top of page

What Is a Botnet? How It Works, Detection & Protection Solutions

4 hours ago
23 min read
Botnet network connecting infected devices to a central command-and-control server.

A home router can spend years working for a stranger. Its owner streams shows, answers email and sees nothing wrong, while the device quietly waits for orders from a remote server. That is the unsettling nature of a botnet: ordinary equipment becomes a silent crowd that someone else controls. In March 2026, the U.S. Justice Department said four botnets had hijacked more than three million devices, mostly cameras, recorders and routers (DOJ, 2026-03-19). This guide explains how botnets work, how to spot one, and which defenses actually help.


TL;DR


  • What it is. A botnet is a group of compromised internet-connected devices that one operator controls remotely, usually without the owners knowing.

  • How devices join. Unpatched software, weak or default passwords, exposed services, malicious apps and, in documented cases, devices infected before purchase.

  • More than DDoS. Botnets also send spam, test stolen passwords, relay criminal traffic and spread other malware.

  • Signs are subtle. Look for unexpected outbound traffic, odd DNS requests and unexplained load. One symptom is a clue, not proof.

  • Different jobs, different tools. Endpoint detection, DNS controls, DDoS mitigation and penetration testing complement each other. None replaces the rest.


Quick answer: What is a botnet?


A botnet is a network of internet-connected devices, such as computers, routers and cameras, that malware has compromised and an attacker controls remotely. The attacker, called a botmaster or bot herder, sends commands to every infected bot at once. Botnets are used for DDoS attacks, spam, credential attacks, fraud and malware delivery.


Security engineer reviewing a web application security assessment on a monitor in a dark office

Know Where Your Applications and APIs Are Exposed

A professionally scoped security assessment can help identify relevant weaknesses in your web applications, APIs and SaaS products, and support informed fix decisions.


Table of Contents



What Is a Botnet? A Plain-English Definition


The word botnet joins “robot” and “network.” NIST’s glossary describes a botnet as a network of infected machines that criminals seize with malware and manage remotely (NIST CSRC Glossary). The FBI describes it as a network of hijacked computer devices used for attacks such as DDoS (FBI IC3).


Each infected device is a bot, also called a zombie device. The person running the network is the botmaster, or bot herder. Owners often notice little, because botnet malware is usually built to stay quiet and to leave everyday use mostly intact.


Not all automation is hostile. Search engines send crawlers to read pages, and companies run bots for testing and support. Automation becomes part of a malicious botnet when it runs on a device without the owner’s consent and takes orders from an attacker. Web scraping and other automated traffic raise separate questions about permission and purpose. They are not the same as device compromise.


A well-documented example is Mirai. Researchers traced its growth to a peak of about 600,000 infections, mostly embedded and IoT devices. In late 2016 it overwhelmed several high-profile targets with DDoS attacks (Antonakakis et al., USENIX Security 2017).


How Does a Botnet Work? The Attack Lifecycle


Botnets differ, so treat the lifecycle below as a common pattern rather than a script every botnet follows. MITRE ATT&CK describes two routes. Adversaries can buy or rent botnet infrastructure, or they can compromise third-party systems to build their own.


  1. Exposure. A device is reachable or vulnerable. Examples include an internet-facing service, an unpatched router or a login that still uses a factory password.

  2. Initial compromise. The attacker exploits a flaw, guesses a weak password or persuades a user to run a Trojan horse. Social engineering often supplies that push.

  3. Installation. Malicious code runs on the device. Some threats leave little on disk, as fileless malware does.

  4. Persistence, where applicable. Some malware survives restarts or hides itself, for example with a rootkit. Other malware does not persist and relies on reinfection.

  5. Command and control (C2). The bot contacts a control server or other bots for instructions. MITRE notes that adversaries hide C2 traffic inside ordinary web, mail or DNS protocols so it blends in (ATT&CK T1071).

  6. Recruitment. Bots may look for more victims. The FBI describes a TheMoon variant that checks in with a C2 server, which can direct a router to find and infect other vulnerable routers (FBI, 2025-05-07).

  7. Coordinated activity. The operator sends tasks to many bots at once, such as a DDoS flood, a spam run or a password-guessing job.

  8. Disruption and recovery. Defenders, internet providers or law enforcement may cut C2, clean devices or seize infrastructure. Operators can rebuild, and new variants appear.


Who controls the bots, and how


The botmaster is the person or group directing the network. Cloudflare explains that in a client-server design, bots contact a central C2 resource. That makes updates easy but creates a weak point defenders can target. Peer-to-peer designs remove that single point of failure (Cloudflare Learning Center).


Botnet lifecycle diagram showing eight stages from exposure and initial compromise through installation, persistence, command and control, recruitment, coordinated activity, and disruption and recovery
Figure 1. A common botnet lifecycle. Real botnets may skip, reorder or repeat stages.

Botnet Architectures and Types


Architecture describes how bots receive orders. Type usually describes what kind of device is infected. Both shape how a botnet behaves and how you can fight it.


In the client-server model, bots connect to a central C2 resource such as a web domain or an IRC channel. Cloudflare also lists star, multi-server and hierarchical layouts. Peer-to-peer bots keep lists of trusted peers and can pass commands along, so no single server holds control. Real botnets can mix ideas, so read these labels as design patterns rather than strict boxes.


Device type changes the risk. IoT devices such as cameras and recorders rarely run security agents. Routers sit at the network edge. Desktops and servers can run endpoint protection. Android-based streaming boxes, projectors and picture frames appear in the FBI’s BADBOX 2.0 alert. Residential proxy networks rent access to infected home connections, so other people’s traffic appears to come from the victim.


Architecture or type

Typical infected systems

How control works

Defensive implication

Centralized client-server

Mixed devices

Bots contact a C2 server or domain

Blocking or seizing C2 can disrupt the whole network

Peer-to-peer

Mixed devices

Bots relay commands to trusted peers

No single server to take down. Focus on endpoints and traffic patterns

IoT botnet

Cameras, recorders, routers

Usually central C2, often weak or default logins

Inventory, patching and segmentation matter most

Desktop and server

Windows, Linux and cloud workloads

Malware installed after exploit or phishing

EDR, patching and egress monitoring can see these hosts

Mobile and Android-based devices

Phones, TV boxes, projectors

Malicious or preinstalled apps

App sources and device certification matter

Residential proxy network

Home computers and devices

Infected hosts relay buyers’ traffic

Victims’ IP addresses are used for crime

Web, mail or DNS control channel

Any

C2 hidden in normal protocols

Needs DNS and flow analysis, not port blocking alone


How Devices Become Part of a Botnet


Infection routes are usually ordinary. Attackers favor weaknesses that are common and cheap to find.


  • Known vulnerabilities and exposed services. The FBI describes a TheMoon variant that targets end-of-life routers with remote administration enabled. It scans for open ports and does not need a password (FBI, 2025).

  • Unsupported hardware. The FBI notes that routers dated 2010 or earlier likely no longer receive software updates. Most routers in the KV Botnet were end-of-life Cisco and NetGear models (DOJ, 2024-01-31).

  • Weak or default credentials. Reporting on the Aisuru botnet says it compromised devices through known vulnerabilities or by brute-forcing weak credentials (BleepingComputer, 2025-12-03).

  • Trojanized software. 911 S5 spread through free VPN apps bundled with pirated games and software (FBI, 2024-05-29).

  • Malicious apps and preinstalled malware. The FBI says BADBOX 2.0 devices can be infected before purchase, during setup, or through apps from unofficial marketplaces (FBI, 2025-06-05).

  • Devices assumed to be shielded. The Justice Department said KimWolf and JackSkid were accused of infecting devices that are normally firewalled from the internet (DOJ, 2026-03-19). A router’s firewall is not a guarantee.


Three situations are easy to confuse. Direct infection means malware runs on the device. Credential compromise means an attacker holds a valid login but may not control the device itself. Other unauthorized control includes a proxy or remote-access tool added through a compromised app or account. Each needs a different response, from reimaging a device to revoking an account.


Security engineer working at a workstation reviewing application authentication and authorization test results

Identify Security Weaknesses Before They Escalate

Attackers often start with a known flaw or a weak login. A scoped assessment can review authentication, access control, exposed functionality and configuration within agreed boundaries.


What Are Botnets Used For?


Operators choose uses that earn money or cause disruption. No botnet does everything.


  • DDoS attacks. Attackers flood a target with traffic from many sources, which is hard to stop at any single source. The FBI notes that DDoS-for-hire services let even inexperienced actors run disruptive attacks (FBI IC3).

  • Spam and phishing. MITRE lists large-scale phishing as a follow-on use of botnets (MITRE T1584.005).

  • Credential attacks. Reporting on Aisuru says customers could rent its residential proxy capacity for spamming, scraping and credential stuffing (SecurityWeek).

  • Fraud and proxying. The FBI describes 911 S5 as giving criminals access to compromised devices so malicious activity looked like it came from the victims’ own devices (FBI, 2024).

  • Spreading more malware and stealing data. Cloudflare lists spam, DDoS and data theft as bot tasks, and notes that bots can spread malware to other devices on a network (Cloudflare).

  • Unauthorized cryptomining. Some malware also uses infected machines’ processing power. See Articsledge’s guide to cryptojacking.


The victim changes the story. A DDoS target loses availability. An account holder faces password guessing. The infected household may find its internet address used to hide someone else’s crime.


Botnet vs. Malware, DDoS, Worms and Legitimate Bots


These terms overlap, which is why buyers choose the wrong tool. A botnet is a controlled group of devices. Malware is the harmful software that often builds it. DDoS is one thing a botnet can do.


Term

What it is

Relationship to a botnet

Botnet

A network of compromised devices under one operator’s control

The controlled arrangement itself

Malware

Any software built to cause harm

Often the tool that infects a bot

Trojan

Malware that pretends to be useful

A common way devices are recruited

Computer worm

Malware that spreads on its own

Can recruit bots without user action

DDoS attack

A flood of traffic meant to knock a service offline

One use of a botnet, not a botnet

Legitimate web bot

Permitted automation such as a search crawler

Not a botnet, though it generates bot traffic

Malicious bot traffic

Unwanted automated requests, such as credential stuffing

May come from a botnet or from rented servers

Residential proxy service

A service that routes traffic through home connections

Can be legitimate or built on infected devices


The distinction guides tool choice. A computer worm or Trojan problem points to endpoint defenses. A traffic flood points to DDoS mitigation. Abusive automated requests to a login page point to bot management and rate limits. See Articsledge’s guides to malware, the Trojan horse and the computer worm for deeper background.


Real-World Botnet Examples and Lessons


Real cases show how botnets form and how defenders respond. Dates below separate the incident period from the publication date of each source.


Mirai (2016): simple infection at scale


Mirai surged in late 2016 and consisted mainly of embedded and IoT devices, peaking at about 600,000 infections. The researchers argue that its simple infection process let novice attackers compromise enough low-end devices to threaten well-defended targets. Lesson: cheap devices with weak security can become powerful weapons in bulk (Antonakakis et al., 2017).


911 S5 (2014–2022, takedown 2024): a residential proxy botnet


The FBI says 911 S5 operated from May 2014, went offline in July 2022, and returned as Cloudrouter in October 2023 (FBI, 2024-05-29). The Justice Department linked it to more than 19 million unique IP addresses and said administrator YunHe Wang was arrested on May 24, 2024. Charges are allegations. Lesson: free software from untrusted sources can carry a hidden backdoor (DOJ press release).


KV Botnet and TheMoon: end-of-life routers


In a January 31, 2024 announcement, the Justice Department described hundreds of small office and home routers, mostly end-of-life Cisco and NetGear models, taken over to hide the origin of state-sponsored hacking. A court-authorized operation removed the malware, but DOJ warned that restarting a router without protective steps can allow reinfection (DOJ). The FBI’s May 7, 2025 alert covers a TheMoon variant that targets similar equipment. Lesson: replace routers that no longer receive updates (FBI).


BADBOX 2.0 (FBI alert 2025-06-05): infected before purchase


The FBI says BADBOX 2.0 reportedly includes millions of infected devices, including TV streaming devices, projectors, aftermarket vehicle infotainment systems and digital picture frames. Infected devices can join residential proxy services. Lesson: unfamiliar brands and uncertified Android devices carry supply-chain risk. For vehicles, see Articsledge’s overview of automotive cybersecurity (FBI).


Aisuru, KimWolf, JackSkid and Mossad (2026): command infrastructure disrupted


On March 19, 2026 the Justice Department announced court-authorized action against the C2 infrastructure of four IoT botnets. Court documents cite more than 200,000 attack commands from Aisuru, 90,000 from JackSkid, 25,000 from KimWolf and 1,000 from Mossad (DOJ). Cloudflare’s Q3 2026 report counted about 18.2 million network-layer DDoS attacks and found attacks above 1 Tbps fell from 805 to 251 between Q2 and Q3, coinciding with the disruption. It stresses this is correlation, and says new variants keep emerging (Cloudflare, 2026-10-29). Lesson: takedowns help, but defenders still need resilient basics.


Risks to Businesses and Individuals


Botnet risk depends on which side of the attack you are on. The impacts below are possible outcomes, not guaranteed ones.


  • Service disruption. A DDoS attack can take a website or API offline. For a store or SaaS product, downtime can mean lost sales and support load.

  • Fraud and financial exposure. The FBI says 911 S5 caused confirmed victim losses in the billions of dollars (FBI).

  • Infrastructure abuse. An infected router or server can be used to attack others, which can lead to blocklisting and awkward questions from partners.

  • Reputation and trust. Customers judge a company by whether its service stays up and whether its devices are used against others.

  • Response burden. Containment, rebuilds, credential resets and communication take staff time, which small teams struggle to spare.

  • Compliance duties. Depending on sector and region, a serious incident may trigger reporting obligations under rules such as the EU’s NIS2 Directive.


Small organizations often feel the pressure most, since one person may handle IT, security and everything else. Understanding the broader cyberattack landscape and the threat actors behind it helps teams decide what to protect first.


Signs of a Possible Botnet Infection


Most botnet infections are quiet, so signs are indirect. The FBI’s router guidance lists overheating, connectivity problems and unrecognized setting changes as warning signs (FBI). For BADBOX 2.0, it lists unexplained or suspicious internet traffic and cautions that an indicator alone does not confirm malicious activity (FBI).


Treat each symptom as a clue. The table shows what it may mean, how to check, and what else could explain it.


Symptom

What it may mean

How to investigate

Alternative explanations

Priority

Unexpected outbound traffic

Bot reporting to C2 or sending attack traffic

Review flow logs by device and destination

Backups, updates, cloud sync

High if sustained and unexplained

DNS lookups to unfamiliar or rarely used domains

C2 lookups or tunneling

Check DNS logs for new domains and timing patterns

New apps, CDN changes, ad networks

Medium to high

Beacon-like, evenly timed connections

Check-ins to a control server

Look for regular intervals and small unbalanced transfers

Monitoring agents, heartbeat checks

High once confirmed

Unexplained CPU or bandwidth use

Attack traffic, proxying or mining

Compare against baseline and inspect processes

Updates, indexing, a busy user

Medium

Router or IoT settings changed

Takeover of an edge device

Compare settings to a known-good copy and check admin logins

Admin or ISP change, factory reset

High

Unknown processes, services or scheduled tasks

Malware installation or persistence

Verify file hashes and parent processes

Legitimate admin tools

High

Security alerts from EDR, IDS or firewall

Known malicious patterns detected

Validate the alert and scope affected hosts

False positives, test traffic

Depends on alert quality

Overheating or connectivity problems

Heavy hidden workload on a router

Check uptime, load and logs; compare with firmware version

Failing hardware, poor ventilation

Low to medium


One symptom proves nothing. Several together, such as odd DNS plus regular beacons plus an unknown process, make a much stronger case.


How to Detect Botnet Activity: A Practical Workflow


Detection means moving from vague suspicion to evidence. Much botnet activity happens outside your network, so MITRE advises focusing on related stages you can see, such as phishing and denial of service (MITRE T1583.005). Work through the steps below in order.


  1. Build an inventory. List every device, owner and location. You cannot spot a rogue camera you never recorded.

  2. Set egress baselines. Learn what normal outbound traffic looks like for servers, offices and IoT segments.

  3. Collect DNS telemetry. Log queries by device and review new, rare or oddly patterned domains.

  4. Analyze flows and logs. Use NetFlow, firewall and proxy logs to find beacons, scans and large unexplained transfers. Sound network security practice starts with this visibility, and a modern firewall supplies much of it.

  5. Watch endpoints. Endpoint protection and EDR show processes, persistence and parent-child behavior that network tools cannot.

  6. Add IDS, IPS and network detection. These flag known patterns and unusual traffic, especially from devices that cannot run agents.

  7. Correlate in a SIEM. A SIEM joins endpoint, DNS and network events so a weak signal becomes a stronger one. Teams that cover cloud workloads also need cloud telemetry.

  8. Enrich with threat intelligence. Compare destinations and file hashes with indicators of compromise, treating a match as a lead rather than a verdict.

  9. Triage false positives. Ask whether a business process explains the traffic before escalating.

  10. Plan for IoT blind spots. Devices without agents need network-based monitoring and strict segmentation.


MITRE’s detection guidance for application-layer C2 points to the same signals: common protocols used by abnormal processes, unusual outbound volume, beacon-like intervals and tunneling inside DNS or TLS (ATT&CK T1071). Broader threat detection and response programs turn these signals into a repeatable process.


A hypothetical triage example


This is an illustrative scenario, not a real incident. A monitoring tool flags an office camera that contacts a newly seen domain every 60 seconds. The analyst checks DNS logs and finds the domain appears nowhere else. Flow data shows small, evenly spaced transfers. A threat-intelligence lookup returns no match. The evidence is moderate, not conclusive, so the analyst isolates the camera on a quarantine VLAN, captures traffic, checks the vendor’s firmware advisories and escalates for review. Certainty rises only if the device is confirmed to be running unexpected code or the domain matches a known C2.


Turn Security Concerns into a Defined Assessment

If this guide exposed gaps in an important web application, API or SaaS system, plan a preventive review while it is calm. This is for planning ahead, not for an active incident.


How to Remove a Botnet Infection and Recover


Removal depends on the device and the evidence you need. Do not reboot or reset everything on reflex. The FBI advises reimaging compromised hosts unless you are told to preserve them forensically (FBI IC3).


  1. Assess scope and severity. Identify which devices talk to suspicious destinations and which accounts they touch.

  2. Isolate where appropriate. Move suspect devices to a quarantine network or take them offline.

  3. Preserve evidence if needed. Capture logs, memory or disk images when legal, regulatory or insurance questions are possible.

  4. Identify exposed credentials. List passwords, keys and tokens that lived on or were reachable from the device.

  5. Remove access by trusted means. Use vendor-supported cleanup or a known-good image rather than guessing.

  6. Reimage or factory-reset when justified. Cloudflare lists antivirus, restoring clean backups, reformatting and factory resets among remediation options (Cloudflare).

  7. Update firmware and software. The FBI advises updating firmware, changing passwords and rebooting a suspected router (FBI).

  8. Replace unsupported hardware. DOJ notes that cleanup can be temporary if the weakness stays (DOJ).

  9. Rotate credentials and revoke access. Reset passwords, keys and tokens, and remove unknown accounts.

  10. Validate and monitor. Confirm no beacons remain, then watch for reinfection.


Devices differ. PCs can often be cleaned or reimaged. Servers need careful rebuilds from trusted images because they hold data and keys. Routers usually need a firmware update, a reset and new admin credentials. IoT devices often cannot be cleaned and may need replacing. Reliable backups and a tested disaster recovery plan make rebuilding far less painful.


Bring in a qualified incident responder when the scope is unclear, sensitive data may be involved, evidence matters legally, or you cannot contain the problem. Penetration testing is a preventive assessment, not incident response. If you are in the United States, report suspected incidents to the FBI’s IC3.


How to Prevent Botnet Infections and Reduce Exposure


Prevention has two separate goals. One is keeping your own devices out of botnets. The other is protecting a public service from traffic that botnets elsewhere generate. Mixing them leads to buying the wrong control. The NIST Cybersecurity Framework 2.0 gives a helpful structure, organizing outcomes into Govern, Identify, Protect, Detect, Respond and Recover.


Keeping your devices out of botnets


  • Know what you own. Keep an inventory and retire end-of-life equipment.

  • Patch and update. Prioritize internet-facing systems and routers. The FBI urges prompt firmware updates (FBI).

  • Remove default logins and use strong, unique passwords. The FBI suggests random passwords of 16 to 64 characters. Enable MFA where supported.

  • Close what you do not use. Disable remote administration and unused services.

  • Segment networks. Put IoT and guest devices on separate segments. In cloud environments a virtual private cloud offers similar boundaries, and cloud security and cloud-native security practices extend them.

  • Install software only from trusted sources. 911 S5 spread through pirated software (FBI).

  • Protect endpoints and credentials. Malware such as a keylogger or spyware steals the logins that attackers reuse.

  • Train people. Regular phishing simulation builds the habit of pausing before clicking.


Home and small office versus business and enterprise


Home and small office: replace unsupported routers, turn on automatic updates, set unique passwords, disable remote management and avoid unbranded streaming boxes. Business and enterprise: add asset management, segmentation, centrally managed endpoint security, DNS filtering, egress monitoring, least-privilege access, supplier risk reviews and an API security program for exposed services. Good overall cybersecurity hygiene supports every one of these.


Protecting a public service from external botnet traffic


The FBI recommends enrolling in a denial-of-service mitigation service, working with your internet provider before an attack, maintaining continuity plans and keeping backups disconnected from the systems they protect (FBI IC3). Patching your own devices does not stop someone else’s botnet from flooding your website.


Botnet Protection Solutions Compared


No single product covers botnet risk. The table compares common categories by purpose, fit, limits and the moment buyers usually act. It describes categories, not Articsledge offerings.


Solution

What it solves

Best fit

Limitations

Buying trigger

Endpoint protection and EDR

Finds and contains malware on hosts

Organizations with laptops and servers

Cannot see agentless IoT

Alerts you cannot investigate

MDR or SOC service

24/7 monitoring and response by specialists

Teams without round-the-clock staff

Needs good data sources; adds cost

Gaps in after-hours coverage

SIEM

Correlates logs from many systems

Larger environments with many log sources

Needs tuning and analysts

Too many disconnected consoles

DNS security and filtering

Blocks known bad domains; reveals lookups

Any network, including remote users

Misses new or encrypted channels

Suspected C2 lookups

Firewall, IDS/IPS and NDR

Controls and inspects network traffic

Sites with internet edges and east-west traffic

Encrypted traffic limits visibility

Need for egress control

WAF, bot management and API controls

Filters abusive web and API requests

Public sites, logins and APIs

Does not clean infected devices

Credential stuffing or scraping

DDoS mitigation

Absorbs flood traffic before it hits you

Revenue-critical public services

Does not remove infections

Outages or repeated floods

Vulnerability management

Finds and tracks known weaknesses

Any organization with assets

Scans find known issues, not logic flaws

Patch backlog

Penetration testing

Tests whether weaknesses are exploitable

Important apps, APIs and SaaS

Point-in-time; not monitoring or cleanup

Release, audit or customer review

IoT security and segmentation

Limits what unmanaged devices can reach

Sites with cameras, sensors, TVs

Hard to retrofit

New device rollouts

Incident-response retainer

Pre-arranged expert help during incidents

Organizations lacking in-house IR

Costs money before any incident

Insurance or board requirement


These controls are complementary, not interchangeable. A DDoS service will not remove a bot from a camera. EDR will not absorb a flood. Pricing varies with the number of devices or users, data volume, coverage hours, contract term, onboarding effort and response commitments, so ask vendors for quotes against your own scope instead of trusting generic figures.


Strengthen Your Security Decisions with Evidence

Fixes are stronger when they are verified. A scoped assessment can produce clear findings and prioritized remediation guidance. Retesting to confirm fixes is agreed and scoped separately.


How to Choose the Right Botnet Protection Strategy


Start with what could hurt you most, then match controls to it. Ask these questions:


  • What do we expose? List public websites, APIs, VPNs, routers and cloud services.

  • What threats apply? Infection of our devices, attacks on our service, or both?

  • What do we already have? Map existing controls to the table above.

  • What can we staff? Decide what to run in-house and what to buy as a managed service.

  • What proof do we need? Boards, customers and insurers often ask for evidence, such as reports and logs.


In-house teams suit organizations with skilled staff and steady workloads. Managed services suit those who cannot cover nights and weekends. Specialist assessments suit a single high-risk application. Many organizations combine all three. Articsledge’s overview of cybersecurity services explains typical service types and costs.


Criterion

What to ask

Score 1–5

Coverage

Does it address our biggest exposure first?

Your score

Visibility

What does it see, and what stays hidden?

Your score

Effort

What staff time and tuning does it need?

Your score

False positives

How are noisy alerts handled?

Your score

Proof

What reports or evidence does it produce?

Your score

Lock-in and cost

What are exit terms and total cost?

Your score


Weight each criterion by importance to you, multiply, and compare shortlisted options. Penetration testing is not the default answer to every botnet risk. It is one useful part of a layered plan.


Where Penetration Testing Fits into Botnet Risk Reduction


Botnets often succeed by finding a weakness someone could have fixed. Authorized, scoped penetration testing can help assess weaknesses in web applications, public APIs, authentication, authorization, business logic, session management and configuration within the agreed scope. Articsledge’s service page describes web application, API and SaaS testing, vulnerability assessments and remediation retesting (penetration testing services). The related guide to penetration testing services covers types, process and costs.


A vulnerability assessment is a lighter way to identify and prioritize known weaknesses. Remediation verification checks whether reported findings were actually fixed. Retesting can be included in the original proposal or quoted separately.


What penetration testing cannot replace


  • DDoS scrubbing or mitigation

  • Continuous endpoint monitoring or managed detection and response

  • Threat hunting

  • Emergency malware removal and forensic incident response

  • A maintained vulnerability-management program


A test is a point-in-time assessment within agreed limits. It does not guarantee security, establish compliance or find every flaw.


How to prepare an inquiry


  • Describe your company, system type, approximate scope and objectives.

  • State your timeline and any standards or report formats you need.

  • Prepare documentation and test accounts, which affect depth and price.

  • Expect to give written testing authorization and agree rules of engagement before active testing.

  • Do not send passwords, private keys or production credentials in the first message.

  • Ask whether any cloud-related review applies to your environment.


When you are ready, review the scoped testing options on the service page. Questions you may still have are answered next.


Ready to Scope Your Application or API Security Test?

Describe the systems involved, your testing objective and your timeline. Availability, scope, deliverables and commercial terms are confirmed before any engagement, and testing starts only with written authorization.


FAQ


What is a botnet in simple terms?


A botnet is a group of infected internet-connected devices controlled by one attacker. The devices keep working for their owners while quietly taking orders from someone else. Attackers use the group to launch attacks, send spam or relay traffic.


How does a device join a botnet?


Attackers exploit unpatched software, guess weak or default passwords, abuse exposed services, or trick people into installing malicious apps. In documented cases such as BADBOX 2.0, devices were infected before purchase or during setup. Once malware runs, the device contacts the attacker’s control system.


Can a phone or router become part of a botnet?


Yes. Routers are a frequent target, especially end-of-life models that no longer get updates. Android-based devices have also been named in FBI alerts, including TV boxes and projectors. Phones can be enrolled through malicious apps, so install software only from trusted sources.


What is the difference between a bot and a botnet?


A bot is a single compromised device, or a piece of automated software. A botnet is the whole controlled network of bots. Some bots are harmless, such as search crawlers, while a botnet in the security sense is malicious by definition.


Are all botnets used for DDoS attacks?


No. DDoS is one common use, but botnets also send spam, relay traffic through proxies, attempt credential attacks and spread malware. The operator decides, and some botnets are built mainly to rent out proxy access.


How can I tell whether my router is infected?


Look for unrecognized setting changes, overheating, connectivity problems and unusual traffic. The FBI lists these as warning signs, but none proves infection on its own. Check the router’s firmware version, compare settings with a known-good copy and look at your provider’s logs if available.


Can antivirus software detect a botnet?


Endpoint security can detect many bot malware families on computers and servers. It cannot see inside most routers, cameras or smart TVs, and it will miss threats with no known signature. Combine it with network monitoring and updates.


How do security teams detect botnet command-and-control activity?


They look for beacon-like connection timing, unusual DNS lookups, odd protocol use and traffic to known bad destinations. DNS logs, network flow data, endpoint telemetry and threat intelligence are correlated to build a case. MITRE’s guidance on application-layer C2 describes these patterns.


Can a botnet survive a device reboot?


Some can and some cannot. The Justice Department warned that after removing KV Botnet malware, a restarted router could be reinfected if it was not protected. The FBI’s advice for a suspected TheMoon infection combines a reboot with a firmware update and new passwords.


Does a VPN protect a device from botnet malware?


No. A VPN encrypts traffic and hides your address, but it does not stop malware from running or block exploitation of your device. Free VPN apps can even be the source: 911 S5 spread through free VPN apps bundled with pirated software.


How can a business reduce its risk of botnet attacks?


Keep an asset inventory, patch internet-facing systems, remove default credentials, segment networks and deploy endpoint and DNS protections. For public services, add DDoS mitigation and bot controls. Test your recovery plan before you need it.


What is the best botnet protection solution?


There is no single best option, because the tools solve different problems. EDR protects hosts, DNS filtering blocks malicious lookups, DDoS mitigation absorbs floods and testing finds exploitable weaknesses. Choose by your exposure and staffing.


What should I do if I suspect an active botnet infection?


Contain first. Isolate the suspect device, preserve logs if evidence may matter, and avoid mass resets before scoping. Bring in a qualified incident responder if the situation is unclear or sensitive data may be involved. In the United States, you can report to the FBI’s IC3.


Can penetration testing identify weaknesses that botnets exploit?


It can help identify weaknesses in the web applications, APIs and SaaS systems that are in scope, such as authentication and authorization flaws. It does not detect active infections, stop DDoS traffic or clean devices. It complements monitoring and response.


What affects the cost of botnet protection?


Cost depends on how many devices and users you cover, the data volume you collect, coverage hours, contract length and onboarding effort. Penetration testing pricing depends on scope, complexity and specialist availability. Ask each provider for a quote against your defined scope.


Is botnet activity illegal?


In most jurisdictions, building or operating a botnet and using it to attack others is a crime. U.S. authorities charged the alleged 911 S5 administrator with computer-fraud offenses, though charges are allegations until proven. This is general information, not legal advice, so consult a qualified lawyer for your situation.


Key Takeaways


  • A botnet is a controlled network of compromised devices, not just a piece of malware.

  • DDoS is one use of botnets. Spam, proxying, credential attacks and malware delivery are others.

  • End-of-life routers, default passwords, exposed services and untrusted apps are recurring entry points.

  • A single symptom is weak evidence. Correlate DNS, flow and endpoint data before concluding.

  • Do not reset everything blindly. Scope and preserve evidence first, then clean or rebuild.

  • Protecting your devices and protecting your service from external botnets are different jobs.

  • EDR, DNS controls, DDoS mitigation and testing are complementary, not substitutes.

  • Penetration testing is a point-in-time assessment, not monitoring, DDoS defense or incident response.


Actionable Next Steps


  1. Today: If you suspect an active infection, isolate the device and contact a qualified incident responder. Do not wait for a sales conversation.

  2. Today: Change default passwords on routers and cameras, and disable remote administration you do not need.

  3. This week: List every internet-connected device, its owner and its update status.

  4. This week: Replace or isolate any device that no longer receives security updates.

  5. This week: Turn on automatic updates and MFA wherever they are supported.

  6. This month: Put IoT and guest devices on a separate network segment.

  7. This month: Enable DNS logging and set a baseline for normal outbound traffic.

  8. This month: Review whether your public services need DDoS mitigation and bot controls.

  9. Ongoing: Test backups and rehearse your response plan twice a year.

  10. When ready: For important web applications, APIs or SaaS systems, consider a scoped security assessment and plan retesting after fixes.


Glossary


  • Botnet. A network of compromised devices controlled remotely by an attacker.

  • Bot. A single compromised device in a botnet, or a piece of automated software.

  • Zombie device. Another name for an infected device that obeys a botmaster.

  • Botmaster. The person or group that controls a botnet, also called a bot herder.

  • Command and control (C2). The server or channel used to send instructions to bots.

  • IoT. Internet of Things: connected devices such as cameras, sensors and smart TVs.

  • DDoS. Distributed denial of service: an attack that floods a target with traffic from many sources.

  • Malware. Software designed to harm systems or steal data.

  • Persistence. Methods that let malware keep running after a restart.

  • Phishing. Deceptive messages that trick people into revealing secrets or running malware.

  • IDS / IPS. Intrusion detection and prevention systems that spot, and sometimes block, suspicious network activity.

  • EDR. Endpoint detection and response: tools that monitor and contain threats on devices.

  • SIEM. Security information and event management: a platform that collects and correlates logs.

  • WAF. Web application firewall: a filter that inspects web requests to protect applications.

  • API security. Controls that protect application programming interfaces from misuse and attack.

  • Network segmentation. Dividing a network into separate zones to limit how far an attacker can move.

  • Vulnerability assessment. A review that finds and prioritizes known weaknesses.

  • Penetration testing. Authorized, scoped testing that attempts to exploit weaknesses to show real risk.

  • Remediation verification. Retesting to confirm that reported weaknesses were fixed.


Sources & References


  1. NIST Computer Security Resource Center (CSRC). Glossary: Botnet. n.d..

  2. National Institute of Standards and Technology. The NIST Cybersecurity Framework (CSF) 2.0. 2024-02-26.

  3. FBI Internet Crime Complaint Center (IC3). Botnets, DDoS, and TDoS. n.d..

  4. FBI IC3. Public Service Announcement I-060525-PSA: BADBOX 2.0 Targets Consumer Home Networks. 2025-06-05.

  5. FBI IC3. Public Service Announcement I-050725-PSA: End-of-Life Routers. 2025-05-07.

  6. FBI, DCIS and U.S. Department of Commerce, via IC3. Public Service Announcement I-052924-PSA: 911 S5 Residential Proxy Service. 2024-05-29.

  7. U.S. Department of Justice. 911 S5 Botnet Dismantled and Its Administrator Arrested in Coordinated International Operation. 2024-05-29.

  8. U.S. Department of Justice. U.S. Government Disrupts Botnet People’s Republic of China Used to Conceal Hacking of Critical Infrastructure. 2024-01-31.

  9. U.S. Attorney’s Office, District of Alaska. Authorities disrupt world’s largest IoT DDoS botnets responsible for record breaking attacks targeting victims worldwide. 2026-03-19.

  10. Cloudflare Cloudforce One. Cloudflare DDoS Threat Report 2026 Q3. 2026-10-29.

  11. Cloudflare Learning Center. What is a DDoS botnet?. n.d..

  12. Antonakakis et al., USENIX Security Symposium. Understanding the Mirai Botnet. 2017-08.

  13. MITRE ATT&CK. T1583.005 Acquire Infrastructure: Botnet. n.d. (page last modified 2025-10-24).

  14. MITRE ATT&CK. T1584.005 Compromise Infrastructure: Botnet. n.d. (page last modified 2026-05-12).

  15. MITRE ATT&CK. T1071 Application Layer Protocol. n.d. (page last modified 2026-09-03).

  16. BleepingComputer. Aisuru botnet behind new record-breaking 29.7 Tbps DDoS attack. 2025-12-03.

  17. SecurityWeek. Aisuru Botnet Powers Record DDoS Attack Peaking at 29 Tbps. n.d..


bottom of page