If you already have a share link, identify the protocol from its prefix instead of forcing a different type in your client. Both VMess and VLESS commonly use a UUID to identify users. VMess has its own protocol-level encryption options, while common VLESS configurations with encryption=none rely on a separately configured connection security layer. Know which fields control the protocol, transport, and connection security to verify imported settings and troubleshoot errors.
Encryption and authentication: VMess vs VLESS
In a client’s node editor, two different security settings are easy to confuse. One belongs to the proxy protocol itself, such as VMess data encryption. The other protects the connection between your client and the server, such as TLS. They are separate settings: VMess can also use TLS, and encryption=none in a typical VLESS configuration does not mean TLS is disabled.
| Setting to check | VMess | VLESS |
|---|---|---|
| Share link prefix | vmess:// | vless:// |
| User credentials | Usually a UUID, used for protocol authentication | Usually a UUID, which the server uses to identify the user |
| Protocol-level encryption | Data may use AES-128-GCM or ChaCha20-Poly1305, depending on the node configuration or the client’s automatic settings | Common configurations use encryption=none, meaning there is no protocol-level encryption here |
| Connection security | TLS can be configured separately | TLS is common; Xray configurations that support it can also use REALITY |
A UUID is usually 36 characters long, including four hyphens. It is neither a server address nor a TLS certificate. Even if two devices connect to the same address and port, a wrong UUID may cause the server to reject the request. Keep the hyphens when copying, and avoid including leading or trailing spaces. When asking for help with a screenshot, hide the full UUID, share link, and subscription URL.
First, identify which setting “encryption” refers to
If a VLESS node shows both encryption=none and security=tls, keep both settings. The first refers to protocol-level encryption; the second means the outer connection uses TLS. Don’t change either one just to make the values look consistent.
Understanding transport: TCP, WebSocket, gRPC, and TLS
Clients usually list TCP, WebSocket, and gRPC under “Transport.” TCP is a transport-layer protocol, while WebSocket and gRPC are higher-level ways to carry data. They appear side by side in client interfaces because each determines how proxy data is delivered to the other end. TLS belongs under connection security. A port number like 443 alone doesn’t tell you which transport a node uses or whether TLS is enabled.
- Address and port: If the original node specifies
8443, don’t change it to443just because TLS is enabled. First verify the server address, then check the port. The node name is only a display label in the list. - WebSocket: After selecting WebSocket as the transport, check the path and any explicitly required Host value. For example, if the original path is
/edge/ws, keep its leading slash and capitalization. Don’t add the path to the server address or use it in place of the TLS server name. - gRPC: Check the service name, which may appear in the client as
serviceName. It is not a WebSocket path. Copying a path from another node may connect to the same port but still fail to find the service. - TLS: Check that the server name (SNI) matches the original configuration. The connection address, WebSocket Host, and SNI serve different purposes. Their values may or may not match; don’t substitute one for another just because they all look like domain names.
A VLESS configuration may also include flow. It is not part of the UUID, nor is it a WebSocket or gRPC name. If the original configuration specifies a flow-control value, check that your client’s core supports that combination. If the original leaves it blank, don’t copy a value from another node.
Choosing a client: check the node requirements and core
v2rayN is a desktop client that lets you choose a supported core based on the node’s requirements. v2rayNG is an Android client that uses the Xray core, while v2flyNG is an Android client that uses the v2fly core. A protocol name appearing in the client only means the interface recognizes that type; it does not confirm support for the required transport, security layer, or flow-control combination.
Existing subscriptions and available nodes
Import using the original node settings
RecommendedKeep VMess for a vmess:// link and VLESS for a vless:// link. If the node requires a specific Xray feature, choose the corresponding core. Changing the protocol name in your client does not change the server’s protocol.
Best for: importing a working node or subscription for the first time, or switching devices
Test two configured nodes separately
If your subscription includes both VMess and VLESS, import each complete configuration and compare how they connect. Don’t copy one node and change only its protocol or link prefix. That tests mismatched settings, not the difference between protocols.
Best for: comparing real-world performance when the server offers both types of connection
Where to import a subscription URL or individual share link
First, identify whether you have a subscription or a share link
Desktop: v2rayN
- Subscription URL: Using the v2rayN 6.60 menus as an example, open “Subscription Group” → “Subscription Group Settings,” add the URL, and save. Then return to the “Subscription Group” menu and update the subscription.
- Individual link: Copy the entire link, then choose “Import batch URLs from clipboard” from the “Servers” menu. Open the imported node and verify its protocol, transport, and connection security settings.
Android: v2rayNG
- Subscription URL: Open “Subscription Group Settings” in the side menu, tap “+” in the top-right corner to add a group, and save. Return to the main screen and update the subscription from the top-right menu.
- Individual link: Copy the link, tap “+” in the top-right corner of the main screen, and choose “Import from clipboard.” Menu labels may vary slightly by version. After importing, check that the new node appears in the current list.
A subscription URL fetches a group of nodes; an individual share link adds one node. If the expected nodes don’t appear after a successful update, check the selected group first, then review the update log.
Troubleshooting after import: read the error, then change one setting at a time
A node appearing in the list only means the client read its configuration; it doesn’t mean the server accepted the connection. Keep the original configuration and check the core log for the latest failure. Use the error to identify which layer failed, then verify the address, connection security, or authentication fields as appropriate. Change one setting at a time so you can tell what fixed the issue.
Still can’t connect after changing protocols? Start here
Restore the original protocol first. For a connection timeout or refusal, check the server address and port against the original node, then confirm that your current network can reach that endpoint. Changing the UUID is unlikely to help at this stage because the connection may not have reached authentication yet.
VMess authentication failed, but the UUID looks right. What next?
Reimport the original share link and compare the UUID. Then check whether automatic system time is enabled and sync the clock. VMess authentication is sensitive to clock skew; changing the time zone alone does not correct the system clock. Modern VMess AEAD configurations typically use alterId=0. If an older configuration specifies another value, ask the node operator about compatibility instead of trying values at random.
The port is reachable, but the TLS certificate name doesn’t match
Open the node editor and compare the SNI with the server name specified by the provider. If the connection address is an IP address and the certificate is issued for a domain, the SNI may not be that IP. Keep certificate verification enabled. If the settings match the original configuration and the problem persists, ask the operator to check the certificate configuration.
Both nodes work. How can I tell which is faster?
Alternate tests using the same device, network, and test target. Run at least three tests per node, recording the median latency and download time for the same file. A responsive test port doesn’t guarantee smooth browsing or downloads. If the nodes use different routes or servers, the results can help you choose the better-performing node, but they can’t be attributed to VMess or VLESS itself.