Insights
DNS sinkhole: how it works and what its logs show
Learn how DNS sinkholes block selected domains, where their limits lie, and how to investigate logs without assuming infection.

On this page
A Domain Name System (DNS) sinkhole answers requests for selected malicious or unwanted domains with a controlled address instead of the real destination. This keeps devices using that answer from reaching the real site and can help you investigate suspicious activity. It doesn't remove malware, and a blocked lookup alone doesn't prove infectionOpens in a new tab.
In short:
- Changed answers. A sinkhole replaces the real destination address for selected domains.
- Scoped blocking. Protection covers selected domains and lookups that reach the control.
- Investigation, not diagnosis. A lookup or connection alone doesn't prove malware is present.
How a sinkhole changes a DNS lookup
A DNS resolver translates domain names into Internet Protocol (IP) addresses that devices use to connect. A sinkhole changes the answer for domains selected by its policy. The replacement address may lead to a controlled monitoring server or prevent a useful connection.
Hypothetical flow: Your device asks its resolver for a domain on the blocklist. The resolver or firewall substitutes the configured answer, so your device doesn't receive the real destination address. If the answer points to a monitoring server, a follow-up connection may provide evidence for investigation.
Some tools described as sinkholes can return NXDOMAINOpens in a new tab instead of an address. NXDOMAIN means the requested name doesn't exist. That response denies the lookup instead of directing the device to a controlled destination.
What you can use a sinkhole for
You can use sinkholing to limit contact with known phishing sites and malware distribution domains. It can also limit access to malware's command and control servers, which provide instructions to infected devices. The effect depends on those domains being included in your policy.
Another use is to block selected ad-serving domains. This applies the same lookup control to unwanted content instead of malicious destinations. A monitored sinkhole can also gather evidence about attempted contact with selected domains.
A local sinkhole changes answers for devices whose queries pass through its deployment. Research sinkholes can be set up when registrars redirect a malicious domain's name-server records to a server run by researchers. Those sinkholes can receive traffic from many networks, not just one local network.
What a sinkhole can't stop
Protection covers domains selected by your policy and queries that reach the control. Apps using an external resolver through DNS over HTTPS (DoH) or DNS over TLS (DoT) may bypass a local sinkhole. These are encrypted ways to send DNS queries.
Encryption and domain blocking solve different problems. Encryption protects queries in transit, while blocking changes the answers for selected names. An approved resolver can still apply domain policy to encrypted queries because it receives those queries. Neither function removes malware or guarantees that every ad, tracking method or malicious connection stops.
If your deployment keeps DNS logs, those records can reveal attempted access to services. Limit who can read them, and set retention rules that fit your investigation needs. Don't assume a sinkhole provides logs, alerts or client identification without checking its actual capabilities.
What to check after a blocked lookup
A sinkhole event is a reason to investigate, not a malware diagnosis. Inspect only networks and devices you own or administer. These checks don't require changing DNS or firewall settings.
1. Record the domain, time and observed activity.
Start with the blocked domain and event time. Note whether the record shows a DNS lookup, a connection to the sinkhole address or both. A lookup proves that a query reached the logging control, not that malware made it.
2. Check what the source address represents.
The recorded source may be a client or a forwarding DNS server. A firewall placed beyond a local DNS server sees that server's request instead of the client's original query. If your source is a forwarding server, don't label that server infected based on the lookup.
3. Use available connection records to narrow the device.
If your deployment records follow-up connections to a controlled sinkhole address, check them against the domain and time. A connection passing through the firewall can reveal the originating client in that deployment. A client might never attempt that connection, so missing connection records don't resolve attribution.
4. Look for evidence tied to malware.
Use available endpoint or request evidence to check what caused the activity. SANS Internet Storm Center guidance says a simple lookup or sinkhole connection shouldn't be enough to classify a host as infected. If attribution is missing, seek client-side evidence instead of treating the forwarding DNS server as the infected device.
5. State what the evidence supports.
Your finding should identify the observed activity, say whether the originating device is known and separate suspicion from evidence of infection. If you have only a lookup or connection, record it as activity to investigate. Keep unresolved attribution or infection status marked as unknown instead of filling the gap with an assumption.
Where doxx.net fits. doxx.net encrypts DNS lookups and blocks unwanted domainsOpens in a new tab, but blocking isn't malware removal. If that fits your browsing needs, create a doxx.net account, download the app and connect a device or agentOpens in a new tab.