Create a policy

While SIA includes a default policy, you can configure a policy to define how your company handles known or suspected threats.

As a best practice, assign Security Connector to the malware and command and control (C&C) categories. A C&C threat indicates that a user's machine is already compromised by the time it's detected. To clean compromised machines, you can use Security Connector to identify infected machines and get the information you need for remediation.

To perform this task, you need to be an SIA administrator.

📘
A strict delegated administrator cannot enable the SIA proxy and complete any step related to the proxy.

To create a policy:

In the Threat Protection menu of Enterprise Center, select Policies > Policies.
On the Policies page, click the plus sign icon.
Complete the Name and Description fields.
In the Policy Type menu, select one one of these options:
- DNS Only. Enables only DNS protection in the policy.
- DNS + Proxy. Enables DNS Protection and SIA Proxy. If you would like to enable the proxy, make sure you select this option.
To use a predefined template, select one of these templates and click Continue:
- Strict. Contains settings that block known and most suspected threat categories. Select this template to apply settings that are a best practice for a policy.
- Monitor-only. Logs and reports threats but it does not block them. This template is ideal for testing or assessing policy impact before using the Strict template.
- Custom. Lets you define policy actions for known and suspected threats.
To assign a location or sub-location, click the link icon for locations or sub-locations, and select one or more. Then click Associate.
Click the Settings tab.
In the Proxy Settings section, complete the steps for these fields:
1. Proxy Authorization. Toggle on to require that SIA Proxy authorizes connections from the on-premises proxy. To use this setting, you need to configure proxy credentials in SIA and in the on-premises proxy. For more information, see Configure proxy authorization.
2. Origin Ports. If you want to allow outbound traffic on a new origin port, enter the port number or port range. Separate each port number or range with a comma. By default, the full web proxy allows outbound traffic to ports 80 to 84, 443, 4443, 8080, 8443, and 8888.
3. Trust XFF Header. Toggle on if you are configuring proxy chaining or the full web proxy. Your organization needs to be licensed for SIA Advanced Threat.
4. Proxy Logging Mode. To change the SIA Proxy logging mode, select a different level. The default Level 1 ensures that detailed data is logged, such as response or request headers in HTTP or HTTPS threat events. For more information, see Proxy logging mode.
5. Bypass Microsoft 365 Traffic. Toggle on to bypass traffic to Microsoft 365 apps and services.
6. Block Incompatible Domains. Toggle on to block domains that are not compatible with TLS encryption. Otherwise, these domains bypass SIA Proxy.
7. Invalid Certificate Response. Select Block - Error Page to block a request if SIA Proxy cannot verify a website's origin certificate. Otherwise, select Bypass to bypass SIA Proxy.
8. Local Breakout for Bypass Domains. Disable this option only if your network has no default route to the Internet, and it cannot directly access origins that are configured for bypass.
In the Payload Analysis section, enable inline payload analysis to scan files that are up to 5 MB before they are downloaded. You need to enable SIA Proxy to use payload analysis. Complete the steps for these fields:
1. Block Unscannable Files. Toggle on if you want to block files that cannot be scanned with SIA Proxy as part of inline payload analysis.
2. Block On Upload Scan Timeout. Toggle on if you want to block requests that cause scanning to take longer than expected. Note: This setting applies to DLP and File Type blocking.
3. Risky File Handling - by file size. If your organization is enabled for Advanced Sandbox:
  - For downloads that range from 5 MB to 2 GB in size, select Allow or Allow and Scan. Otherwise, select Block - Error Page. For more information, see Static malware analysis of large files.
  - If you select Allow and Scan, the Dynamic Analysis toggle is available. To enable dynamic analysis, toggle this setting to on. For more information, see Dynamic malware analysis.
  - For files that are greater than 2 GB (huge files), select an action. You can select Block - Error Page or Allow. For more information, see Payload analysis.
In the Browsing Restrictions section, complete the steps for these fields:
1. Safe Search. Toggle on to block explicit results from Google and Bing searches.
2. YouTube. Select Strict or Moderate to enable YouTube Restricted Mode. Otherwise, select Unrestricted to allow unrestricted access to YouTube content.
In the Other Settings section, complete the steps for these fields:
1. Forward Public IP to Origin. Toggle on to forward the user's public IP address to authoritative DNS servers and web servers. This setting identifies the geolocation of clients. Make sure you also enable this setting if you enabled the Bypass Microsoft 365 Traffic option.
2. Enable IP Intelligence. Make sure this setting is enabled to use IP Intelligence. IP Intelligence detects threats based on IP address. This setting is enabled by default. However, you can disable it if you prefer to not use this feature.
3. Authentication Mode. Select Require to require authentication, Optional to give users the option to skip authentication, or None. This mode defines whether users are prompted to authenticate when accessing allowed websites or web applications.
4. Identity Provider. Select an IdP if you selected Require or Optional as an authentication mode.
If you've installed ETP Client or Zero Trust Client on devices in your network, you can complete the steps for these fields:
1. Disable Client. Enable this option to disable the client in the locations that are associated with the policy.
2. Avoid Local DNS Resolvers. Enable this option to have the client query the local DNS resolver only for domains that are configured on the Local Bypass Settings page. All other traffic is directed to SIA instead of the local resolver. To configure the domains that you prefer bypass SIA, see Configure local bypass settings.
📘
It's recommended that you don’t enable this setting when the client is on the network. When the client is on the network, local traffic should be directed to the local resolvers.
1. Walled Garden Exceptions. Enable this option to block all traffic when the client is in an unprotected state. Only the domains and IP addresses that are specified as walled garden exceptions in the network configuration are allowed when walled garden is enabled. This setting also makes the client the device web proxy. As a result, Yes is automatically selected for Overwrite Device Proxy Settings. To configure the domains and IP addresses that are exceptions to a walled garden, see Configure walled garden exceptions.
2. Overwrite Device Proxy Settings. Select Yes or Only if there's no local proxy if you want to enable ETP Client or Zero Trust Client as a proxy on the client computer or device. Otherwise, select No.
3. DNS-over-TLS Mode. Defines whether the client uses DNS over TLS (DoT) to protect DNS traffic it forwards to SIA. Select one of these modes:
  - Attempt. Indicates the client always attempts to use DoT. If DoT is not available, the client falls back to plain DNS.
  - Required. Indicates that DoT is required. If DoT is not available, DNS traffic is directed from the client to the local DNS resolver.
  - Disabled. Indicates that DoT is not used to secure DNS traffic from the client.
4. DNS-over-TLS Port. Select the port for DoT connections.
5. Support multiple users per device. If this feature is available for your organization, you can allow multiple users to use the same device. After a user authenticates from the client and provides their Login Portal credentials, a logout button is available for the user to log out of their account. This setting is supported on Zero Trust Client 6.1 with the Threat Protection service only. It’s also supported on ETP Client 4.4 or later when the client is on a Windows device. For more information about this setting, see Policy settings.
Define policy actions for a threat category. Click the Threat tab and complete the action based on threat type:
1. Known. If you want to assign the same policy action to all known threat categories, select an action in the Action column. Otherwise, make sure the Known option is expanded to show the threat categories.
  - For each threat category, select an action. For more information, see Policy actions.
  - If you select Block, select a specific response to the user. The Response to User column is available when the Block action is selected.
  - If Error Page is selected and you want to direct traffic to Security Connector, select a security connector in the Security Connector field. Otherwise, select None.
2. Suspected. If you want to assign the same policy action to all suspected threat categories, select an action in the Action column. Otherwise, make sure the Suspected option is expanded to show the threat categories and complete the fields as described in the previous step.
To send alerts for a threat, toggle Alerts to on.
Click the Access Control tab and complete these steps:
1. Click the AUP & Shadow IT tab and complete the steps described in Configure application visibility and control.
2. Click the DLP tab and complete the steps described in Select user and group exceptions for DLP scanning and Assign a DLP dictionary to a policy.
3. If you want to block or monitor the download or upload of specific file types, click the File Types tab and follow the instructions described in Access by file type.
To configure custom headers, click the Custom Header tab and follow the instructions in Add a custom header.
To add a list to the policy, see Add a Block list to a policy or Add an exception to a policy.
To configure Firewall rules for Internet-bound traffic, click the Firewall tab and complete the steps in Create firewall rules.
Click Save. If you want to save and deploy the policy, click Save and Deploy.

Next steps

If you haven’t deployed the policy, make sure you deploy it to the SIA network. For instructions see Deploy configuration changes.