Decryption Policies

The following topics provide an overview of decryption policy creation, configuration, management, and logging.

About Decryption Policies

A decryption policy determines how the system handles encrypted traffic on your network. You can configure one or more decryption policies, associate a decryption policy with an access control policy, then deploy the access control policy to a managed device. When the device detects a TCP handshake, the access control policy first handles and inspects the traffic. If it subsequently identifies a TLS/SSL-encrypted session over the TCP connection, the decryption policy takes over, handling and decrypting the encrypted traffic.

Create a decryption policy using the wizard

You can use a wizard to create the following types of decryption policies:

  • Outbound protection (Decrypt - Resign rule action). If traffic matches this rule, the system re-signs the server certificate with the CA certificate, then acts as a man-in-the-middle.

    Three rules with a Do Not Decrypt action are added to the policy at the same time, saving you the trouble of doing it later. These rules correspond to any decryption exclusions you configure when you create the policy (for example, you can choose to bypass decryption for applications that are known to use certificate pinning.

    For more information, see Create a Decryption Policy with Outbound Connection Protection.

  • Inbound protection (Decrypt - Known Key rule action). You can associate one or more server certificates and paired private keys with the action. If traffic matches the rule, and the certificate used to encrypt the traffic matches the certificate associated with the action, the system uses the appropriate private key to obtain the session encryption and decryption keys.

    Three rules with a Do Not Decrypt action are added to the policy at the same time but these rules are disabled by default. These rules correspond to any decryption exclusions you configure when you create the policy (for example, you can choose to bypass decryption for applications that are known to use certificate pinning.

    For more information, see Create a Decryption Policy with Inbound Connection Protection.

  • Any other decryption rule action (such as blocking or monitoring).

    For more information, see Create a Decryption Policy with Other Rule Actions.

The wizard automatically creates a separate rule for each certificate you specify. For example, an inbound protection rule might specify one certificate for traffic coming to the Finance department internal network and a different certificate for traffic coming to the Engineering network.

The wizard creates additional rules for outbound and inbound protection policies as follows:

  • Outbound protection (Decrypt - Resign rule action): The wizard creates Do Not Decrypt rules for traffic that matches exceptions you specify in the wizard. For example, you can choose to not decrypt traffic from undecryptable applications, typically those using certificate pinning.

    The Do Not Decrypt rules are placed first in the decryption policy so that traffic is passed through the firewall with minimal processing.

  • Inbound protection (Decrypt - Known Key rule action): The wizard does not allow you to choose any exceptions but it adds the Do Not Decrypt rules to the policy and disables them; that way, you can enable those exceptions later if you need to.

Do Not Decrypt policy example

Following is an example decryption policy with a Do Not Decrypt rule action:

A Do Not Decrypt rule action passes the traffic through the firewall without trying to decrypt it; the access control policy then determines whether or not the traffic is inspected

The simplest decryption policy, as shown in the following diagram, directs the device where it is deployed to handle encrypted traffic with a single default action. You can set the default action to block decryptable traffic without further inspection, or to inspect undecrypted decryptable traffic with access control. The system can then either allow or block the encrypted traffic. If the device detects undecryptable traffic, it either blocks the traffic without further inspection or does not decrypt it, inspecting it with access control.

To get started, see Create a Decryption Policy

Requirements and Prerequisites for Decryption Policies

Supported Domains

Any

User Roles

  • Admin

  • Access Admin

  • Network Admin

Create a Decryption Policy

You can create any of the following types of decryption policies:

  • Outbound protection policy with rules that protect outbound connections; that is, the destination server is outside your protected network. This type of rule has a Decrypt - Resign rule action. We also create additional rules with a Do Not Decrypt action that excludes traffic you specify (such as traffic that uses certificate pinning.)

    See Create a Decryption Policy with Outbound Connection Protection

  • Inbound protection policy with a rule that protects inbound connections; that is, the destination server is inside your protected network. This type of rule has a Decrypt - Known Key rule action. We also create additional rules with a Do Not Decrypt action that excludes traffic you specify (such as traffic that uses certificate pinning.) These rules are disabled initially but you can modify and enable them later if you wish.

    See Create a Decryption Policy with Inbound Connection Protection

  • Other actions (including Do Not Decrypt, Block, and Block with Reset).

    See Create a Decryption Policy with Other Rule Actions

Create a Decryption Policy with Outbound Connection Protection

This task discusses how to create a decryption policy with a rule that protects outbound connections; that is, the destination server is outside your protected network. This type of rule has a Decrypt - Resign rule action.

When you create a decryption policy, you can create multiple rules at the same time, including multiple Decrypt - Known Key rules and multiple Decrypt - Resign rules.

If you enabled Change Management, you must create and assign a ticket before you can create a decryption policy. Before the decryption policy can be used, the ticket and all associated objects (like certificate authorities) must be approved. For more information, see Creating Change Management Tickets and Policies and Objects that Support Change Management.

Before you begin

You can optionally must upload an internal CA certificate for your managed device before you can create a decryption policy that protects outbound connections. You can do this in any of the following ways:

  • Create an internal CA certificate object by going to Objects > Object Management > PKI > Internal CAs and referring to PKI.

  • At the time you create this decryption policy.

Procedure

Step 1

Log in to Secure Firewall Management Center if you haven't already done so.

Step 2

Click Policies > Access Control > Decryption.

Step 3

Click Create Decryption Policy.

Step 4

Give the policy a unique Name and, optionally, a Description.

Step 5

Click the Outbound Connections tab.

Your decryption policy can cover outbound servers with a Decrypt - Resign rule or inbound servers with a Decrypt - Known Key rule. We create one decryption rule per combination of certificate and networks/ports, if any.

Step 6

From the Internal Certificates list, upload or choose certificates for the rules.

For more information about internal certificates, see Generate an Internal CA for Outbound Protection and Upload an Internal CA for Outbound Protection.

Step 7

(Optional.) Choose networks and ports.

For more information:

Step 8

Click Next.

Step 9

Continue with Decryption Policy Exclusions.

What to do next

Create a Decryption Policy with Inbound Connection Protection

This task discusses how to create a decryption policy with a rule that protects inbound connections; that is, the destination server is inside your protected network. This type of rule has a Decrypt - Known Key rule action.

When you create a decryption policy, you can create multiple rules at the same time, including multiple Decrypt - Known Key rules and multiple Decrypt - Resign rules.

Before you begin

You can optionally must upload an internal certificate for your internal server before you can create a decryption policy that protects inbound connections. You can do this in any of the following ways:

  • Create an internal certificate object by going to Objects > Object Management > PKI > Internal Certs and referring to PKI.

  • At the time you create this decryption policy.

If you enabled Change Management, you must create and assign a ticket before you can create a decryption policy. Before the decryption policy can be used, the ticket and all associated objects (like certificate authorities) must be approved. For more information, see Creating Change Management Tickets and Policies and Objects that Support Change Management.

Procedure

Step 1

Log in to Secure Firewall Management Center if you haven't already done so.

Step 2

Click Policies > Access Control > Decryption.

Step 3

Click Create Decryption Policy.

Step 4

Give the policy a unique Name and, optionally, a Description.

Step 5

From the Internal CA list, upload or choose certificates for the rules.

For more information about internal CA certificates, see Internal Certificate Authority Objects.

Step 6

(Optional.) Choose networks and ports.

For more information:

Step 7

Click the Inbound Connections tab.

Your decryption policy can cover outbound servers with a Decrypt - Resign rule or inbound servers with a Decrypt - Known Key rule.

Step 8

Click Next.

Step 9

Continue with Decryption Policy Exclusions.

Decryption Policy Exclusions

This task discusses how to exclude from decryption certain types of traffic. We create Do Not Decrypt rules in your decryption policy for these although the rules are initially enabled only for an outbound decryption policy (that is, one that uses the Decrypt - Resign policy action).

Before you begin

You can optionally must upload an internal CA certificate for your managed device before you can create a decryption policy that protects outbound connections. You can do this in any of the following ways:

  • Create an internal CA certificate object by going to Objects > Object Management > PKI > Internal CAs and referring to PKI.

  • At the time you create this decryption policy.

Procedure

Step 1

Complete the tasks discussed in:

Step 2

The exclusions page provides the following options. All options are enabled for an outbound protection policy (Decrypt - Resign rule action) and disabled for all other decryption policy actions.

Item

Description

Bypass decryption for sensitive URL categories

Check the box to not decrypt traffic from the indicated categories. Depending on the laws in your area, decryption certain traffic, such as finance or health-related, might be prohibited. Consult an authority in your area for more information.

Click Add to add more categories.

Click Delete (delete icon) to remove categories.

Bypass decryption for undecryptable distinguished names

Check the box to not decrypt traffic when re-signing the certificate is likely to cause the connection to fail. Typically, this behavior is associated with certificate pinning, which is discussed in TLS/SSL Certificate Pinning Guidelines.

The list of undecryptable distinguished names is maintained by Cisco.

Bypass decryption for undecryptable applications

Check the box to not decrypt traffic when re-signing the certificate is likely to cause the connection to fail. Typically, this behavior is associated with certificate pinning, which is discussed in TLS/SSL Certificate Pinning Guidelines.

Undecryptable applications are updated automatically in the Vulnerability Database (VDB). You can find a list of all applications on the Secure Firewall Application Detectors page; the undecryptable tag identifies applications Cisco determines are undecryptable.

The list of undecryptable applications is maintained by Cisco.

The following figure shows default options.

For an outbound decryption policy, you can specify which (if any) types of traffic to exclude from decryption. For example, you can exclude traffic going to sensitive website categories such as finance or health.

For an outbound decryption policy, you can specify which (if any) types of traffic to exclude from decryption. For example, you can exclude traffic going to sensitive website categories such as finance or health.

Step 3

Click Create Policy.

The following figure shows a sample outbound protection policy.

This sample decryption policy has one Decrypt - Resign rule applied to traffic that matches the networks and ports chosen. It decrypts traffic and subsequently re-encrypts it by signing with the internal certificate object you specified. We add Do Not Decrypt rules for traffic you specified to be excluded from decryption.

In the preceding example, the Do Not Decrypt rules corresponding to your choices for rule exclusions are automatically added before the Decrypt - Resign rule. The rule for sensitive URL categories is disabled because, by default, that exclusion is disabled. Had you selected the Bypass decryption for sensitive URL categories check box, the rule would have been enabled.

Step 4

Click Create Policy.

What to do next

Generate an Internal CA for Outbound Protection

This task discusses how you can optionally generate an internal certificate authority when you create a decryption rule that protects outbound connections. You can also perform these tasks using Objects > Object Management as discussed in Uploading a Signed Certificate Issued in Response to a CSR.

Before you begin

Make sure you understand the requirements for generating an internal certificate authority object as discussed in Internal Certificate Authority Objects.

Procedure

Step 1

Log in to Secure Firewall Management Center if you haven't already done so.

Step 2

Click Policies > Access Control > Decryption.

Step 3

Click Create Decryption Policy.

Step 4

Enter a name for the policy in the Name field and an optional description in the Description field.

Step 5

Click the Outbound Connections tab.

Step 6

From the Internal CA list, click Create New > Generate CA.

Step 7

Give the internal CA a Name and provide a two-letter Country Name.

Step 8

Click Self-Signed or CSR.

For more information about these options, see Internal Certificate Authority Objects.

Step 9

Enter the requested information in the provided fields.

Step 10

Click Save.

Step 11

If you chose CSR, after the signing request has been completed, click Install Certificate as follows:

  1. Repeat the preceding steps in this procedure.

  2. Edit the CA from the Internal CA list as follows.

    To install an internal CA from a certificate signing request, click Edit next to the name of the internal CA

  3. Click Install Certificate.

  4. Follow the prompts on your screen to complete the task.

Step 12

Continue creating the policy as discussed in Create a Decryption Policy with Inbound Connection Protection.

Upload an Internal CA for Outbound Protection

This task discusses how you can optionally upload an internal certificate authority when you create a decryption rule that protects outbound connections. You can also perform these tasks using Objects > Object Management as discussed in Uploading a Signed Certificate Issued in Response to a CSR.

Before you begin

Make sure you understand the requirements for generating an internal certificate authority object as discussed in Internal Certificate Authority Objects.

Procedure

Step 1

Log in to Secure Firewall Management Center if you haven't already done so.

Step 2

Click Policies > Access Control > Decryption.

Step 3

Click Create Decryption Policy.

Step 4

Enter a name for the policy in the Name field and an optional description in the Description field.

Step 5

Click the Outbound Connections tab.

Step 6

From the Internal CA list, click Create New > Upload CA.

Step 7

Give the internal CA a Name.

Step 8

Paste or browse to locate the certificate and its private key in the provided fields.

Step 9

If the CA has a password, select the Encrypted check box and enter the password in the adjacent field.

Step 10

Continue creating the policy as discussed in Create a Decryption Policy with Outbound Connection Protection.

Upload an Internal Certificate for Inbound Protection

This task discusses how to upload an internal certificate authority when you create a decryption rule that protects outbound connections. You can also upload the internal CA using Objects > Object Management as discussed in Importing a CA Certificate and Private Key.

Before you begin

Make sure you have an internal certificate authority in one of the formats discussed in Internal Certificate Authority Objects.

Procedure

Step 1

Log in to Secure Firewall Management Center if you haven't already done so.

Step 2

Click Policies > Access Control > Decryption.

Step 3

Click Create Decryption Policy.

Step 4

Enter a name for the policy in the Name field and an optional description in the Description field.

Step 5

Click the Inbound Connections tab.

Step 6

From the Internal Certificates list, click Add (add icon).

Step 7

Click Upload.

Step 8

Give the internal CA a Name.

Step 9

Paste or browse to locate the certificate and its private key in the provided fields.

Step 10

If the certificate has a password, select the Encrypted check box and enter the password in the adjacent field.

Step 11

Continue creating the decryption policy as discussed in Create a Decryption Policy with Inbound Connection Protection.

Create a Decryption Policy with Other Rule Actions

To create a decryption rule with a Do Not Decrypt, Block, Block With Reset, or Monitor rule action, create a decryption policy and edit the policy to add the rule.

When you create a decryption policy, you can create multiple rules at the same time, including multiple Decrypt - Known Key rules and multiple Decrypt - Resign rules.

If you enabled Change Management, you must create and assign a ticket before you can create a decryption policy. Before the decryption policy can be used, the ticket and all associated objects (like certificate authorities) must be approved. For more information, see Creating Change Management Tickets and Policies and Objects that Support Change Management.

Procedure

Step 1

Log in to Secure Firewall Management Center if you haven't already done so.

Step 2

Click Policies > Access Control > Decryption.

Step 3

Give the policy a unique Name and, optionally, a Description.

Step 4

Click Next.

Step 5

The bypass page is provided for your information only; you cannot bypass traffic for other types of decryption (such as Block).

Step 6

Click Create Policy.

Step 7

Wait for the policy to be created.

Step 8

Click Edit (edit icon) next to the decryption policy name.

Step 9

Click Add Rule.

Step 10

Give the rule a Name.

Step 11

From the Action list, click a rule action and see one of the following sections for more information:

Step 12

Click Save.

What to do next

Decryption Policy Default Actions

The default action for a decryption policy determines how the system handles decryptable encrypted traffic that does not match any non-monitor rule in the policy. When you deploy a decryption policy that does not contain any decryption rules, the default action determines how all decryptable traffic on your network is handled. Note that the system does not perform any kind of inspection on encrypted traffic blocked by the default action.

To set the decryption policy default action:

  1. Log in to the Secure Firewall Management Center if you haven't already done so.

  2. Click Policies > Access Control > Decryption.

  3. Click Edit (edit icon) next to the name of the decryption policy.

  4. In the Default Action row, click one of the following actions from the list.

Table 1. Decryption Policy Default Actions

Default Action

Effect on Encrypted Traffic

Block

Block the TLS/SSL session without further inspection.

Block with reset

Block the TLS/SSL session without further inspection and reset the TCP connection. Choose this option if traffic uses a connectionless protocol like UDP. In that case, the connectionless protocol tries to reestablish the connection until it is reset.

This action also displays a connection reset error in the browser so the user is informed that the connection is blocked.

Do not decrypt

Inspect the encrypted traffic with access control.

Default Handling Options for Undecryptable Traffic

Table 2. Undecryptable Traffic Types

Type

Description

Default Action

Available Action

Compressed Session

The TLS/SSL session applies a data compression method.

Inherit default action

Do not decrypt

Block

Block with reset

Inherit default action

SSLv2 Session

The session is encrypted with SSL version 2.

Note that traffic is decryptable if the ClientHello message is SSL 2.0, and the remainder of the transmitted traffic is SSL 3.0.

Inherit default action

Do not decrypt

Block

Block with reset

Inherit default action

Unknown Cipher Suite

The system does not recognize the cipher suite.

Inherit default action

Do not decrypt

Block

Block with reset

Inherit default action

Unsupported Cipher Suite

The system does not support decryption based on the detected cipher suite.

Inherit default action

Do not decrypt

Block

Block with reset

Inherit default action

Session not cached

The TLS/SSL session has session reuse enabled, the client and server reestablished the session with the session identifier, and the system did not cache that session identifier.

Inherit default action

Do not decrypt

Block

Block with reset

Inherit default action

Handshake Errors

An error occurred during TLS/SSL handshake negotiation.

Inherit default action

Do not decrypt

Block

Block with reset

Inherit default action

Decryption Errors

An error occurred during traffic decryption.

Block

Block

Block with Reset

When you first create a decryption policy, logging connections that are handled by the default action is disabled by default. Because the logging settings for the default action also apply to undecryptable traffic handling, logging connections handled by the undecryptable traffic actions is disabled by default.

Note that if your browser uses certificate pinning to verify a server certificate, you cannot decrypt this traffic by re-signing the server certificate. For more information, see Decryption Rule Guidelines and Limitations.

Set Default Handling for Undecryptable Traffic

You can set undecryptable traffic actions at the decryption policy level to handle certain types of encrypted traffic the system cannot decrypt or inspect. When you deploy a decryption policy that contains no decryption rules, the undecryptable traffic actions determine how all undecryptable encrypted traffic on your network is handled.

Depending on the type of undecryptable traffic, you can choose to:

  • Block the connection.

  • Block the connection, then reset it. This option is preferrable for connectionless protocols like UDP, which keep trying to connect until the connection is blocked.

  • Inspect the encrypted traffic with access control.

  • Inherit the default action from the decryption policy.

Procedure

Step 1

Log in to Secure Firewall Management Center if you haven't already done so.

Step 2

Click Policies > Access Control > Decryption.

Step 3

Click Edit (edit icon) next to the name of the decryption policy.

Step 4

In the decryption policy editor, click Undecryptable Actions.

Step 5

For each field, choose either the decryption policy's default action or another action you want to take on the type of undecryptable traffic. See Default Handling Options for Undecryptable Traffic and Decryption Policy Default Actions for more information.

Step 6

Click Save to save the policy.

What to do next

Decryption Policy Advanced Options

A decryption policy 's Advanced Settings page has global settings that are applied to all managed devices that are configured for Snort 3 to which the policy is applied.

A decryption policy advanced settings are all ignored on any managed device that runs:

  • A version earlier than 7.1

  • Snort 2

Block flows requesting ESNI

Encrypted Server Name Indication (ESNI (link to draft proposal)) is a way for a client to tell a TLS 1.3 server what the client is requesting. Because the SNI is encrypted, you can optionally block these connections because the system cannot determine what the server is.

Disable HTTP/3 advertisement

This option strips HTTP/3 (RFC 9114) from the ClientHello in TCP connections. HTTP/3 is part of the QUIC transport protocol, not the TCP transport protocol. Blocking clients from advertising HTTP/3 provides protection against attacks and evasion attempts potentially burried within QUIC connections.

Propagate untrusted server certificates to clients

This applies only to traffic matching a Decrypt - Resign rule action.

Enable this option to substitute the certificate authority (CA) on the managed device for the server's certificate in cases where the server certificate is untrusted. An untrusted server certificate is one that is not listed as a trusted CA in the Secure Firewall Management Center. (Objects > Object Management > PKI > Trusted CAs).

Enable TLS 1.3 Decryption

Whether to apply decryption rules to TLS 1.3 connections. If you do not enable this option, the decryption rules apply to TLS 1.2 or lower traffic only. See TLS 1.3 Decryption Best Practices.

Enable adaptive TLS server identity probe

Automatically enabled when TLS 1.3 decryption is enabled. A probe is a partial TLS connection with the server, the purpose of which is to obtain the server certificate and cache it. (If the certificate is already cached, the probe is never established.)

If TLS 1.3 Server Identity Discovery is disabled on the access control policy with which the decryption policy is associated, we attempt to use the Server Name Indication (SNI), which is not as reliable.

The adaptive TLS server identity probe occurs on any of the following conditions as opposed to on every connection as in earlier releases:


Note

Enable adaptive TLS server discovery mode is not supported on any Secure Firewall Threat Defense Virtual deployed to AWS. If you have any such managed devices managed by the Secure Firewall Management Center, the connection event PROBE_FLOW_DROP_BYPASS_PROXY increments every time the device attempts to extract the server certificate.

QUIC Decryption

Whether to apply decryption rules to connections that use the HTTP/3 over the QUIC protocol. When you decrypt QUIC connections, the system can inspect the contents of the sessions for intrusions, malware, or other issues. You can also apply granular control and filtering of decrypted QUIC connections based on specific criteria in the access control policy. QUIC support is in line with RFC 9000, 9001, 9002, 9114, 9204.

Consider the following when implementing QUIC decryption:

  • QUIC decryption is not supported on high availability or clustered devices. Multi-instance is supported.

  • Rules that apply to QUIC traffic would include the UDP protocol with destination port 443.

  • Access control rules that apply to QUIC traffic would include the HTTP/3 or QUIC protocols, either explicitly or by implication.

The following limitations apply to QUIC decryption:

  • QUIC decryption applies to Threat Defense 7.6+ only. Devices running a lower release cannot decrypt QUIC connections.

  • Connections from browsers using the Chromium stack (Google Chrome, Opera, Edge) cannot be decrypted for outbound traffic. But inbound traffic from the same browsers can be decrypted.

  • Connection Migration as described in RFC 9000 is not supported. The concept of Connection ID in QUIC allows endpoints to retain the same connection in the event of address change.

  • Key update, session resumption, and QUIC version 2 are not supported.

  • Interactive Block and Interactive Block with Reset (in access control rules) is not supported. These actions will work as Block and Block with Reset.

  • The active connection-ID per connection is limited to 5. The maximum stream support per connection is limited to 25. If necessary, you can modify these limits using the system support quic-tuning and system support quic-tuning-reset commands in the device CLI.

TLS 1.3 Decryption Best Practices

Recommendation: When to enable advanced options

Both the decryption policy and the access control policy have advanced options that affect how traffic is handled, whether the traffic is being decrypted or not.

The advanced options are:

  • Decryption policy:

    • TLS 1.3 decryption

    • TLS adaptive server identity probe

  • Access control policy: TLS 1.3 Server Identity Discovery

    The access control policy setting takes precedence over the decryption policy setting.

Use the following table to decide which option to enable:

TLS adaptive server identity probe setting (decryption policy)

TLS 1.3 Server Identity Discovery setting (access control policy)

Result

Recommended when

Enabled

Disabled

Adaptive probe sent if decryption policy contains any rule conditions specified in Decryption Policy Advanced Options and if the server certificate is not cached.

  • You're not using application or URL conditions in access control rules

  • You're decrypting traffic

Enabled

Enabled

Probe is always sent if the server certificate is not cached.

Use only if your access control rules have URL or application conditions

Disabled

Enabled

Probe is always sent if the server certificate is not cached.

Not recommended.

Disabled

Disabled

Probe is never sent.

Very limited usefulness; use only if not decrypting traffic and not using application or URL conditions in the access control rule


Note

A cached TLS server's certificate is available to all Snort instances on a particular threat defense. The cache can be cleared with a CLI command and is automatically cleared when the device is rebooted.

Reference

For more information, see the discussion of TLS server identity discovery on secure.cisco.com.