Glossary index / 06 categories · 28 terms

V2Ray Glossary

Look up a term from the client menu, node list, or JSON config, then browse by category. Each term has its own link, useful for annotating config notes or troubleshooting logs.

01 / PROTOCOLS

Protocols & Transport

Protocols define how the two sides identify requests; transport determines how those requests travel over the network. When importing a node, both must match the server.

VMess

Protocols & Transport

VMess is a proxy protocol in the Project V ecosystem that typically identifies users with a UUID. Transport methods such as WebSocket must be configured separately; the VMess name alone doesn't tell you which one is in use. If an older config fails authentication after migration, check the clocks on both ends and whether the core still supports the legacy parameters, as well as verifying the identifier.

VLESS

Protocols & Transport

VLESS is a proxy protocol that typically identifies users with a UUID. Seeing encryption: none only means that VLESS protocol-layer encryption is disabled in this config; it doesn't tell you whether the entire connection is encrypted. Also check whether transport security is enabled with TLS or REALITY.

Trojan

Protocols & Transport

Trojan is a password-authenticated proxy protocol commonly used to carry proxy traffic over a TLS connection. The password identifies the user to the server, while certificate verification confirms the server's identity. If the logs show a certificate or hostname verification error, changing the password won't fix it.

Shadowsocks

Protocols & Transport

Shadowsocks is an encrypted proxy protocol. Typical configs include the server address, port, encryption method, and password. Enter the parameters as specified by the server; checking only the password isn't enough. Before choosing a newer encryption method, make sure the core used by the client supports it.

02 / CORES & CLIENTS

Cores & Clients

The client provides the interface; the core runs the config. Choose a platform first, then check that it supports the protocols and core features you need.

Xray

Cores & Clients

Xray is a proxy core developed from the V2Ray ecosystem, with support for features such as REALITY and XTLS Vision. A client being able to import a share link doesn't mean its current core recognizes every parameter in that link. If a config fails to load, first check which core the client actually launched.

V2Fly

Cores & Clients

V2Fly is a community project continuing Project V's open-source development and maintaining the core V2Ray implementation. In menus, v2fly usually refers to this core; v2flyNG is a GUI client. When reading config documentation, first check whether the author means the project, the core, or the client.

Core

Cores & Clients

A core is the program that reads the config and handles connections. It accepts traffic through inbounds, then selects an outbound based on the rules. A working interface only means the client is running; check the status or listen information in the logs to confirm the core started successfully.

GUI client

Cores & Clients

A GUI client manages nodes, edits settings, and starts or stops the core. The menus and available features differ across v2rayN, v2rayNG, and v2flyNG. If you can't find an option shown in a screenshot, check the client comparison, then verify the menu name in the relevant product.

03 / JSON STRUCTURE

Config Structure

Read the config by following the flow: where traffic enters, what rules match it, and where it goes next. See the config reference for the field hierarchy and complete examples.

inbounds

Config Structure

inbounds is an array of inbound configs that defines how the core accepts connections. Each inbound can have its own protocol, listen address, and port—for example, separate SOCKS and HTTP entry points. If an app uses the wrong entry point, local requests may never reach the core, even when the remote node is reachable.

outbounds

Config Structure

outbounds is an array of outbound configs that defines how the core handles connections to be sent. An outbound can forward traffic to a remote server, connect directly to the destination, or block the request. When troubleshooting, don't just check that an outbound exists; verify that the matching rule actually references it.

listen address

Config Structure

listen specifies the local address an inbound binds to; port specifies the port number. 127.0.0.1 is the IPv4 loopback address for local programs, not the node's server address. If you change the local listening port, update the port in any apps that use that entry point manually.

outboundTag reference

Config Structure

In a routing rule, outboundTag references an outbound's tag. For example, if a rule references direct, find the outbound tagged direct and check which protocol it uses. A tag is just a name, so if you rename the outbound, update its references too.

04 / ROUTING & DNS

Routing & DNS

First check whether app traffic reaches the client, then check how domains are resolved and which routing rule matches. These three steps don't necessarily follow the same path.

Routing rule

Routing & DNS

Routing rules select an outbound based on conditions such as the domain, destination address, or inbound tag. In common core configs, rules are checked from top to bottom, and evaluation stops at the first match. When adding an exception, check whether an earlier, broader rule already catches the connection.

Traffic routing

Routing & DNS

Traffic routing sends different connections through different outbounds. For example, an office intranet can connect directly while other specified requests go through a proxy outbound. The inbound mode determines which connections the client can receive; routing only determines where those connections go next.

GeoIP

Routing & DNS

GeoIP matches IP address ranges grouped by region or purpose in routing rules. It matches the connection's destination address, not the user's location. If a domain resolves to multiple addresses, the result also depends on whether the core resolves the domain for that rule and which addresses it gets at the time.

GeoSite

Routing & DNS

GeoSite references a group of domain matches by name, making them reusable across multiple rules. Which domains are included depends on the data file in use; the results aren't queried in real time. If a connection only provides a destination IP, GeoSite alone can't tell you which domain it originally belonged to.

FakeDNS

Routing & DNS

FakeDNS assigns a synthetic IP address to a domain and stores the mapping in the core. When a later connection returns to the same core, the core can use that mapping to recover the domain and route by domain. The synthetic address isn't the remote host's address and can't be treated as an ordinary DNS result outside this mapping.

DNS leak

Routing & DNS

A DNS leak occurs when domain queries that should follow a specified route are sent to an unintended resolver. Web requests going through a proxy doesn't mean the app's DNS queries use the same route. To troubleshoot, compare the app's resolution method, system DNS, and client DNS—not just the public IP seen by websites.

System proxy

Routing & DNS

A system proxy is an operating-system setting used by apps that read it. When the client enables the system proxy, apps that follow the setting send requests to a local proxy entry point; apps with their own network settings may ignore it. If only a few apps have connection problems, compare their proxy settings first.

TUN mode

Routing & DNS

TUN mode receives IP packets routed through a virtual network interface by the operating system. Apps don't need to read the system proxy, so TUN can also handle traffic from some apps that ignore it. The actual coverage depends on routing, exclusions, and permissions.

05 / SUBSCRIPTIONS & NODES

Subscriptions & Nodes

Updating a subscription, selecting a node, and testing a connection are three separate actions. For the steps in order, see the getting started guide.

Subscription

Subscriptions & Nodes

A subscription lets a client fetch or update a set of node configs, usually via a link. Refreshing a subscription updates the config list; it doesn't select a node or check whether one works. Subscription links may contain access credentials, so avoid posting the full link publicly when troubleshooting.

Node

Subscriptions & Nodes

A node is a selectable connection profile in a client's list; it doesn't necessarily correspond to a separate server. It typically includes an address, port, authentication, and transport settings, and several nodes may share the same address. The remarks field is just for identifying entries; renaming it doesn't change the connection parameters.

Latency

Subscriptions & Nodes

Latency is the time taken by a specific test, usually shown in milliseconds. A client might measure network round-trip time or the time to establish a TCP connection; these figures aren't directly comparable. Latency isn't download speed either. For a fair comparison between nodes, use the same test method.

Connection test latency

Subscriptions & Nodes

Connection test latency usually measures the time it takes to send a request through the selected node to a test address and receive a response. Compared with a port-only test, it also covers the proxy protocol and an actual request, but results still depend on the test address and current network conditions. One successful test only confirms that request worked; it doesn't mean other destinations are reachable.

06 / TRANSPORT SECURITY

Encryption & Security

Protocol authentication, transport encryption, and flow control operate at different layers. If the logs show a handshake error, first identify which layer failed, then check that layer's settings.

TLS

Encryption & Security

TLS encrypts transport connections and protects their integrity. It typically verifies server identity using a certificate. The client checks the certificate's validity period, trust chain, and expected server name. If the port is reachable but certificate verification fails, check these conditions first rather than disabling verification.

REALITY

Encryption & Security

REALITY is a transport security mechanism provided by Xray and commonly used with VLESS. VLESS defines how proxy requests are handled, while REALITY helps establish a protected connection. They often appear in the same node, but VLESS is a protocol option and REALITY is a transport security option.

XTLS Vision

Encryption & Security

XTLS Vision is a flow-control mode in Xray, commonly specified in configs as xtls-rprx-vision. Under the right conditions, it can reduce redundant processing of data that's already encrypted. It isn't a standalone proxy protocol and can't replace the TLS or REALITY handshake.

Handshake

Encryption & Security

A handshake is the process in which two parties exchange parameters and establish a session before transmitting application data. A successful TCP handshake only means the transport connection is established; it doesn't mean the subsequent TLS handshake or proxy authentication succeeded. When recording an error, include the stage name—it's more useful for troubleshooting than simply saying "can't connect."

Download a client