Edit

Best practices for Web Application Firewall (WAF) on Azure Front Door

  • Article
  • 4 minutes to read

This article summarizes best practices for using the web application firewall (WAF) on Azure Front Door.

General best practices

Enable the WAF

For internet-facing applications, we recommend you enable a web application firewall (WAF) and configure it to use managed rules. When you use a WAF and Microsoft-managed rules, your application is protected from a range of attacks.

Tune your WAF

The rules in your WAF should be tuned for your workload. If you don't tune your WAF, it might accidentally block requests that should be allowed. Tuning might involve creating rule exclusions to reduce false positive detections.

While you tune your WAF, consider using detection mode, which logs requests and the actions the WAF would normally take, but doesn't actually block any traffic.

For more information, see Tuning Web Application Firewall (WAF) for Azure Front Door.

Use prevention mode

After you've tuned your WAF, you should configure it to run in prevention mode. By running in prevention mode, you ensure the WAF actually blocks requests that it detects are malicious. Running in detection mode is useful while you tune and configure your WAF, but provides no protection.

Define your WAF configuration as code

When you tune your WAF for your application workload, you typically create a set of rule exclusions to reduce false positive detections. If you manually configure these exclusions by using the Azure portal, then when you upgrade your WAF to use a newer ruleset version, you need to reconfigure the same exceptions against the new ruleset version. This process can be time-consuming and error-prone.

Instead, consider defining your WAF rule exclusions and other configuration as code, such as by using the Azure CLI, Azure PowerShell, Bicep or Terraform. Then, when you need to update your WAF ruleset version, you can easily reuse the same exclusions.

Managed ruleset best practices

Enable default rule sets

Microsoft's default rule sets are designed to protect your application by detecting and blocking common attacks. The rules are based on a various sources including the OWASP top 10 attack types and information from Microsoft Threat Intelligence.

For more information, see Azure-managed rule sets.

Enable bot management rules

Bots are responsible for a significant proportion of traffic to web applications. The WAF's bot protection rule set categorizes bots based on whether they're good, bad, or unknown. Bad bots can then be blocked, while good bots like search engine crawlers are allowed through to your application.

For more information, see Bot protection rule set.

Use the latest ruleset versions

Microsoft regularly updates the managed rules to take account of the current threat landscape. Ensure that you regularly check for updates to Azure-managed rule sets.

For more information, see Web Application Firewall DRS rule groups and rules.

Rate limiting best practices

Add rate limiting

Front Door's WAF enables you to control the number of requests allowed from each client's IP address over a period of time. It's a good practice to add rate limiting to reduce the impact of clients accidentally or intentionally sending large amounts of traffic to your service, such as during a retry storm.

For more information, see the following resources:

Use a high threshold for rate limits

It's usually a good practice to set your rate limit threshold to be quite high. For example, if you know that a single client IP address might send around 10 requests to your server each minute, consider specifying a threshold of 20 requests per minute.

High rate limit thresholds avoid blocking legitimate traffic, while still providing protection against extremely high numbers of requests that might overwhelm your infrastructure.

Geo-filtering best practices

Geo-filter traffic

Many web applications are designed for users within a specific geographic region. If this situation applies to your application, consider implementing geo-filtering to block requests that come from outside of the countries you expect to receive traffic from.

For more information, see What is geo-filtering on a domain for Azure Front Door Service?.

Specify the unknown (ZZ) location

Some IP addresses aren't mapped to locations in our dataset. When an IP address can't be mapped to a location, the WAF assigns the traffic to the unknown (ZZ) country. To avoid blocking valid requests from these IP addresses, consider allowing the unknown (ZZ) country through your geo-filter.

For more information, see What is geo-filtering on a domain for Azure Front Door Service?.

Logging

Add diagnostic settings to save your WAF's logs

Front Door's WAF integrates with Azure Monitor. It's important to save the WAF logs to a destination like Log Analytics. You should review the WAF logs regularly. Reviewing logs helps you to tune your WAF policies to reduce false-positive detections, and to understand whether your application has been the subject of attacks.

For more information, see Azure Web Application Firewall monitoring and logging.

Send logs to Microsoft Sentinel

Microsoft Sentinel is a security information and event management (SIEM) system, which imports logs and data from multiple sources to understand the threat landscape for your web application and overall Azure environment. Front Door's WAF logs should be imported into Microsoft Sentinel or another SIEM so that your internet-facing properties are included in its analysis. For Microsoft Sentinel, use the Azure WAF connector to easily import your WAF logs.

For more information, see Using Microsoft Sentinel with Azure Web Application Firewall.

Next steps

Learn how to create a Front Door WAF policy.