Malware on a device chops data into tiny chunks and encodes it in subdomains.
Your resolver forwards those lookups to the attacker’s authoritative DNS server.
The attacker decodes the data from the queries or sends back instructions in DNS answers.
Result: command-and-control and data exfiltration that looks like normal DNS traffic.
Lots of very long or random-looking domains (gibberish labels).
Unusual spikes in TXT record queries or NXDOMAIN responses.
Constant DNS traffic to a single strange domain or newly registered zones.
Endpoints making DNS queries at odd hours with steady, periodic bursts.
Block the suspicious domain/NS at DNS and firewall; capture samples.
Isolate affected hosts and run a full malware/EDR sweep.
From a clean admin box, rotate credentials and tokens used on that host.
Review logs to scope what left and who else is talking to the same domain.
Use a DNS filter/firewall with tunneling detections (length, entropy, TXT/NXDOMAIN heuristics).
Egress control: only allow DNS to approved resolvers; block outbound 53/853 elsewhere.
Turn on DoH/DoT to your resolver; log queries centrally and alert on anomalies.
Segment networks and least-privilege access so one host isn’t a gateway to all data.
Keep endpoints patched and monitored (EDR) to stop the malware that starts the tunnel.