VPN Glossary: Subscriptions, Nodes, Protocols, and Traffic Routing Explained
A quick glossary for beginners covering subscription links, nodes and route types, common protocols, traffic rules, and when to use global or rule-based mode.
This VPN glossary starts with your client settings: subscriptions provide configurations, nodes determine what you connect to, protocols define how the two ends communicate, and traffic rules decide which requests use the proxy. Understanding these roles helps you work out whether a configuration failed to import, a route failed to connect, or an app's traffic isn't using the expected exit.
Subscription links: Importing a configuration doesn't establish a connection
A plan subscription refers to billing and usage benefits; a subscription link tells the client where to retrieve its configuration. Configurations typically include node addresses, ports, and protocol settings. Some also include proxy groups or rules, depending on the subscription format and client support. Pasting the link into a browser may display text or start a download, but it won't connect you to the proxy.
- Sign in to the user panel and get a subscription link compatible with your client. For your first setup, follow the getting started guide to find the import option for your platform.
- Add the link in your client's subscription or configuration manager, then update it. Once successful, you should see a configuration or node list. If the list is empty, first check that you copied the full link and that its format is supported.
- Enable the imported configuration, select a node, then turn on the client's system proxy or traffic-tunneling feature. Test with the app that was having trouble rather than relying only on the client's connection icon.
Nodes and routes: entry points, paths, and exits
A node is a connection configuration you can select in your client. The entry point is the server your device connects to first; the exit is the endpoint that sends requests to the destination website. They may be in different regions. A node's region label usually distinguishes routes, but check the provider's naming conventions to see whether it refers to the entry or exit. Websites generally see the exit address making the request, not the name in your client's list.
| Type | How the route works | What to consider |
|---|---|---|
| Direct | Your device connects directly to the remote proxy server, without an additional relay entry point provided by the service. | Performance depends on the public internet route between your local ISP and the remote server. Test with the app you intend to use instead of relying only on speed-test labels. |
| Relayed | Your device connects to an entry point first, and the provider forwards traffic from there to the exit. The entry-to-exit leg may use the public internet or a dedicated link. | A fast response from the entry point doesn't guarantee a fast onward route. During a choppy call, watch for both broken audio and frozen video. |
| IEPL dedicated line | An international Ethernet private line describes a communications link. It can carry traffic between an entry point and an exit, but it isn't a client protocol. | Check which segment the private line covers. The connection from your device to the entry point, or from the exit to the destination, may still use the public internet. A private line doesn't necessarily cover the entire route. |
“Direct” in a route name is different from “direct” as a traffic-routing action. The former still uses a proxy, just without an extra relay entry point; the latter means the request bypasses this client's proxy. If your browser works but a meeting app doesn't, first check whether the client's traffic capture includes the meeting app before switching routes.
Common protocols: check compatibility before network conditions
A protocol is an agreed way for a client and server to communicate. Shadowsocks is an encrypted proxy protocol. VMess and VLESS are both used for proxying, but they are different protocols; VLESS doesn't provide traffic encryption by itself, so assess connection security based on the transport and security layers actually configured. Trojan typically carries authenticated proxy requests over a TLS connection. After importing a configuration, don't change it to another protocol just to try to improve speed. A mismatch with the server settings will prevent a connection.
Hysteria2 and TUIC both use QUIC over UDP, but they aren't compatible with each other. On networks that restrict UDP, even nodes supporting these protocols may fail to connect. And a route that connects successfully isn't necessarily faster on every network. For video calls, whether audio cuts out during a sustained call is a more useful measure than how quickly a webpage loads once.
The same subscription can behave differently across devices. Desktop clients may offer system proxy and virtual network adapter options for capturing traffic; mobile clients often use the system's VPN interface and require permission. Importing a configuration doesn't guarantee that your client version supports every protocol and transport option it uses. If fields are unrecognized or nodes are missing, check the client version and subscription format first.
Traffic routing rules, global mode, and DNS
Rule-based mode uses conditions such as domain names and destination addresses to decide whether traffic goes through the proxy, connects directly, or is blocked. Some clients also let you match apps or processes. Many rule engines evaluate rules in order, applying the first match and using a fallback policy if nothing matches. A webpage may request content from separate login, image, and API domains. If the page loads but sign-in fails, check which rule matched the failing request.
Global mode usually sends traffic captured by the client through the selected proxy; it doesn't mean every connection on the device is captured. For example, apps that ignore system proxy settings may be unaffected when only the system proxy is enabled. Rule-based mode is a better fit if you need local network access or want some sites to connect directly. To troubleshoot rules, temporarily switch to global mode on the same node, retest in the same app, then switch back.
DNS translates domain names into addresses. A connection may use the proxy while DNS queries are still sent by the system, browser, or app to a different resolver. That's a DNS leak in this context: the query didn't follow the DNS route you expected. A domain explicitly set to resolve locally for direct access isn't, by itself, evidence of a problem. If the proxy exit is correct but a domain won't load, check the client's DNS settings and the browser's secure DNS settings separately. An exit IP lookup can't confirm that DNS queries followed the right route.
Troubleshooting connection issues in order
- ✅ Empty list: First check whether the subscription updated successfully and whether the client supports its format. No need to switch regions yet.
- ✅ Node connects but an app doesn't: Check which traffic the client captures and which rules match. If possible, change only the mode to compare.
- ✅ Unexpected region: Open IP lookup in the same browser to check the exit, then confirm that the browser is using the proxy. This result doesn't represent the exit used by other apps.
- ❌ Don't change the node, protocol, and DNS settings all at once. Even if the connection recovers, you won't know which change made the difference.