What Is Split Tunneling?
Learn the difference between full-tunnel and split-tunnel VPN routing, why people use split tunneling, and the privacy, compatibility and troubleshooting tradeoffs.
Last reviewed: September 1, 2026
A VPN kill switch is designed to prevent selected traffic from silently falling back to the ordinary internet path when the VPN connection is no longer available.
If a VPN tunnel disappears, an operating system may normally continue using the regular network connection. That can restore connectivity quickly, but traffic that was expected to use the VPN may now leave through the ordinary route.
A kill switch adds a policy: if the protected VPN path is unavailable, traffic covered by that policy should be blocked or restricted until the VPN is restored. The exact implementation can be firewall-based, route-based, application-specific, or integrated with operating-system VPN controls.
The tunnel can still fail because Wi-Fi drops, the device sleeps, or the network changes. The kill switch is about what happens after that failure. It does not improve a weak Wi-Fi signal or repair an outage.
Some designs try to block all ordinary internet traffic while the VPN is unavailable. Others only protect selected applications or a particular VPN profile. Users should understand which behavior their client actually implements rather than assuming every feature with the same name works identically.
If your VPN drops frequently, investigate the cause instead of treating the kill switch as the fix. The VPN Keeps Disconnecting guide focuses on the underlying stability problem.
A useful kill-switch design makes the blocked state clear. Otherwise a user may think the internet is broken when the client is intentionally preventing fallback traffic. Good status information should distinguish “VPN disconnected and traffic blocked” from “local network unavailable.”
If you test kill-switch behavior, use a controlled non-sensitive activity and verify what happens to the public IP or connectivity when the tunnel is deliberately disconnected. The goal is to confirm the documented behavior, not to create repeated unstable network conditions during important work.
No. Its purpose is to control fallback traffic when the VPN is unavailable.
No. Split tunneling decides which traffic should use the VPN. A kill switch controls what happens when a protected VPN path is unavailable.
That depends on how important it is that selected traffic never fall back to the ordinary route and on the behavior of the specific device and client.
Learn the difference between full-tunnel and split-tunnel VPN routing, why people use split tunneling, and the privacy, compatibility and troubleshooting tradeoffs.
A practical guide to repeated VPN disconnects, unstable Wi-Fi, network changes, sleep and roaming, server reachability, and how to isolate the real cause.
Understand DNS, IPv6, IP and WebRTC leaks and how to test them.
Important limits: malware, phishing, accounts, tracking and device security.
Practical VPN security habits for Windows users.
Readers who want to test these concepts with a VPN client can use the ZBEVPN Download page. For connection-specific problems, start with the Guides & Knowledge Center.