Common issues
Connected but no internet (or unstable)
The VPN reports a connected state, but internet access is missing or drops in and out.
Symptoms
- The client status shows "Connected", but web browsers display timeout errors.
- Internet routing functions intermittently or drops completely every few minutes.
- IP verification websites (e.g.
ifconfig.me,icanhazip.com) fail to load.
Resolution steps
macOS and Linux with GUI
- Switch exit nodes: click the location dropdown and select an alternative node with the lowest available latency (ms).
- Verify wallet balance: click the wallet icon in the top-right corner. Ensure your account balance is sufficient; top up if it is depleted.
- Hard restart: completely terminate the application. Right-click the Gnosis VPN tray icon, select Quit, and relaunch the application.
Linux without GUI
- Check your node and wallet status — verify your Safe and channel balances with your active exit node:
gnosis_vpn-ctl balance
- If your balance is sufficient, restart the background service to flush stuck routing tables:
# Linuxsudo systemctl restart gnosisvpn.service# macOSsudo launchctl kickstart -k system/com.gnosisvpn.gnosisvpnclient
If this behavior persists, attach your client logs to your support ticket. See Finding your logs.
Connection handshake takes over 60 seconds
The client hangs indefinitely in a "Connecting" state when attempting to establish a tunnel with an exit node.
Symptoms
- The interface or terminal status stays frozen in the connecting phase for over 1–2 minutes.
- Specific geographic locations fail to hand over traffic, while others connect instantly.
- Connection eventually succeeds, but only after an unacceptable delay.
Resolution steps
macOS and Linux with GUI
- Click the wallet icon to verify your account balance.
- Pick an alternate exit node with lower latency.
- Restart the app via the tray icon menu if it remains frozen.
Linux without GUI
- Inspect the handshake state:
gnosis_vpn-ctl status
- Force a reconnection to a different node endpoint:
gnosis_vpn-ctl connect --node <alternative_node_id>
If this behavior persists, attach your client logs to your support ticket. See Finding your logs.