V2Ray Glossary
Browse 24 common terms by configuration task. If you see VMess or REALITY in a node editor, start with protocols and transports. If you're unsure which connections the client handles, compare system proxy and TUN mode.
Each term heading has its own link, making it easy to cite in notes or discussions.
Protocols & Transports
A protocol defines how two sides exchange proxy requests; transport settings determine how the data is delivered. Node editors usually configure these separately.
- VMess
- VMess is a proxy protocol in the V2Ray ecosystem with user authentication and data encryption. The client and server need compatible protocol and transport settings. If a configuration that worked before suddenly fails authentication, check the node details and make sure the device clock is accurate.
- VLESS
- VLESS identifies users with a UUID (a unique identifier) and uses a different message format from VMess. Common deployments require a separate connection security mechanism, such as TLS or REALITY; the protocol name alone doesn't tell you how the connection is protected. Even if two configurations use the same UUID, you can't switch between them simply by changing the protocol name.
- Trojan
- Trojan is a proxy protocol that authenticates clients with a password and typically protects traffic with TLS. A correct password and a valid server certificate are separate requirements. If the log reports a certificate name mismatch, check the server name and certificate rather than repeatedly changing the password.
- REALITY
- REALITY is a connection security mechanism from Xray based on the TLS handshake, often used with VLESS. Configuration screens list VLESS as the protocol and REALITY as a connection security option; they operate at different layers and aren't interchangeable. After importing a configuration, make sure the active core supports this mechanism.
Cores & Ecosystem
Being able to import a configuration doesn't mean the selected core can handle every field in it. When checking compatibility, consider the client and core separately.
- Project V
- Project V is a historic project name in the V2Ray ecosystem, with early V2Ray as one of its core components. It's neither a proxy protocol nor a client you can select from a node list. When older guides mention Project V, check their publication date and context to see whether they mean the original project or a specific V2Ray core.
- V2Fly
- V2Fly is an open-source community project that continues development and maintenance of the V2Ray core. Client documentation often uses V2Fly to distinguish this core implementation from Xray. They share some configuration concepts, but check the active core and its version to see whether a specific setting is supported.
- Xray
- Xray is an open-source core developed from a V2Ray-related code branch, with its own features and release cycle. It handles connections; it isn't a node list or subscription manager. If a setting isn't recognized after switching cores, first check whether it belongs only to the previous implementation.
- Core
- The core is the program that establishes connections, parses protocols, and applies routing rules; the graphical client displays settings and manages configurations. An open window doesn't mean the core started successfully. When troubleshooting, first distinguish between a client interface issue, a core startup failure, and a node handshake failure, then check the logs for the relevant stage.
Subscriptions & Nodes
A subscription update checks whether the list can be retrieved; a node test checks whether a specific connection can be established. Success at one stage doesn't mean the other is working.
- Subscription
- A subscription lets a client retrieve and update node configurations in bulk. The client reads a list from the subscription URL and adds its entries to the relevant groups. A successful update only means the list was fetched and parsed; it doesn't confirm that every node can connect. Subscription URLs may contain access credentials, so hide the full URL when sharing screenshots.
- Node
- A node is a connection-ready configuration in a client, typically including a server address, port, protocol, authentication, and transport settings. One server can have multiple node configurations, and the name shown in the list may be just a label. When troubleshooting duplicates or import errors, compare the configuration fields rather than the names.
- Latency
- Latency is the time a specific network test takes, usually shown in milliseconds. A client may measure the connection time to a node's port or the time it takes to reach a test address through the node. Compare results using the same test method and destination; lower latency doesn't necessarily mean faster downloads.
- Real connection latency
- Real connection latency usually measures the time to reach a test address through a selected node, testing more of the proxy request path than a simple server-port check. Clients may use different start and end points for timing, and results also depend on the test site and current network conditions. If a test times out, try another known-working test address to compare.
Routing & Traffic Splitting
Routing selects how connections that reach the core are handled. An “outbound” is a handling method, such as forwarding a connection to a proxy server or connecting directly to its destination.
- Routing rule
- A routing rule links matching criteria to an outbound—for example, using a domain, destination IP, or port to decide how a connection is handled. V2Ray-family cores typically check rules in order and use the outbound specified by the first match. If a specific rule doesn't seem to work, check whether a broader rule earlier in the list matches first.
- Traffic splitting
- Traffic splitting sends different connections along different paths—for example, sending some requests directly, routing others through a proxy, or blocking selected connections. It doesn't split bandwidth for a single download, and it won't automatically capture connections that never reach the core. To check the result, first confirm the request reached the client, then see which rule it matched.
- GeoIP
- In routing configurations, GeoIP usually refers to sets of IP addresses and subnets grouped by region or purpose. It matches address ranges, not domains, and doesn't establish a server's actual physical location. Set contents can vary by maintainer and data version; when investigating a false match, check which data file is loaded.
- GeoSite
- GeoSite is a categorized set of domain rules that routing can reference by set name instead of listing domains individually. Depending on the data file, entries may match full domains, suffixes, or other patterns. If the core only sees the destination IP and hasn't recovered the domain, a domain set alone can't match it.
DNS & Networking
Separate two questions: which resolver handles the domain lookup, and which route the connection takes after an address is returned.
- DNS
- DNS (the Domain Name System) is commonly used to look up IPv4 or IPv6 addresses for domain names, as well as other record types. Queries may come from an app, the operating system, or the proxy core, and each may use different settings. If a site won't load, first determine whether the domain failed to resolve or the connection failed after an address was returned.
- FakeDNS
- FakeDNS returns a synthetic address from a predefined pool and records its mapping to the original domain. When a subsequent connection reaches the core, the core can use that mapping to recover the domain and apply routing rules. The synthetic address isn't the destination server's real address and can't be used on its own without the processing chain that maintains the mapping.
- DNS leak
- A DNS leak occurs when a domain lookup that should follow a specified route instead reaches an unintended resolver. Routing application traffic through a proxy doesn't mean DNS queries follow the same path, since apps may resolve domains themselves. To check for a leak, first identify the intended resolver, then compare the test results with the client logs.
- DNS cache
- A DNS cache temporarily stores lookup results so they don't have to be resolved again on every visit. Browsers, operating systems, and resolvers may each keep their own caches, so apps may not immediately use new results after DNS settings change. To verify a change, test a domain that hasn't been looked up yet, so an old cached result isn't mistaken for a configuration issue.
Client Features
System proxy and TUN determine how connections reach the client; multiplexing affects how connections are carried, while logs help pinpoint where a problem occurs.
- TUN mode
- TUN mode uses a virtual network interface and system routes to capture IP traffic and pass it to the client. It doesn't rely on apps reading the system proxy settings, but the traffic it captures still depends on routes, exclusions, and permissions. Even with TUN enabled, not every connection that reaches the client is necessarily proxied; subsequent routing rules determine the outbound.
- System proxy
- A system proxy is a set of proxy settings provided by the operating system. Apps that follow these settings send relevant requests to the configured proxy address. It isn't a virtual network adapter, and apps that ignore system settings may still connect directly. If a browser works but another app doesn't, check whether that app has separate network settings.
- Mux multiplexing
- Mux carries multiple logical connections over fewer underlying connections to reduce the overhead of repeatedly opening new ones. It doesn't add bandwidth, and congestion on a shared underlying connection can affect multiple requests at once. Decide whether to enable it based on compatibility at both ends and how requests perform in practice.
- Runtime log
- Runtime logs record events as the client or core starts up, resolves domains, and establishes connections. To troubleshoot, reproduce the issue and check the log around that time for the first error; later errors may be consequences of the initial one. Before sharing logs, redact sensitive information such as subscription URLs and authentication details.
From terminology to configuration
New to proxy clients? Follow the Getting Started guide to import a subscription, choose a node, and test the connection. Once the basics work, see the routing and DNS sections in Advanced Setup to choose which requests connect directly and which go through a proxy.