavior analysis, and threat analysis for your devices. If you use PAN-OS 10.0 or later, you can use Device-ID to obtain IP address-to-device mappings to view device context for network events, use IoT Security to obtain policy rule recommendations for these devices, and gain visibility for devices in reports and the ACC.
You can create a device-based Security policy on any Panorama or firewall that uses PAN-OS version 10.0 or later. To enforce the Security policy, the device must have a valid IoT Security license.
To identify and classify devices, the IoT Security app uses metadata from logs, network protocols, and sessions on the firewall. This does not include private or sensitive information or data that is not relevant for device identification. Metadata also forms the basis of the expected behavior for the device, which then establishes the criteria for the policy rule recommendation that defines what traffic and protocols to allow for that device.
When a firewall imports security policy rule recommendations and IP address-to-device mappings from IoT Security, the firewall sends its device certificate to an edge server to authenticate itself. The edge server authenticates itself to the firewall by sending its own certificate. The firewall uses Online Certificate Status Protocol (OCSP) to validate the server’s certificate by checking it against the following sites using HTTP on TCP port 80:
  • ocsp.int-x3.letsencrypt.org
  • isrg.trustid.ocscp.identrust.com
  • crl.identrust.com
Panorama performs the same check to validate the edge server’s certificate when Panorama imports policy rule recommendations from IoT Security.
After IoT Security identifies and classifies the devices in your network using the Palo Alto Networks firewalls already in your network, so you don’t have to implement new devices or third-party solutions, Device-ID can leverage this data to match devices with policy rules and provide device context for network events. Through the visibility that the firewall or Panorama provides for traffic, apps, users, devices, and threats, you can instantly trace network events back to individual devices and obtain Security policy rule recommendations for securing those devices.
All firewall platforms that support PAN-OS 10.0 support Device-ID and the IoT Security app with the exception of the VM-50 series, the VM-200, the CN series, and Prisma Access.
There are six levels of classification (also known as attributes) for devices:
Attribute
Example
Category
ATM Machine
;
3D Printer
Profile
Palo Alto Networks Device
Model
iPad
OS Version
iOS 9.9.3
OS Family
Android
;
iOS
Vendor
ASUS
;
Philips
To obtain policy rule recommendations for devices in your network, the firewall observes traffic to generate Enhanced Application logs (EALs). The firewall then forwards the EALs to the Cortex Data Lake (CDL) for processing. The IoT Security app on the hub receives logs from CDL for analysis, provides IP address-to-device mappings, and generates the latest policy rule recommendations for your devices. Using the IoT Security app, you can review these policy rule recommendations and create a Security policy for these devices. After you activate the policy rules in the IoT Security app, import them to the firewall or Panorama and commit your Security policy.
The firewall must be able to observe DHCP broadcast and unicast traffic on your network to identify devices. The more traffic the firewall can observe, the more accurate the policy rule recommendations are for the device and the more rapid and accurate the IP address-to-device mappings are for the device. When a device sends DHCP traffic to obtain an IP address, the firewall observes this type of request, it generates EALs to send to the Cortex Data Lake for processing and then analysis by IoT Security.
To observe traffic on an L2 interface, you must configure a VLAN for that interface. By allowing the firewall to treat the interface as an L3 interface for a DHCP relay, it can observe the DHCP broadcast traffic without impacting traffic or performance.
Because the firewall needs to both detect the devices based on their traffic and then enforce Security policy for those devices, the firewall acts as both a sensor to collect metadata from devices and an enforcer by enforcing your Security policy for the devices. The IoT Security app automatically detects new devices as soon as they send DHCP traffic and can identify 95% of devices within the first week.
Each application has an individual recommendation that you import to the firewall or Panorama as a rule. When you import the recommendation, the firewall or Panorama creates at least two objects to define the device behavior from the recommendation:
  • A source device object that identifies the device where the traffic originates
  • One or more destination objects that identify the permitted destinations for the traffic, which can be a device, IP address, or Fully Qualified Domain Name (FQDN)
If any of the device objects already exist on the firewall or Panorama, the firewall or Panorama updates the device object instead of creating a new device objects. You can use these device objects in Security, authentication, decryption, and Quality of Service (QoS) policies.
Additionally, the firewall assigns two tags to each rule:
  • One that identifies the source device, including the category (such as
    NetworkDevice - TrendNet
    ).
  • One that indicates that the rule is an IoT policy rule recommendation (
    IoTSecurityRecommended
    ).
Because the tags that the firewall assigns to the rule are the only way to restore your mappings if they become out of sync, do not edit or remove the tags.
For optimal deployment and operation of Device-ID, we recommend the following best practices:
  • Deploy Device-ID on firewalls that are centrally located in your network. For example, if you have a large environment, deploy Device-ID on a firewall that is upstream from the IP address management (IPAM) device. If you have a small environment, deploy Device-ID on a firewall that is acting as a DHCP server.
  • During initial deployment, allow Device-ID to collect metadata from your network for at least fourteen days. If devices are not active daily, the identification process may take longer.
  • Write device-based policy in order of your most critical devices to least critical. Prioritize by:
    1. Class (secure networked devices first)
    2. Critical devices (such as servers or MRI machines)
    3. Environment-specific devices (such as fire alarms and badge readers)
    4. Consumer-facing IoT devices (such as a smart watch or smart speaker)
  • Enable Device-ID on a per-zone basis for internal zones only.

Recommended For You