Technical Reference / Glossary

V2Ray Glossary

From protocol fields in a subscription to routing and traffic-capture options in a client. Browse terms by category and distinguish concepts that are easy to confuse.

Protocols and transports

Protocols define how two sides exchange proxy traffic; transports determine how that traffic is carried, while transport security protects the connection.

VMess

Proxy protocol

VMess is a proxy protocol in the V2Ray ecosystem, with user authentication and traffic encryption. Clients typically identify users with a UUID, and protocol validation is sensitive to clock differences between the two systems. If the settings look correct but the connection fails, check the system clocks first, then review the transport and security fields.

VLESS

Proxy protocol

VLESS is a proxy protocol that typically identifies users with a UUID and does not encrypt traffic on its own. Security mechanisms such as TLS or REALITY are configured separately from VLESS and should be set according to the parameters provided by the server. Recognizing VLESS when importing a configuration does not mean the current core supports every other field.

Trojan

TLS-based proxy protocol

Trojan is a proxy protocol that typically authenticates clients with a password over a TLS connection. A correct password does not guarantee a successful TLS handshake; server name and certificate verification are separate steps. When troubleshooting, distinguish handshake failures from password authentication failures rather than treating both as an incorrect password.

Shadowsocks

SS / encrypted proxy protocol

Shadowsocks (often abbreviated as SS) is a family of encrypted proxy protocols that use a shared key. After importing a share link, make sure the port, encryption method, and password match the server configuration. Implementations may differ in which encryption methods and extended parameters they support, so successfully parsing a link does not guarantee a working connection.

REALITY

Transport security mechanism

REALITY is a transport security mechanism provided by Xray, often used with VLESS. VLESS is a protocol; REALITY is a security-layer option, and they belong in separate configuration fields. Use the original configuration as the source of truth for parameters such as the server public key. Before switching cores, confirm the target core recognizes those fields.

Cores and ecosystem

Clients manage subscriptions and interface settings; cores apply the network configuration. A setting appearing in the client does not mean every available core supports it.

Project V

Open-source project

Project V is the open-source networking project behind V2Ray. In documentation, “V2Ray” may refer to the project or more broadly to the proxy ecosystem built around it. To identify the software actually installed, check the graphical client’s name and the core it is currently using.

V2Fly

V2Ray community project

V2Fly is a community project that continues V2Ray development; its core is commonly known as v2ray-core. The core reads configuration and handles proxying, routing, and DNS, while the client provides the subscription group’s update button. When checking compatibility, refer to the core version in use, not just the client version.

Xray

Independent proxy core

Xray is an independent core developed from the V2Ray core. It shares some protocols and configuration fields with V2Fly, but their security options and extended features are not fully interchangeable. After switching cores, check not only that the core starts, but also that existing nodes and routing rules work as expected.

Subscriptions and nodes

Subscriptions control how the list is updated, nodes determine which configuration is used to connect, and speed tests reflect only the result of a particular test at a particular time.

Subscription

Where the client updates the node list

A subscription lets a client fetch and update a set of node configurations from a single URL. Saving the subscription URL usually just sets up the update source; the list is fetched when you run an update. Subscription URLs may contain access credentials, so hide the full URL and QR code when reporting an issue.

Node

Remote connection configuration

In a client’s list, a node usually refers to a selectable remote proxy configuration. It records the server address, port, protocol, and related parameters; it is not the same as a standalone server. One address can have multiple configurations, and similar names in the list do not guarantee identical connection settings.

Latency

Test duration

Latency is the time a network test takes from start to finish, usually measured in milliseconds. A TCP test may measure only the time it takes to connect to a remote port; it does not confirm that proxy authentication succeeded. Compare nodes using the same test method, and don’t assume lower latency means faster downloads.

Connectivity-test latency

Request test through the proxy route

Connectivity-test latency usually measures how long a request sent through the selected node to a test URL takes to receive a response. Results also depend on the test site, handshake, and temporary network conditions, so latency is not a fixed property of a node. A single timeout may be caused by an issue with the test target and does not, by itself, prove the node is unavailable.

Routing and traffic splitting

To troubleshoot where traffic goes, first check whether the core sees a domain name or an IP address, then identify the matching rule and its outbound route.

Routing rule

Matching criteria and outbound route

Routing rules map matching criteria—such as domains, addresses, and ports—to a specified outbound route. Rules are usually checked in list order, so a broad rule placed too early may match before a more specific one. To find out what happens when no rule matches, check the current configuration’s default behavior; don’t assume the last item in the list determines it.

GeoIP

IP range dataset

GeoIP is a dataset of IP ranges grouped by region or purpose, used by routing rules to match destination addresses. It classifies address ranges; it does not indicate a device’s precise physical location. If a request includes only a domain name, whether an IP-based rule matches also depends on the core’s domain resolution strategy.

GeoSite

Domain category dataset

GeoSite is a categorized domain dataset used by routing rules to match destination domains. The domains included in a category can vary across data sources and versions, and a service’s domains may not all belong to the same category. To troubleshoot a missed match, check the domain actually requested rather than just the service name.

Traffic routing

Choose an outbound route based on rules

Traffic routing sends traffic to different outbound routes—such as direct, proxy, or reject—based on rules. It only handles traffic that reaches the core; apps that are not connected to the client will not automatically follow its rules. If an app is not behaving as expected, check how it connects first, then review the rule content and order.

DNS resolution

Domain name lookup

In client settings, DNS resolution usually means looking up the IP address associated with a domain name. Lookups may be initiated by the app, operating system, or proxy core, and each layer may use a different cache. A domain that fails to resolve and a connection to the destination that times out are different problems; check lookup results and connection results separately.

Client settings

Understand how app connections reach the client, where domains are resolved, and what determines the outbound route once traffic is captured.

System proxy

Proxy settings available to apps

A system proxy is the proxy address and port that the operating system provides to apps that support the setting. Browsers may follow the system settings, while command-line tools often have their own proxy options. If the system proxy is enabled but a program still connects directly, first check whether that program reads the system settings.

TUN mode

Virtual network interface

TUN mode receives IP packets routed by the operating system through a virtual network interface. Apps do not need to read the system proxy settings, but the traffic captured still depends on routing, exclusions, and platform permissions. Whether captured traffic follows the proxy rules also depends on which outbound route the core selects for those packets.

FakeDNS

Temporary address and domain mapping

FakeDNS returns a temporary mapped address for a domain, then uses that address to recover the original domain for subsequent connections. This lets the core match rules against the original domain even when an app’s connection contains only an IP address. The mapped address is not the destination site’s real address; subsequent connections must pass through a path that can recognize the mapping.

DNS leak

Unexpected DNS resolution path

A DNS leak occurs when domain lookups take an unexpected path, exposing queries to an unintended resolver. Routing an app’s connection through a proxy does not mean its DNS queries take the same route. To assess a leak, first identify the intended resolver, then check the resolution settings actually used by the browser, operating system, or client.

Local listening port

Local connection entry point for the client

A local listening port is where a client accepts proxy connections, such as HTTP or SOCKS, on the local machine. It serves a different purpose from the remote port in a node configuration, so changing one does not require changing the other. If you change the local port, also update any browser or command-line tool manually configured to use the old port.