DNS leaks are easy to misunderstand because a VPN can appear connected while some name-resolution requests still use the local network provider. A browser may show the expected public address, yet DNS queries, WebRTC connection candidates, or applications outside the browser may follow a different path. This does not automatically mean that every connection is exposed, but it does mean that the privacy configuration has not been verified completely.
This practical guide explains how to check DNS routing, identify WebRTC exposure, review kill switch behavior, and fix common configuration mistakes. The goal is not to rely on one website or one button. A useful test combines the VPN client status, operating-system settings, browser behavior, DNS results, and what happens when the encrypted connection is interrupted.
What a DNS leak means when a VPN is connected
DNS, or the Domain Name System, translates a hostname such as example.com into an IP address. Every time an application needs to reach a hostname, it may first ask a DNS resolver for the destination address. Depending on the operating system, browser, VPN client, and application, that request can be handled by the system resolver, a browser-level resolver, a remote VPN resolver, or a manually configured third-party service.
A DNS leak occurs when a request that should be protected by the VPN is sent to a resolver outside the intended tunnel. The visible website may still load through the VPN’s exit route, so checking only the public IP address can miss the problem. A resolver may learn which domains are being requested even when it cannot see the complete contents of the connection. For privacy-focused use, the resolver path is therefore part of the overall connection design rather than a minor technical detail.
DNS leaks can have several causes. The client may not capture DNS traffic in its current mode. The operating system may retain an older resolver after the tunnel starts. An application may use its own encrypted DNS method. Split routing rules may intentionally send local or selected domains outside the tunnel. Another VPN, proxy, security product, virtual machine, or manually configured network adapter may also change the path.
4
Main checks in this guide
90+
Countries available from MeeVPN
200+
Available routes
14 days
Refund period
DNS leaks and WebRTC exposure are different problems
DNS testing and WebRTC testing are often placed on the same privacy checklist, but they examine different mechanisms. A DNS test observes which resolvers answer name-resolution requests. WebRTC is a browser technology used by real-time communication features such as voice, video, and peer-to-peer connection setup. During connection negotiation, a browser may reveal network candidates that help another endpoint discover possible routes.
Modern browsers apply restrictions to local and public address discovery, and the exact result depends on browser version, permissions, extensions, operating-system networking, and the site’s WebRTC implementation. A displayed address is not automatically proof that a complete browsing identity has been exposed. It is a signal that deserves investigation: compare it with the VPN exit address, identify whether it is a local, private, public, or IPv6 address, and repeat the test after changing the browser’s relevant settings.
DNS protection is normally controlled by the VPN client, operating system, or browser DNS settings. WebRTC behavior is mainly controlled by the browser and by the way the client captures traffic. Disabling browser WebRTC can reduce exposure, but it may affect meeting platforms or browser-based calling. A better approach is to understand the trade-off and test the browser you actually use instead of assuming that one setting applies to every application.
| Check | What it examines | Typical control | Possible false assumption |
|---|---|---|---|
| Public IP | The apparent internet exit address | VPN route and client connection | The expected IP does not confirm DNS protection |
| DNS test | Resolvers receiving hostname requests | VPN DNS handling, system resolver, browser DNS | One clean result covers every application |
| WebRTC test | Browser connection candidates | Browser privacy settings and traffic capture mode | Every displayed address represents a serious public leak |
| Kill switch test | Behavior when the tunnel is unavailable | Client firewall or block rules | A disconnected icon always means traffic is blocked |
How to run a reliable DNS leak and WebRTC test
Testing is most useful when the starting conditions are recorded. Close applications that may generate background traffic, note whether the client is using global routing, rules, or TUN mode, and disconnect other VPN or proxy tools. If the device is on public Wi-Fi, first confirm that the local network itself is working normally. Do not use a single screenshot as permanent evidence because resolver selection and browser behavior can change after reconnecting.
Prepare the test environment
- Restart the VPN client or disconnect and reconnect it so that the test begins with a fresh tunnel.
- Record the selected route, connection mode, protocol, and whether split routing is enabled.
- Open a private browser window without unfamiliar extensions, then visit a reputable IP and DNS testing service.
- Check the public IP first. It should correspond to the selected VPN exit region rather than the local internet connection.
- Run the DNS test more than once, including a standard test and an extended test when available.
- Review every resolver shown, not just the headline result. Look for the network provider, workplace network, hotel network, or home broadband operator.
- Run a WebRTC test in the browser you normally use and compare the candidates with the public IP and local interface addresses.
Testing while the VPN is disconnected provides a useful baseline. It shows which resolvers and addresses belong to the ordinary network. Reconnect the VPN and compare the results rather than judging an unfamiliar resolver name without context. A resolver operated by the VPN provider may be expected; a resolver belonging to the local access provider may indicate that DNS requests are bypassing the tunnel.
Check applications beyond the browser
Browsers can use their own DNS-over-HTTPS or DNS-over-TLS settings, while desktop applications may use the operating system resolver. Command-line tools, software updaters, game launchers, messaging clients, and virtual machines can also follow different network paths. If privacy matters for a particular application, test that application’s behavior separately. A clean browser result cannot certify software that does not use the browser’s resolver.
On Windows, inspect the active adapter and DNS configuration after connecting. On macOS and Linux, review the active network service or resolver manager. On Android and iOS, check private DNS, per-network DNS, and any browser-specific secure DNS options. The menu names vary by version, so focus on whether the device is using a manually configured resolver, a local gateway, or the VPN client’s DNS route.
How to fix common DNS leak causes
Start with the VPN client rather than immediately editing system files. Look for options named DNS leak protection, use VPN DNS, prevent DNS outside the tunnel, or similar. Enable the setting, reconnect, clear the local DNS cache if appropriate, and repeat the test. The wording differs between official clients and compatible clients such as Clash Verge, sing-box, or Shadowrocket, so do not assume that a similarly named switch has identical behavior.
Next, review the traffic-capture mode. A system proxy usually covers applications that honor the operating-system proxy. A TUN or virtual-adapter mode can capture a broader range of IP traffic, but it may conflict with firewalls, virtual machines, endpoint security tools, or another virtual adapter. Rule-based routing may deliberately keep local services direct. Confirm that DNS requests are handled by the intended policy and are not excluded accidentally by a rule.
On the operating system, remove obsolete manual DNS entries only if you understand why they were added. A public resolver configured for ordinary browsing is not necessarily a leak when the VPN is disconnected, but it may bypass the VPN’s preferred resolver after connection. Reconnect the network after changes so the device does not continue using stale resolver information. If the problem returns after every restart, inspect startup applications and network adapters for competing software.
Browser secure DNS requires special attention. If the browser sends encrypted DNS directly to a provider, the request may not use the VPN client’s DNS policy. This can be acceptable for some users, but it makes the browser’s behavior different from the rest of the device. Either configure the browser to follow the intended VPN policy or deliberately document that it uses a separate resolver. Consistency matters when you are trying to understand what information leaves the device.
- ✅ Enable the client’s DNS protection option when available.
- ✅ Reconnect after changing DNS, TUN, or routing settings.
- ✅ Test the operating-system resolver and browser resolver separately.
- ✅ Keep only one active VPN or proxy capture layer during diagnosis.
- ❌ Do not assume global mode captures applications that ignore the system proxy.
- ❌ Do not paste a subscription link into public troubleshooting posts or screenshots.
Review the kill switch instead of trusting the connection icon
A kill switch is intended to block selected or all traffic when the VPN tunnel is unavailable. The exact scope differs between clients. Some block traffic only after an established connection drops; others apply a firewall rule as soon as the feature is enabled. Some allow local-network access, while stricter modes block local devices as well. Read the client description and identify which applications or destinations remain exempt.
Perform a controlled failure test. Connect the VPN, open a simple page, and confirm the route is active. Then use the client’s disconnect function or temporarily change the network connection in a way that does not damage the device. Try a new browser request and observe whether it is blocked or falls back to the ordinary network. Reconnect the VPN and confirm that normal traffic returns. Do not interrupt a work call or an important file transfer while learning how the setting behaves.
Also test a client restart and a sleep-and-wake cycle. A setting that works after a manual disconnect may behave differently during startup, network switching, or wireless roaming. On mobile devices, background restrictions and battery management can suspend the VPN service. On desktop systems, firewall permissions and administrator rights may affect whether the block rule is installed correctly.
A kill switch cannot correct every privacy issue. It may not control a browser’s own DNS service, an application using a separate network stack, or traffic intentionally excluded by split tunneling. It also does not make an unsafe Wi-Fi network trustworthy. Treat it as one layer in a broader configuration: secure account credentials, current software, sensible browser settings, and careful handling of sensitive links remain important.
Apply the checklist to public Wi-Fi, banking, and everyday browsing
Public Wi-Fi creates a different risk profile from a trusted home network. Before connecting, disable automatic access to unknown networks where possible, confirm the network name, and avoid entering sensitive credentials until the connection is established securely. After starting the VPN, check the public IP and DNS behavior. If the client repeatedly disconnects, enable the kill switch or stop sensitive work until the route is stable. Local printer, casting, or captive-portal access may require a carefully documented exception.
For banking and other high-value accounts, privacy is only one part of the decision. Banks may challenge a sign-in from an unfamiliar region, device, or IP reputation. Keep multi-factor authentication enabled, use the official banking application or a carefully verified website, and do not bypass a security warning simply because the VPN is connected. If a financial service rejects the selected exit route, try a different legitimate route or use the normal connection according to the institution’s policy.
For everyday browsing, rule-based routing can reduce unnecessary disruption to local services while sending selected traffic through the VPN. However, every exception is also a decision about which traffic is outside the tunnel. Review rules after importing or updating a configuration. For privacy-focused use, choose a service and client that clearly explain supported platforms, DNS handling, protocol behavior, and kill switch scope rather than selecting only by the number of listed routes.
| Scenario | Recommended checks | Important caution |
|---|---|---|
| Public Wi-Fi | VPN status, DNS result, kill switch, captive portal | Local-network exceptions may be necessary but should be understood |
| Banking | Verified domain, MFA, device security, route consistency | A bank may reject unfamiliar VPN exits or request extra verification |
| Daily browsing | Browser DNS, WebRTC behavior, split-routing rules | Local exceptions can send selected requests outside the tunnel |
| Privacy-focused selection | Platform support, DNS controls, protocols, kill switch documentation | More routes do not automatically mean better privacy or compatibility |
Choose a VPN client with verifiable privacy controls
For most users, an official client is the simplest starting point because its DNS handling, update path, platform permissions, and kill switch are documented together. MeeVPN supports Windows, macOS, iOS, Android, and Linux. After signing in, users can obtain the client or import a subscription link into a compatible application. Compatible clients may offer more detailed routing control, but they also require a clearer understanding of profiles, DNS modes, rule providers, TUN permissions, and update behavior.
Protocol choice can affect compatibility and transport behavior, but a protocol name alone does not prove that DNS is protected. Shadowsocks, VMess, Trojan, Hysteria2, and WireGuard describe different connection technologies; the client still needs to route DNS correctly and apply the selected policy. Check whether the software supports the protocol and configuration format you receive, whether IPv6 is handled, and whether the kill switch covers the traffic you care about.
MeeVPN provides access to 90+ countries and 200+ routes, with unlimited simultaneous devices. Those figures can help when comparing route availability and household use, but they should not replace a local test. Try the routes relevant to your location and tasks, observe whether the client reconnects cleanly, and repeat DNS and WebRTC checks after updates. A service is easier to evaluate when its refund terms are clear; MeeVPN offers a 14-day refund promise for an unsatisfactory first payment.
5
Supported platform families
Unlimited
Simultaneous devices
90+
Countries
200+
Routes
Finally, protect the subscription link itself. It can contain access information that lets a client retrieve route configurations. Store it in a password manager or another private location, avoid sending it through public channels, and revoke or refresh it through the user panel if you believe it has been exposed. Privacy is not only about packet routing; credential handling is part of the same security model.
Frequently asked questions
Does an expected VPN IP prove that there is no DNS leak?
No. The public IP test and DNS test examine different paths. Run a DNS test and inspect the resolver names, then repeat the check for the browser and operating-system applications that matter to you.
Should I disable WebRTC completely?
Not necessarily. Disabling or restricting WebRTC can reduce browser address discovery, but it may affect browser-based voice and video features. Test the browser and services you use, then choose a setting that matches your privacy and communication requirements.
Why is the browser clean while another application shows a different resolver?
The browser may use its own secure DNS setting while the application uses the operating-system resolver. Check application-specific settings, VPN capture mode, split-routing rules, and competing network adapters instead of assuming that one result represents the whole device.
What should I do if the leak returns after reconnecting?
Record the client mode and resolver before changing anything. Then check startup software, manual DNS entries, other VPNs or proxies, virtual adapters, browser secure DNS, and kill switch permissions. Change one item at a time and retest after a fresh connection.
A repeatable privacy checklist
Use the following checklist whenever you install a new client, import a new configuration, change browser settings, or move between home internet and public Wi-Fi. The most reliable result comes from repeating the same checks under the same conditions and keeping a short record of the selected route and mode.
- ✅ Confirm the VPN connection and selected route before opening sensitive services.
- ✅ Check the public IP, DNS resolvers, and WebRTC behavior separately.
- ✅ Test both the browser and applications that use the operating-system network stack.
- ✅ Verify what the kill switch blocks during disconnect, restart, and network changes.
- ✅ Review split-routing exceptions and browser secure DNS settings.
- ✅ Keep client software, account credentials, and subscription links private.
- ❌ Do not treat one clean test or one connection icon as permanent proof of privacy.