To check whether your VPN is working, compare your public IP before and after connecting on the same device, then check DNS and confirm that the apps you need are using the expected route. For example, your team may route a US store dashboard through a US server while keeping a local shipping tool on a direct connection. Different exit IPs may be exactly what you expect.
Compare your public IP before and after connecting
Your public IP is the address websites see as the source of your connection. It’s not your device’s local network address, and it may differ from the server address shown in your VPN client. To see whether your VPN changes your public IP, keep the device, network, browser, and test page the same; change only the VPN connection status.
- On a network where direct access is allowed, disconnect the VPN and open IP Lookup. Note the public IP and address type. For a baseline, you only need to visit the test page; there’s no need to open a store dashboard or work system while disconnected. If your organization requires an always-on connection, don’t disable protection for testing. Check the exit details provided by your administrator instead.
- Connect to a server in the intended region. Once the client shows that it’s connected, reload the same test page. If needed, quit and reopen the browser so it doesn’t reuse a connection from before the switch. Record the new address, current server, and client mode.
- Compare like with like. If both your device and the test service support IPv4 and IPv6, test each exit separately. An IPv4-only result doesn’t tell you whether IPv6 traffic also uses the VPN. Mark any untested address type as unverified.
A changed IP means the test request used a different exit. Before attributing the change to your selected server, rule out browser proxies, extensions, and other VPNs. Location labels come from IP geolocation databases and may be out of date. If a label looks wrong, repeat the test under the same conditions, then check the server settings in your client instead of relying on the city shown on a map.
What the result tells you: An expected exit IP on a test page confirms only that this browser request to that page used the expected exit. To confirm that your store dashboard uses the same route, you’ll need to check the dashboard’s own requests too.
Distinguish DNS resolution from web traffic
DNS translates domain names into IP addresses; your web connection carries the site’s data. These requests don’t necessarily take the same route. Your system may use a resolver supplied by the network or configured by the VPN client, while your browser may resolve names independently using Secure DNS. A system-level tool’s results don’t necessarily show which resolver your browser is using.
First, check whether your VPN client has a DNS interception option, then review your browser’s Secure DNS settings. To test the browser’s actual behavior, use a trusted DNS test page. It typically has the browser request a unique test domain, then reports the recursive resolvers that queried it. Keep the same browser settings before and after connecting, and run a fresh test so cached results aren’t mistaken for new queries.
A DNS test reports the resolvers observed by the test provider; it doesn’t show every step along the query path. Seeing a public DNS service doesn’t mean the query bypassed your VPN: encrypted DNS requests from your browser may also travel through the VPN. Conversely, if your settings require DNS to use the VPN but client diagnostics show test-domain queries going through your local network, investigate a possible DNS leak. If your settings allow local DNS, judge the result against those rules rather than the resolver’s name alone.
Check split tunneling app by app
Write down which route each app should use, then check which rules the client actually applies. System proxy settings work only for programs that use that proxy; a browser extension also affects only its own scope, not the whole device. Even when a client routes traffic through a virtual network adapter, domain rules, excluded apps, and IP address types can still affect the route.
- ✅ Browser: Use the browser profile you actually use for store tasks, send a fresh request, and check the client’s connection log for the destination domain and selected route at that time. A rule matched by a test page may not apply to your store’s domain.
- ✅ Desktop apps: Quit and relaunch the app, then perform a low-risk action such as refreshing a list. If the client shows the process and destination, check whether the expected rule was applied. If the service provides login or access logs, you can also compare the source of that request.
- ✅ Command-line tools: Check whether the tool reads proxy environment variables and whether it has its own proxy settings or direct-connection exceptions. A successful browser test doesn’t mean a terminal, sync tool, or its child processes use the same settings.
- ❌ Don’t assume traffic used the VPN just because a page loads. An existing connection, a cached page, or an endpoint that’s normally accessible directly can all make a page look fine. If your tools can’t show evidence of the route, mark the result as unverified.
If the home page loads but uploads or sign-ins fail, check the actual destination of the failed request. Dashboards, authentication, and file uploads may use different domains with different routing rules. When retesting, stick to necessary, low-risk actions. Don’t repeatedly submit forms or place transactions just to check the route.
Common issues and retest order
| What you see | Check first | Next step |
|---|---|---|
| Connected, but IP unchanged | Check whether the test domain is set to connect directly and whether the browser is reusing an old connection | Reopen the browser and send a fresh request; check the rule-match log for that domain |
| Browser IP changed, desktop app IP didn’t | Check whether the proxy applies only to the browser or the app is excluded | Check the app’s own settings, then retest with a fresh request from that app |
| IP changed, but some domains fail to resolve | Check whether the resolver is reachable and whether the failing domain belongs to your organization’s internal network | Test public and internal domains separately; keep any DNS settings required by your organization |
| Results change after switching networks | Check whether the client has reconnected and whether the new network uses different routing or DNS settings | Wait for the connection to stabilize, then retest the IP, DNS, and target app in order |
Change just one variable per retest. First confirm your current network and server, then check the mode and rules, and finally review browser- or app-specific settings. If you switch servers, change DNS, and reinstall the client all at once, even a fix won’t tell you what solved the problem. When contacting the Help Center, include your device OS, client mode, test time, affected app, and rule-match results. Before sharing screenshots, hide subscription links, account details, and business data.