DNS (the system that turns a site name such as example.com into the IP address computers use to reach each other) is used in almost every connection, so attackers use it both to trick users and to control malware. Protective DNS (a DNS service that refuses to resolve known malicious domains) is an easy measure to deploy and cuts connections to those domains. This article explains what Protective DNS can and cannot do, how it differs from encrypted DNS and DNSSEC (digital signing of DNS data), and where an organization should start, based on NCSC (the UK National Cyber Security Centre), CISA (the US cybersecurity agency) and RFC 9076.
Why DNS matters for security
Before a device opens a site or sends data anywhere, it asks a DNS resolver (a server that receives questions about domain names and answers with IP addresses). The questions therefore reveal which services a user is contacting. RFC 9076 (an IETF (Internet Engineering Task Force) document from July 2021, informational rather than a binding standard) states that the queried name and the source IP address can reveal user behavior, and that resolvers see most of a user's DNS traffic. Attackers exploit this, for example with look-alike domains in phishing emails or malware that calls back to the attacker's domain. For defenders, DNS is a control and monitoring point that covers every device.
How Protective DNS works
Protective DNS is a resolver that compares each domain name to a list of malicious domains before answering. If the domain is on the list, the resolver does not return the real address. The user sees a block page, and a device infected with malware cannot reach its operator. NCSC describes its Protective DNS service as preventing access to known malicious domains by not resolving them. The blocking rules are stored in a Response Policy Zone or RPZ (a policy list the resolver uses to decide whether to answer or block), built from commercial, internal and open sources and reviewed to reduce wrongly blocked sites. CISA and NSA (a US national security agency) state in Selecting a Protective DNS Service (March 2021) that this can significantly weaken attacks such as ransomware, phishing and botnets (networks of hijacked machines under remote control), and that the DNS query logs can support incident response and threat hunting.
- What it blocks: domains known to be linked to malware, ransomware, spyware and phishing, including call-backs from malware already on a device, as NCSC states.
- What it adds: dashboards and logs that can feed a SIEM (a system that collects and analyzes logs from many sources), showing which device tried to contact a malicious domain.
- What it does not cover: brand-new domains not yet on the list, and connections that use an IP address directly without asking for a name. Our view: treat it as a layer that complements email and endpoint security, not a replacement.
Encrypted DNS and DNS data validation do different jobs
- DoT and DoH: DNS over TLS (a protocol that encrypts communication) and DNS over HTTPS encrypt DNS questions between the device and the resolver to stop eavesdroppers on the path. RFC 9076 notes that encryption does not reduce what the resolver sees, so trust shifts to the resolver you choose.
- DNSSEC (DNS Security Extensions): adds digital signatures to DNS data so a resolver can check who published it and that it was not altered in transit. RFC 4033 states that DNSSEC does not provide confidentiality and does not protect against denial-of-service attacks.
- Protective DNS: decides whether the destination domain is malicious, which neither of the two above does.
Protects the DNS data
- DoT and DoHencrypt queries on the path
- DNSSECchecks data was not altered
Judges the destination and keeps evidence
- Protective DNSblocks known malicious domains
- Resolver logsfor hunting and investigation
CISA published its Encrypted DNS Implementation Guidance on 16 May 2024 for US federal agencies. It recommends a phased rollout and using Protective DNS together with encryption. CISA says other organizations can use it as a benchmark. One practical caution: if user devices send DoH straight to an outside resolver, their DNS traffic bypasses the organization's Protective DNS. Organizations should therefore require devices to use only the resolver the organization has chosen.
Choosing a service and deploying it
- Threat data source: ask where the malicious-domain list comes from, how often it updates and how wrongly blocked sites are reviewed.
- Logs and SIEM integration: check that logs can be exported, searched back in time and how long they are kept.
- Users outside the office: ask whether devices off the corporate network are covered. NCSC offers a roaming service for Windows, macOS and iOS using DoH.
- Exceptions and fixing wrong blocks: system owners need a fast way to report and allow a wrongly blocked domain.
- Data location: DNS logs reveal user behavior, so check the provider's data terms and confirm with legal counsel where PDPA (Thailand's personal data protection law) applies.
According to NCSC, its Protective DNS service is mandated for UK central government and other organizations must meet eligibility criteria, so it is cited here as an example of the principle. Thai organizations must choose a provider that fits their needs, which may be a commercial resolver or the DNS filtering feature in a firewall or SASE (cloud-delivered networking plus security) service.
Where to start
- List the resolvers in use: check where devices and servers send DNS queries and whether any bypass to an outside resolver.
- Start in monitor mode: enable alert-only blocking for a pilot group, then review wrongly blocked sites before enforcing everywhere. This is the same approach described in Segmentation and Micro-segmentation: Start Safely.
- Send DNS logs to monitoring: use them with What Is NDR: Detecting Threats from Network Traffic.
- Protect your own domain: set up SPF (a list of servers allowed to send mail for the domain), DKIM (a digital signature on email) and DMARC (a policy for mail that fails those checks) as in DMARC, SPF and DKIM: Setting Them Up Correctly to reduce domain spoofing in email.
- Assess the whole picture: use the readiness assessment to see where DNS sits in your overall defenses.