Augmenting VPN security with ZTNA tags

FortiGate's integration of ZTNA tags into the VPN infrastructure offers a powerful solution to enhance VPN security. ZTNA tags are a feature exclusively offered with the licensed FortiClient versions (FortiClient EMS). ZTNA tags are objects that are assigned to the FortiClient endpoints in real-time. ZTNA tags are used in the firewall policies on the FortiGate to allow or deny access to the VPN and network resources based on the organization’s security compliance regulations. These compliance regulations are enforced in real-time, thereby safeguarding the organization against constantly evolving security threats.

The following table compares features of the free VPN-only standalone FortiClient versus a licensed FortiClient managed by EMS for security compliance.

Feature	Free VPN-only standalone FortiClient	Licensed FortiClient
Basic VPN connection	Yes	Yes
Managed remote access profiles	No	Yes
Compliance using ZTNA tags: Allow or block VPN connections based on ZTNA security posture Per-firewall policy security posture checks using ZTNA tags	No	Yes

For more detailed information, see Feature comparison of FortiClient standalone and licensed versions in the FortiClient Administration Guide.

ZTNA tags

ZTNA tags (formerly FortiClient EMS tags in FortiOS 6.4 and earlier) are tags synchronized from FortiClient EMS as dynamic address objects on the FortiGate. FortiClient EMS uses zero-trust tagging rules to automatically tag managed endpoints based on various attributes detected by the FortiClient. When the FortiGate establishes a connection with the FortiClient EMS server through the EMS Fabric connector, it pulls zero-trust tags containing device IP and MAC addresses and converts them to read-only dynamic address objects. It also establishes a persistent WebSocket connection to monitor for changes in zero-trust tags, which keeps the device information current. These zero-trust tags can then be used in SSL VPN firewall rules to perform security posture checks to restrict or allow access to network resources, enabling role-based access control.

Security Fabric configuration

The FortiGate needs to be connected to FortiClient EMS in order to retrieve the ZTNA tags so they can be used in firewall policies. This is done by configuring a FortiClient EMS Security Fabric connector on the FortiGate to connect to FortiClient EMS. See Configuring FortiClient EMS for more information.

Creating ZTNA tags and ZTNA rules in FortiClient EMS

You can create, edit, and delete zero-trust tagging rules for endpoints. You can also view and manage the tags used to dynamically group endpoints.

The following process occurs when using zero-trust tagging rules with EMS and FortiClient:

1. EMS sends zero-trust tagging rules to endpoints through telemetry communication.

2. FortiClient checks endpoints using the provided rules and sends the results to EMS.

3. EMS receives the results from FortiClient.

4. EMS dynamically groups endpoints together using the tag configured for each rule. The dynamic endpoint groups can be viewed on the Zero Trust Tags > Zero Trust Tag Monitor page. See Zero Trust Tag Monitor in the FortiClient EMS Administration Guide for more information.

In this topic, two zero-trust tagging rule sets are created:

ZTNA tag	ZTNA tagging rule
AD-Joined	Apply if a remote user has OS version Windows 8.1 or Windows 10 and is a part of the AD group, FORTI-ARBUTUS.LOCAL/IT/IT.
Vulnerable	Apply if critical vulnerabilities are detected on a remote user.

These tags will be applied in two scenario examples (see Scenario 1 and Scenario 2). For more information about zero-trust tagging rule settings, see Adding a Zero Trust tagging rule set and Zero Trust tagging rule types in the FortiClient EMS Administration Guide.

To create a zero-trust tagging rule set in FortiClient EMS:

1. Go to Zero Trust Tags > Zero Trust Tagging Rules, and click Add.

2. Create the AD-Joined tagging rule set:

   1. In the Name field, enter AD-Joined.

   2. In the Tag Endpoint As dropdown list, enter AD-Joined and press Enter.

      EMS uses this tag to dynamically group together endpoints that satisfy the rule, as well as any other rules that are configured to use this tag.

   3. Toggle Enabled on to enable the rule.

   4. Configure the user in AD group rule:

      1. Click Add Rule.

      2. Set OS to Windows.

      3. Set the Rule Type to User in AD Group.

      4. Set the AD Group to FORTI-ARBUTUS.LOCAL/IT/IT.

      5. Click Save.

   5. Configure the OS rule:

      1. Click Add Rule.

      2. Set OS to Windows.

      3. Set the Rule Type to OS Version and select Windows 8.1.

      4. Click the + button and select Windows 10.

      5. Click Save.

   6. By default, an endpoint must satisfy all configured rules to be eligible for the rule set. You may want to apply the tag to endpoints that satisfy some, but not all, of the configured rules. In this example, you need to modify the rule set logic to apply the same tag to endpoints that fulfill one of the following criteria:

      • Running Windows 8.1 or 10

      • Is part of an AD group called FORTI-ARBUTUS.LOCAL/IT/IT

      With the default rule set logic, an endpoint would be eligible for the rule set if it is running Windows 8.1 or 10 and is part of an AD group called IT. To modify the rule set logic, do the following:

      1. Click Edit Logic.

      2. Clicking Edit Logic assigns numerical values to each configured rule. You can use and and or to define the rule logic. You cannot use not when defining the rule logic. You can also use parentheses to group rules.

         In the Rule Logic field, enter 1 and (2 or 3) to indicate that endpoints that satisfy that they are part of the AD IT group (rule 1) and Windows 8.1 (rule 2) or Windows 10 (rule 3) satisfy the rule set.

         

   7. Click Save.

3. Create the Vulnerable tagging rule set:

   1. Click Add.

   2. In the Name field, enter Vulnerable.

   3. In the Tag Endpoint As dropdown list, enter Vulnerable and press Enter.

   4. Toggle Enabled on to enable the rule.

   5. Configure the vulnerable devices rule:

      1. Click Add Rule.

      2. Set OS to Windows.

      3. Set the Rule Type to Vulnerable Devices.

      4. Set the Security Level to Critical.

      5. Click Save.

         

   6. Click Save.

For more information about editing, deleting, and importing ZTNA rules, see Zero Trust Tagging Rules in the FortiClient EMS Administration Guide.

Connecting FortiClient to FortiClient EMS using telemetry

After FortiClient software installation is complete on an endpoint, you can connect FortiClient to FortiClient EMS. Depending on the way the FortiClient installation is performed, you can either manually or automatically connect to FortiClient EMS, see Connecting FortiClient Telemetry after installation in the FortiClient Administration Guide for more details.

Once FortiClient connects to the FortiClient EMS, the Status shows up as Connected in the Zero Trust Telemetry tab.

After FortiClient telemetry connects to EMS, FortiClient endpoints receive the ZTNA tags if they satisfy any of the required ZTNA rules configured on the FortiClient EMS.

Monitoring ZTNA tags on the FortiGate

After the FortiGate is connected and authorized to and by FortiClient EMS, the ZTNA tags that were created in the zero-trust tagging rules are retrieved by the FortiGate. To view the tags on the FortiGate, go to Policy & Objects > ZTNA and select the ZTNA Tags tab.

If the tags are not visible on the FortiGate, ensure that the FortiClient EMS is configured to share tagging information with the FortiGate. See Configuring EMS to share tagging information with multiple FortiGates in the FortiClient EMS Administration Guide for more details.

Monitoring ZTNA tags in FortiClient and FortiClient EMS

To view the ZTNA tags assigned to the FortiClient endpoints by FortiClient EMS, click the user avatar and locate the Zero Trust Tags section. The following FortiClient endpoint is assigned the AD-Joined tag.

Ensure that the Show Zero Trust Tag on FortiClient GUI is enabled on FortiClient EMS (Endpoint Profiles > System Settings in the profile's Advanced view) so the tags are visible in FortiClient. See System Settings in the FortiClient EMS Administration Guide for more details.

ZTNA tags can also be monitored on FortiClient EMS from the endpoint's details in the Endpoints pane.

To view ZTNA tag information in the endpoint details:

1. Go to Endpoints, and select All Endpoints, a domain, or workgroup. The list of endpoints for the selected domain or workgroup displays.

2. Click an endpoint to display details about it in the content pane.

3. In the Summary pane, you can see the Zero Trust Tags associated with the endpoint. For example, this user has the AD-Joined and all_registered_clients tags.

   

   For detailed descriptions of the options in the Endpoints content pane, see Viewing the Endpoints pane in the FortiClient EMS Administration Guide.

Example: using ZTNA tags to augment VPN security

ZTNA tags can be used to augment VPN security using the following methods:

• Restrict an endpoint to connect to the VPN tunnel based on the ZTNA tag (see Scenario 1).

• Control access to network resources by allowing or denying traffic passing through the FortiGate using the IP/MAC Based Access Control field in the firewall policy (also known as ZTNA IP MAC based access control, see Scenario 2).

Both methods are demonstrated in the following example.

In this example, Off-net-Client is the FortiClient endpoint connected to and managed by FortiClient EMS. The telemetry traffic passes through the FortiGate using a virtual IP. The two ZTNA rule tagging sets configured previously (AD-Joined and Vulnerable) are applied.

Enterprise Core is the FortiGate that acts as the SSL VPN server. To configure SSL VPN, refer to SSL VPN and SSL VPN security best practices. Critical Assets are network resources that the off-net user tries to access after connecting to the VPN. SSL VPN is used in this example, but a similar configuration also applies to dialup IPsec VPN where the FortiGate acts as a dialup server.

Scenario 1: using ZTNA tags to restrict access to FortiClient endpoints connecting to the VPN

FortiClient endpoint profiles can be configured to allow or block an endpoint from connecting to a VPN tunnel based on its applied zero-trust tag. This feature is only available for Windows endpoints.

In this scenario, the endpoint profile is configured to prohibit the Off-net-Client (with a Windows OS) from connecting to the VPN if the endpoint has a Vulnerable ZTNA tag. The Vulnerable tag was configured previously (see Creating ZTNA tags and ZTNA rules in FortiClient EMS).

To configure the remote access profile in FortiClient EMS:

1. Go to Endpoint Profiles > Remote Access, and edit an existing profile or add a new one.

2. In the General section, enable Enable Secure Remote Access.

3. In the VPN Tunnels section, edit an existing VPN tunnel or add a new one.

4. Configure the following under Advanced Settings:

   1. For the Tag field, select Prohibit from the first dropdown.

   2. Select the Vulnerable tag from the second dropdown.

   3. Enable Customize Host Check Fail Warning.

   4. Enter a message to display to users when their connection to the VPN tunnel is prohibited due to critical vulnerabilities on their device.

   5. Configure the other VPN tunnel settings as needed.

      

   6. Click Save.

5. Configure the other remote access profile settings as needed.

6. Click Save.

   After the next communication between FortiClient EMS and FortiClient, endpoints with this profile applied are unable to connect to this VPN tunnel if they have critical vulnerabilities.

To verify the configuration using a vulnerable endpoint:

1. On the Off-net-Client endpoint, open FortiClient.

2. Click Vulnerability Scan, then click Scan Now to initiate a manual scan.

   Vulnerability scans can also be scheduled. See Vulnerability Scan in the FortiClient EMS Administration Guide for more details.

3. Wait a few minutes for the scan to complete.

4. Click the user avatar and locate the Zero Trust Tags section.

   FortiClient discovered the vulnerability and added a Vulnerable ZTNA tag in addition to the existing AD-Joined tag.

   

5. Click Remote Access to try to connect to the VPN:

   1. Enter the Username and Password.

   2. Click Connect.

      Based on the remote access profile configuration, the endpoint's access is denied due to the assigned Vulnerable ZTNA tag. The message configured in the remote access profile appears as a notification above the FortiTray icon (in the Windows system tray).

      

      See FortiTray in the FortiClient Administration Guide for more details about this icon.

Scenario 2: using ZTNA tags in firewall policies for role-based network access control

ZTNA tags are used in firewall policies to control access to network resources with the IP/MAC Based Access Control field.

In this scenario, if the Off-net Client is tagged with a Vulnerable tag, then it is not allowed to access Finance Server 1 (10.100.77.200). If the Off-net Client is tagged with an AD-Joined tag and no Vulnerable tag, then it is allowed to access Finance Server 1. Two firewall policies are configured as follows.

• Deny Vulnerable Endpoints: use IP/MAC based access control with the Vulnerable ZTNA IP tag to deny access.

• SSL VPN to DMZ: use IP/MAC based access control with the AD-Joined ZTNA IP tag to allow access.

These polices use a source address and group that have already be configured for SSL VPN users and authentication. See User groups for more information.

To configure the Deny Vulnerable Endpoints policy:

1. Go to Policy & Objects > Firewall Policy and click Create New.

2. Configure the following settings:

   Name	Deny Vulnerable Endpoints
   Type	Standard
   Incoming Interface	ssl.root
   Outgoing Interface	port2
   Source	SSLVPN_TUNNEL_ADDR1, AD-Joined VPN Users
   IP/MAC Based Access Control	Vulnerable
   Destination	DMZ Subnet
   Schedule	always
   Service	ALL
   Action	DENY
   Log Violation Traffic	Enable this setting.
   Enable this policy	Enable this setting.

3. Click OK.

To verify the configuration using an off-net client with a Vulnerable tag:

This verification assumes that a vulnerability scan was performed, the endpoint has a critical vulnerability, and the Vulnerable zero-trust tag was added.

1. On the endpoint, open FortiClient and click Remote Access to connect to the VPN:

   1. Enter the Username and Password.

   2. Click Connect.

2. Once the FortiClient endpoint is connected to the VPN, try to access Finance Server 1 using the web server. The connection times out because the traffic is denied by the firewall policy.

   

3. Verify the forward traffic log:

   1. In the GUI, go to Log & Report > Forward Traffic.

   2. In the CLI, enter the following:

      # execute log filter category 0
      # execute log filter field policyname "Deny Vulnerable Endpoints"
      # execute log display
      
      date=2023-10-24 time=17:04:19 eventtime=1698192258985043569 tz="-0700" logid="0000000013" type="traffic" subtype="forward" level="notice" vd="root" srcip=10.212.134.200 srcport=53801 srcintf="ssl.root" srcintfrole="undefined" dstip=10.100.77.200 dstport=80 dstintf="port2" dstintfrole="dmz" srcuuid="697b0036-37db-51ee-162f-5fed6735b06e" dstuuid="2e024fe8-57d2-51ee-73a7-63e15a078456" srccountry="Reserved" dstcountry="Reserved" sessionid=25809 proto=6 action="deny" policyid=3 policytype="policy" poluuid="c715492c-72aa-51ee-481d-09eef1adf713" policyname="Deny Vulnerable Endpoints" user="markgilbert" service="HTTP" trandisp="noop" duration=0 sentbyte=0 rcvdbyte=0 sentpkt=0 rcvdpkt=0 appcat="unscanned" crscore=30 craction=131072 crlevel="high"

To configure the SSL VPN to DMZ policy:

1. Go to Policy & Objects > Firewall Policy and click Create New.

2. Configure the following settings:

   Name	SSL VPN to DMZ
   Type	Standard
   Incoming Interface	ssl.root
   Outgoing Interface	port2
   Source	SSLVPN_TUNNEL_ADDR1, AD-Joined VPN Users
   IP/MAC Based Access Control	AD-Joined
   Destination	DMZ Subnet
   Schedule	always
   Service	ALL
   Action	ACCEPT
   Log Allowed Traffic	Enable this setting and select All Sessions.
   Enable this policy	Enable this setting.

3. Configure the other settings as needed.

4. Click OK.

To verify the configuration using an off-net client with an AD-Joined tag:

1. On the endpoint, open FortiClient, click Remote Access to connect to the VPN:

   1. Enter the Username and Password.

   2. Click Connect.

2. Once the FortiClient endpoint is connected to the VPN, try to access Finance Server 1 using the web server. The traffic is allowed by the firewall policy, and the server is accessible.

   

3. Verify the forward traffic log:

   1. In the GUI, go to Log & Report > Forward Traffic.

   2. In the CLI, enter the following:

      # execute log  filter category 0
      # execute log filter field policyname "SSL VPN to DMZ"
      # execute log display
      
      date=2023-10-24 time=14:07:05 eventtime=1698181625479969117 tz="-0700" logid="0000000020" type="traffic" subtype="forward" level="notice" vd="root" srcip=10.212.134.200 srcport=51841 srcintf="ssl.root" srcintfrole="undefined" dstip=10.100.77.200 dstport=80 dstintf="port2" dstintfrole="dmz" srcuuid="697b0036-37db-51ee-162f-5fed6735b06e" dstuuid="2e024fe8-57d2-51ee-73a7-63e15a078456" srccountry="Reserved" dstcountry="Reserved" sessionid=6656 proto=6 action="accept" policyid=5 policytype="policy" poluuid="8393d776-72b0-51ee-b955-68588258f38d" policyname="SSL VPN to DMZ" user="markgilbert" group="AD-Joined VPN Users" service="HTTP" trandisp="noop" duration=155 sentbyte=1196 rcvdbyte=1084 sentpkt=9 rcvdpkt=7 appcat="unscanned" sentdelta=1196 rcvddelta=1084 dstdevtype="Computer" dstosname="Debian" masterdstmac="02:09:0f:00:01:04" dstmac="02:09:0f:00:01:04" dstserver=0