Malware Takedown Process: The Nuance of Removing Malicious Domains

Published • Updated • 5 min read

A malware takedown is the process of getting a domain that distributes malware or receives stolen data suspended by its registrar or registry. The outcome depends less on detecting the malware and more on proving, in plain terms, how that domain is used to cause harm. This article is part of our series on how the phishing takedown process works.

Summary

  • Malware rarely relies on fixed IP addresses. It reaches its operators through domains, often generated in bulk, which makes the domain the point where a takedown disrupts the operation.
  • Since April 2024, ICANN contracts require registrars and registries of generic domains to act promptly once they have actionable evidence of DNS abuse, and malware is part of that definition.
  • What decides the outcome is whether the person reading the abuse report can see the harm: a sandbox screenshot, a multi-engine scan and one plain sentence work better than a 20-page technical report.

Why does malware depend on domains?

Malware depends on domains because it needs a reliable way to talk to its operators, and domains are cheaper to replace than any other piece of the setup. The DNS Abuse Framework uses the Internet & Jurisdiction Policy Network definition of malware: malicious software installed on a device without the user’s consent to disrupt it, steal information or gain access to private systems, including viruses, spyware and ransomware.

Modern malware is rarely built only to destroy a system. It steals data or recruits the device into a botnet, a network of infected machines controlled remotely, to launch further attacks. Both goals require regular contact with a server run by the attacker, known as a command and control (C2) server.

Attackers avoid hard-coding IP addresses into their malware because a fixed address is easy to block. They use domain names instead, often produced by a Domain Generation Algorithm (DGA): code inside the malware that generates a list of domains to try on a given day. The malware checks in with those domains, and when one of them answers, it receives instructions or uploads what it stole.

Domains can be registered and abandoned within minutes, which keeps the attacker’s infrastructure resilient. Domains are also where a takedown hurts most: once the domain the malware calls is suspended, the stolen data has nowhere to go.

What do registrars need before they suspend a malware domain?

Registrars need actionable evidence: proof that a specific domain is being used for malware, clear enough to act on without further investigation. Since April 5, 2024, ICANN’s DNS abuse amendments require registrars and registries of generic top-level domains (gTLDs, such as .com or .online) to promptly take appropriate action once they have actionable evidence that a domain is used for DNS abuse. Malware is one of the five forms of DNS abuse those contracts name. Country-code domains (ccTLDs, such as .ru or .br) follow national rules instead, as we cover in how takedowns work for ccTLDs.

The contracts do not say which action to take or how fast. The registrar decides, weighing the harm against the risk of collateral damage, and that judgment call is where most malware reports fail.

The person reading a registrar’s abuse inbox is often not a security researcher. At many registrars, abuse reports are handled by support, legal or compliance staff as one task among several. If that reviewer cannot follow the evidence, the safe choice is to take no action, because suspending a legitimate customer’s domain by mistake carries its own liability.

A report full of hex dumps and packet captures does not look thorough to that reviewer. It looks confusing, and a confused reviewer rarely approves a suspension.

How do you write a malware abuse report a registrar will act on?

Write the report for a reader with no technical background, and make the invisible data flow visible. Take a common case: an executable installs a keylogger and sends every keystroke to example[.]com. The report should rest on four elements:

  1. A sandbox screenshot that shows the connection. Run the file in a sandbox such as ANY.RUN or Joe Sandbox and capture the process tree where the malicious file connects to the target domain. One screenshot proves more to a non-specialist than pages of raw logs.
  2. Third-party validation. Link to the file’s VirusTotal scan. The registrar may not recognize your organization, but it trusts a result that many independent detection engines agree on.
  3. One plain sentence about the harm. Instead of “The DGA-seeded binary exfiltrates via C2”, write “This file steals the victim’s passwords and sends them to this domain.”
  4. A simple link chart. Three boxes (the malicious file, the reported domain, the server it reaches) let the reviewer follow the path in seconds.

What should you leave out of a malware abuse report?

Leave out anything the reviewer does not need to decide on this one domain:

  • Raw packet captures (PCAPs): too dense for a first report. Keep them ready in case the registrar asks for more.
  • Malware family names: the classification label matters less than what the file does to the victim.
  • Tangential evidence: related infrastructure or campaign history pulls attention away from the domain you want suspended.

How PhishFort applies this to takedowns

For the attacker, the domain is the infrastructure of the crime. For the registrar, the same domain is a paying customer’s asset. A takedown request has to bridge that gap, and evidence a non-specialist can verify in a few minutes is what makes the decision easy to justify.

PhishFort analysts file takedown requests through direct channels with registrars, hosts and platforms, and follow each case until the content is offline. PhishFort has a 99%+ takedown success rate.

Before you send a malware abuse report

  1. Can someone with no security background see the harm in under a minute?
  2. Does the report point to one domain and show how the malware uses it?
  3. Does it include validation from multiple independent detection engines?
  4. Is it addressed to the abuse contact of the registrar that sponsors the domain?

If impersonation domains targeting your brand need to come down at the source, see PhishFort’s phishing takedown service.

Sources

How PhishFort Protects Brands From Phishing

PhishFort finds the phishing sites, fake apps and impersonation accounts that use your brand and takes them down, with evidence for every case.

Book a Demo Explore PhishFort Takedowns