Smiling person using laptop, surrounded by security shield icons, emphasizing cybersecurity and engagement.

DNS Leak Test: 7 Smart DNS and WebRTC Checks

A DNS leak test checks which DNS resolvers answer your browser’s requests; a WebRTC leak test checks which IP addresses your browser exposes during connection discovery. Compare both with your VPN off and on, then check IPv4, IPv6, reconnects, and browser DNS settings. This guide explains how to test DNS and WebRTC leaks, interpret confusing results, and fix unexpected exposure without breaking useful browser features.

Your VPN’s connected icon tells you a tunnel exists. It does not prove that every application, DNS request, or browser connection follows it. The useful question is simple: does this setup behave the way you intended?

Start with the checks below on your own device. You do not need penetration-testing software, a paid scanner, or an account on a leak-test website. You do need a baseline, a few minutes, and the willingness to question a suspicious result before declaring your router haunted.

Affiliate disclosure: This article includes NordVPN affiliate links. If you buy through them, I may earn a commission at no extra cost to you. You can complete every test here with your existing VPN.

CheckExpected resultInvestigate if
DNS leak testYour intended VPN or custom resolverUnexpected ISP DNS appears
WebRTC leak testVPN exit address, local candidates, or no candidatesYour baseline public IP appears
IPv6 and reconnectTraffic stays tunneled or is blockedYour ISP-assigned public address returns

Use the table to triage results, then follow the full method. A resolver name or country alone cannot establish the route your DNS traffic took.

Key Takeaways

  • Don’t trust the VPN icon—run a DNS leak test and a WebRTC leak test every time you change networks.
  • Leaks expose domains and real IPs even over HTTPS; incognito won’t save you.
  • Baseline first, then compare with VPN on using ipleak.net and dnsleaktest.com.
  • Use the VPN’s recommended DNS configuration first; a public resolver address alone does not enable encrypted DNS.
  • Tame WebRTC in the browser or use privacy-focused options like Brave.
  • Check IPv6 separately: your VPN should tunnel it or reliably block it while connected.
  • Re-test after OS, browser, or VPN updates; today’s clean can be tomorrow’s oops.

What a DNS Leak Test and WebRTC Check Reveal

A VPN changes the route of traffic covered by its tunnel. DNS chooses where names are resolved, while WebRTC discovers possible paths for calls and data connections. Those jobs overlap, but they are not the same setting.

A DNS leak test and a WebRTC leak test answer different questions. Run both rather than interpreting one clean result as complete protection. An official VPN application, a router tunnel, and a browser-only proxy can cover different traffic even when they all display “connected.”

HTTPS protects content, not everything. And browsers? They love shortcuts.

How to Test DNS WebRTC Leaks

Why a DNS Leak Test Matters Even With HTTPS

HTTPS encrypts what you say to a website — but not always who you’re asking for.

Before a page loads, your system asks a DNS resolver for the IP address. If that request goes to your ISP instead of through your VPN tunnel, your browsing habits are suddenly very visible.That’s why VPN DNS leak protection matters.

DNS carried inside an encrypted VPN tunnel is protected on the device-to-VPN leg even if it uses ordinary DNS within that tunnel. DoH or DoT can add encryption to a resolver, but they are not a prerequisite for every correctly configured VPN. A DNS leak test identifies observed resolvers; it does not prove that DoH or DoT is active.

How WebRTC Can Reveal Your Real IP via Browser APIs

WebRTC is great for video calls and peer-to-peer connections. It’s also very good at betraying you.

Browsers like Chrome, Firefox, Edge, and Brave use WebRTC to negotiate connections via STUN servers. If misconfigured, this can expose your real IP address, even while a VPN is active.

That’s why I rerun a WebRTC leak test after browser updates. Settings change. Flags reset. Profiles drift.If my home IP shows up, I either hard-disable WebRTC or switch to a hardened browser profile.

DNS Leak Test Warnings: Common Configuration Causes

Most leaks aren’t “hacks”. They’re configuration ghosts:

  • Misconfig: the OS might switch to ISP DNS after an update, or the VPN might forget to use its own DNS.
  • ISP tricks: they might use a transparent DNS proxy, forcing your traffic to go through their servers.
  • IPv6 gaps: the VPN might cover IPv4 but leave IPv6 queries open; sometimes, Teredo can make things worse.
  • Windows quirks: Smart Multi-Homed Name Resolution can send DNS queries to multiple servers and choose the fastest one.

These causes are possibilities, not a diagnosis. First confirm the unexpected address, then check routing, DNS settings, browser overrides, and exclusions. Encrypted DNS changes who can read a query on its path; it does not automatically force that path through the VPN.

Beginner Pitfalls: Incognito Mode & False Confidence

Incognito mode doesn’t make you anonymous. It just keeps your browser from remembering what you did — not what the network saw.

A green VPN icon is not proof of privacy. Comparing a DNS leak test and a WebRTC leak test before and after connection gives you observable evidence. Private browsing changes local history storage; it does not replace a VPN or hide your public IP from websites.

Privacy isn’t a toggle. It’s a habit.

“It is not enough to be intelligent. One must be intelligent intelligently.”

Stephen Fry

Vibrant green shield with i symbol; protection, security, information theme.

DNS Leak Test: Run the Baseline and VPN Comparison

I test leaks like a burglar checks windows — quietly, methodically, and twice before trusting anything. This guide shows how to test DNS WebRTC leaks without guesswork. We’ll run a DNS leak test, a WebRTC leak test, and compare clean baselines with VPN-on results. I rely on proper VPN DNS leak protection to keep things boring — which is exactly what you want.

Baseline vs. VPN-on comparison using DNS leak test sites (ipleak.net, dnsleaktest.com)

Start with a short, controlled VPN-off baseline. Pause downloads and close sensitive sessions first, because this step intentionally uses your normal connection. If an always-on policy prevents a baseline, use previously recorded addresses or compare with your network information rather than bypassing a required security policy.

I visit tools like:

  • IPLeak for public addresses, DNS, and WebRTC candidates
  • DNSLeakTest for standard and extended resolver checks

I note three things:

  • Public IP
  • DNS resolvers
  • Reported location

Record the exact addresses, resolver names, and time locally. Do not post an unredacted screenshot of your baseline IP in a public comment. IP geolocation is approximate; the address itself is the more useful comparison.

Connect the VPN and reload the tests. The public IP should reflect the VPN exit for traffic you intend to tunnel. DNS should reach the resolver you deliberately selected: the provider’s service by default, or an approved custom resolver. The DNS operator does not have to own the VPN exit IP.

Unexpected ISP resolvers are a warning worth investigating, especially when they match your baseline. Check whether they were intentionally configured and whether the VPN uses upstream resolvers before drawing a conclusion.

Running a WebRTC Leak Test in Modern Browsers

Next, open the WebRTC panel on IPLeak in the same browser profile you normally use. Compare its public candidates with your VPN-off public IPv4 and IPv6. A repeatable appearance of your baseline public address while the VPN is connected is the key warning; a private LAN address is a different type of disclosure.

Chrome, Firefox, Brave, and Microsoft Edge can produce different candidates because their policies and profiles differ. Modern browsers may hide local host candidates behind an mDNS .local name. That is not your public ISP address. A test reporting no candidates may mean WebRTC is restricted, the test was blocked, or discovery did not complete; it is not a certificate of anonymity.

Let the DNS leak test finish, then run its extended option if available. It generates more lookups and may expose additional resolver instances. If a test remains blank, check JavaScript and content-blocking settings or use a second trusted tool. Do not disable all browser protection just to obtain a green badge.

I test twice, switch VPN servers once, and test again. Consistency matters more than panic.

Reading Your DNS Leak Test and WebRTC Results

A bad DNS leak shows:

  • ISP DNS resolvers
  • An unexpected resolver confirmed against your baseline; a country mismatch alone is not proof

A bad WebRTC leak shows:

  • My real public IP
  • Your baseline globally routable IPv6 address, rather than just a private or link-local address

An address such as 192.168.1.20 is a private LAN candidate, not your ISP-issued public IP. IPv6 link-local addresses beginning with fe80: and unique-local addresses in fc00::/7 are also not public internet addresses. Local exposure may matter for your privacy goals, but it does not by itself prove that traffic escaped the VPN.

DNS Leak Test False Positives and Retesting

Resolver geolocation can be inaccurate, and providers may use shared infrastructure, upstream resolvers, or several egress addresses. Browser DoH may also deliberately select a different resolver. Compare ownership and configuration rather than expecting every country label to match. Multiple DNS servers do not automatically mean multiple leaks.

  • I flush the DNS cache and restart the browser.
  • I re-run the DNS leak test and WebRTC leak test after tweaks.
  • I test again after OS, browser, or VPN updates—things shift overnight.

Things break quietly. Testing keeps them honest.

That baseline → VPN-on comparison is my go-to method.

Fast, repeatable, and boring — the holy trinity of good security.

What a DNS leak test cannot prove

A DNS leak test usually triggers unique hostname lookups and identifies the recursive resolvers that contact the test’s authoritative servers. It does not directly inspect packets leaving your physical network adapter. It cannot establish the complete route, encryption on every hop, a provider’s logging policy, or whether every other application uses the same resolver.

A browser test also does not validate your email client, game launcher, torrent client, or every VM. Repeat relevant checks inside the application or guest environment you actually protect. A clean result is evidence for that configuration at that moment; cookies, logged-in accounts, and browser fingerprinting remain separate privacy concerns.

“Arguing that you don’t care about privacy because you have nothing to hide is like saying you don’t care about free speech because you have nothing to say.”

Edward Snowden

Stylized portrait of a man in bold colors, highlighting modern digital art trends.

7 Smart DNS Leak Test and WebRTC Checks

Most people stop after one green checkmark. That’s how leaks survive. These seven checks catch what VPN dashboards quietly ignore.

1. Repeat the DNS leak test after reconnecting

Run a DNS leak test after reconnecting and after changing VPN servers. A client may reapply routing and DNS settings during connection changes. If an unexpected baseline resolver returns, investigate the transition rather than assuming the connected icon guarantees the old configuration survived.

2. Check IPv6 alongside your DNS leak test

Check public IPv6 separately from IPv4. A VPN may tunnel IPv6, block it, or leave it outside its protection depending on the application and platform. A blocked IPv6 test can be expected; an ISP-assigned global IPv6 address while connected needs investigation. Do not disable IPv6 merely because a test displays an address you have not identified.

3. WebRTC leaks in secondary browser profiles

Your main browser might be clean. Your other profile often isn’t. Run a WebRTC leak test in:

  • Private windows
  • Secondary profiles
  • Fresh browser installs

Leaks love forgotten profiles.

4. Check the kill switch, then repeat the DNS leak test

Pause sensitive activity before testing. On your own device, enable the VPN’s kill switch and perform a controlled network interruption, such as briefly turning Wi-Fi off and on. Check whether traffic is blocked during reconnection and whether the expected VPN public IP and DNS return afterward. Deliberately clicking Disconnect may behave differently from an unexpected tunnel failure; consult the client’s documentation.

A DNS leak test launched after reconnection cannot prove that no packets escaped during the interruption. Stronger validation uses a packet capture on the physical interface with a qualified tester. Browser tests are a useful snapshot, not a continuous audit.

5. Run a DNS leak test after meaningful updates

After a significant browser, operating-system, or VPN update, verify your saved configuration and re-run:

  • DNS leak test
  • WebRTC leak test

Treat this as a regression check, not evidence that every update resets your settings.

6. Check split tunneling against your DNS leak test

Split tunneling deliberately excludes selected traffic. Temporarily test a full-tunnel setup, then restore your intended exclusions and test each relevant browser. A browser you intentionally excluded can show your ISP IP without proving that protected applications leak. DNS handling varies by client; document the policy you actually want.

7. Compare DNS leak test results with your intended policy

Compare the exact public addresses and resolver operators from both runs. The same DNS resolver can be intentional when you use custom DNS, so identical resolver names alone do not prove a leak. Identical public IPs for a browser you meant to tunnel are a stronger warning.

Why this matters? A VPN icon is not proof. Only repeated testing confirms VPN DNS leak protection actually works.

Quiet checks. Repeatable results.

That’s how you keep your setup honest.

“Privacy isn’t broken by hackers — it leaks because people stop checking.”

Robin Kool, HackersGhost (that’s me )

Follow my lab notes & reflections on Facebook

Fixing Leaks Properly: Browser, OS, and VPN Settings

I don’t fix leaks for five minutes of peace. I fix them so they survive updates, reboots, and bad decisions at 2 a.m. My rule is simple: lock the VPN first, harden the OS second, tame the browser last — then verify everything.

Assume nothing is private until you test it.

Fix DNS Leak Test Warnings in the VPN Client

Start at the source: the VPN app itself. I enable DNS leak protection, keep the kill switch on, and force the VPN to use its own private resolvers inside the tunnel.

A supported full-tunnel configuration is the best starting point. Update the client, reconnect, and run a DNS leak test before changing custom settings. If the problem remains, collect the unexpected address, client version, browser profile, and relevant exclusions for the provider’s support team. Trust is nice. Verification is better.

For the specific combination of NordVPN and custom AdGuard DNS, use my NordVPN DNS and AdGuard checks. That article covers the product-specific setup; this one remains a general testing guide.

Follow the DNS path when NordVPN and AdGuard DNS are used together. These checks help distinguish resolver choices from unintended exposure.

DNS Leak Test Differences Caused by Browser Secure DNS

Next layer: the operating system, browser, or router. Browser Secure DNS can override the operating system’s resolver. This may show a third-party DNS service even when its HTTPS connection travels through the VPN. It can also bypass VPN-provided filtering. Decide which resolver and route you want before treating a difference as a failure.

I manually set secure resolvers like:

  • Cloudflare 1.1.1.1
  • Google 8.8.8.8

Entering 1.1.1.1 or 8.8.8.8 only selects a resolver; it does not automatically enable DoH or DoT. Prefer the VPN’s supported configuration before overriding OS or router DNS. If you intentionally use encrypted DNS, check the client’s encryption status and fallback behavior as well as the observed resolver. Keep in mind that the chosen resolver still receives your queries.

Disable or restrict WebRTC safely

WebRTC is helpful — and noisy.

  • Firefox desktop: if you need to disable WebRTC completely, open about:config, locate media.peerconnection.enabled, and set it to false. Restore true to re-enable it. This can break browser calls and other WebRTC applications.
  • Chrome and Edge: there is no universal normal-settings switch that disables WebRTC. A reputable, compatible extension can limit IP handling, but check its permissions and test the result. Restricting address handling is not the same as disabling WebRTC.
  • Brave: search Settings for WebRTC IP Handling Policy. Disable Non-Proxied UDP restricts direct UDP connections and can affect calls; it does not turn WebRTC off completely. Retest both privacy and call functionality.

After changing a browser policy, run a WebRTC leak test again. If your baseline public IP still appears, check the VPN route and split-tunnel exclusions before adding more extensions. Then test a browser call you rely on: a privacy fix that silently breaks your meetings needs a better configuration.

Handle IPv6 before it handles you

IPv6 deserves its own check because a clean IPv4 result tells you nothing about another address family.

HackersGhost Note: My setup includes Windows as the host, VMware guests, and Proton VPN through WireGuard on a Cudy WR3000. That makes the test location important: a result on the host is not a result inside Parrot OS, and a router tunnel only protects devices routed through it. Check the active guest adapter and gateway before blaming the VPN.

Prefer the VPN client’s documented IPv6 handling: tunnel it when supported, or block it while the VPN is active. Manual OS-wide disabling is a troubleshooting fallback, not the first requirement for everyone. Router WireGuard profiles and third-party clients may behave differently from the provider’s official application.

Do not change transition technologies or network adapters indiscriminately. On the latest Windows version, start with the VPN’s documented configuration and test the actual public-address result before making broader network changes.

Review NordVPN’s features and practical limitations alongside your own leak-test results before deciding whether it fits your setup.

Read my NordVPN review for product considerations before deciding.

Conclusion: Act on Your DNS Leak Test Results

Start with a baseline, connect the VPN, and run a DNS leak test alongside a WebRTC leak test. Investigate an unexpected ISP public address first. Then check DNS ownership, custom resolvers, IPv6 handling, and split-tunnel exclusions. Change one setting at a time and keep the result you can reproduce.

If a DNS leak test shows your intended custom resolver, that can be correct. If WebRTC shows only a private candidate or mDNS name, do not mistake it for your public IP. Fix the exposure you actually found rather than disabling half your network because a country label looks odd.

Retest after reconnects, network changes, and meaningful updates. These checks can reveal configuration problems; they cannot make logged-in accounts, cookies, or fingerprinting disappear. Boring, documented results beat a reassuring icon every time.

If you are choosing a VPN, NordVPN is one option to consider. The same independent testing method applies whichever service you choose.

Young man smiling with laptop; colorful cybersecurity-themed backdrop with shield and lock icons.

Frequently Asked Questions

How do I run a DNS leak test without false positives?

What’s the fastest way to do a DNS leak test?

How to fix DNS leak issues when using a VPN?

How to disable WebRTC safely without breaking your browser?

Why do I pass the DNS leak test but still fail the WebRTC leak test?

VPN & Network Infrastructure Cluster

ⓘ

Some links in this article are affiliate links. If you use them, I may earn a small commission — at no extra cost to you. I only recommend tools I’ve actually tested inside my own cybersecurity lab. Read the full disclaimer.

In many cases, these links unlock better deals than you’ll find on your own.
No paid reviews. No sponsored opinions. Just real testing and real setups.

If you decide to use them, you’re not just getting a discount — you’re helping keep this lab running.

Leave a Reply

Your email address will not be published. Required fields are marked *