Configuration Reference / Glossary

V2Ray Glossary

First distinguish the protocol, core and client; then identify whether the issue lies with traffic entry, routing or the node connection.

Protocol Types / Transports / Security Layers

Protocols & Transports

The protocol defines how the two endpoints exchange proxy data; the transport determines how that data is carried; the security layer protects the connection. Clients usually show these as separate fields, and using the same protocol name does not mean the other settings match.

VMess

Proxy protocol

VMess is an early proxy protocol in the V2Ray ecosystem, with user authentication and data protection mechanisms. The client and server must use the same protocol and agree on the transport and security settings. Authentication depends on time information, so a significantly incorrect device clock can prevent a session from being established even when the server address is reachable.

VLESS

Proxy protocol

VLESS is a lightweight proxy protocol that typically identifies users by UUID. The protocol itself does not encrypt traffic; connection security is provided by a separate layer such as TLS or REALITY. A node marked VLESS does not reveal its transport, and its settings cannot be substituted for a VMess configuration.

Trojan

Proxy protocol

Trojan typically runs over a TLS connection and uses a password for proxy authentication. A successful TLS handshake and successful password authentication are separate steps; the former does not guarantee the latter. When reading connection logs, distinguish certificate or server-name errors from proxy authentication errors.

Shadowsocks

SS · Encrypted Proxy Protocol

Shadowsocks, usually abbreviated to SS, is an encrypted proxy protocol that uses a pre-shared key. The client and server must support and use the same encryption method; sharing a protocol name does not make settings from different generations interchangeable. If the encryption method is blank after import, first check the link format and client support.

REALITY

Transport security option

REALITY is a transport security option provided by Xray, often paired with VLESS. It does not replace VLESS user authentication and cannot be swapped directly with a standard TLS configuration. Treat the protocol type and security option as separate fields when reading a configuration.

See the protocol guide →

Project Background / Configuration Compatibility

Cores & Ecosystem

The graphical client imports subscriptions, saves settings and manages processes; the core handles the actual connections. A client recognizing a link does not mean the selected core can run the configuration generated from it.

Project V

Project background

Project V provides context for understanding V2Ray and its derivatives; it is not a connection protocol. It separates inbound, outbound and routing settings into distinct configuration sections, a model later projects also adopted. When a source cites Project V, distinguish early design documentation from syntax supported by the current core.

V2Fly

Community project

V2Fly continues V2Ray development, and its core handles proxy connections, routing and DNS. It shares a history with Xray, so configuration terms often overlap, but that does not mean the two are fully compatible. For unknown fields, check the core and version currently in use.

Xray

Proxy core

Xray originated in the V2Ray codebase and has since evolved as an independent core, adding features such as REALITY. The graphical client and Xray have separate version numbers because they are different components. When startup fails, configuration parsing errors in the core log are usually more useful than the connection status shown in the interface.

Traffic Entry / Connectivity Tests

Client Features

Proxy modes answer how app traffic enters the client; speed tests answer whether a particular test path works. They provide different clues and are best checked separately.

System proxy

Application-level traffic entry

The system proxy sets a proxy address through the operating system, primarily affecting apps that read and follow that setting. Programs with their own network settings may still connect directly or use a separate proxy. If one app works and another does not, first check whether both use the same settings.

TUN mode

Virtual network interface

TUN mode routes traffic through a virtual network interface, then hands it to the client for processing. Enabling TUN does not mean all traffic goes through the proxy; the core can still choose a direct connection or another outbound based on its rules. Interface permissions and the settings of other network software can also affect which traffic is captured.

Real connection latency

Proxy path test

Real connection latency is the time taken by a test request sent through a node; the exact measurement depends on the client. It is closer to everyday connectivity than pinging the server alone, but does not indicate sustained download throughput. Compare nodes using the same test URL so differences in the destination site's response are not mistaken for node performance.

See the client setup guide →

Match Conditions / Outbound Selection / Name Resolution

Routing & DNS

App connections and domain lookups may take different paths. When checking traffic routing, verify the information available to the core, the rule that matched, and the resolver handling the lookup.

Routing rules

Connection matching & outbound selection

Routing rules send connections to a specified outbound based on conditions such as the destination domain or IP address. Common cores evaluate rules in order, so an earlier match can prevent a more specific rule later in the list from running. Check the active rule order when troubleshooting, not just whether a rule appears in the list.

Traffic routing

Route traffic by condition

Traffic routing selects a path based on connection attributes—for example, sending some connections direct and others through a proxy outbound. System proxy and TUN are different ways traffic enters the client; routing happens after the connection enters the relevant processing flow. If traffic is not taking the expected path, first confirm that it reached the client.

GeoIP

IP address range collection

GeoIP groups IP address ranges by region or other labels for bulk matching against destination addresses. It describes how addresses are categorized in the dataset, not necessarily where a website is based. If different resolvers return different addresses for the same domain, GeoIP matches may also differ.

GeoSite

Domain rule collection

GeoSite groups domain rules by service, category or other labels, avoiding the need to maintain each domain individually. These labels identify rule groups in the dataset; they do not indicate where a server is located. If the core only has the destination IP and cannot recover the domain, domain rules cannot match directly.

FakeDNS

Synthetic addresses & domain mapping

FakeDNS returns a synthetic address for a query and records the mapping between that address and the domain. When the app connects to that address, the core handling the connection can recover the domain for routing decisions. Receiving a synthetic address alone does not prove the mapping is working; the later connection must also pass through a processing path that can use it.

DNS leak

Unexpected DNS resolution path

A DNS leak occurs when domain lookups take an unexpected path, potentially exposing queries to resolvers that were not meant to receive them. Even if a website connection uses a proxy, the app or operating system may resolve the domain through a different path. First identify which resolver is expected to handle the lookup, then compare it with the actual query path.

Configuration Source / Update Scope / Connection Entries

Subscriptions & Nodes

Updating a list involves three steps: reaching the subscription URL, parsing its response format and displaying its entries in the group. Once entries appear, check separately whether each node can connect.

Subscription

Bulk configuration retrieval

A subscription lets the client retrieve and update a group of node configurations from one URL. A reachable URL does not guarantee that its response uses a format supported by the client, and a successful update does not mean every node is online. Subscription URLs may contain access credentials; redact the full URL before sharing screenshots or logs.

Node

Individual connection profile

A node is a selectable connection profile in the client, usually containing a server address, port, protocol and authentication settings. One host can provide multiple profiles, so the number of nodes is not the number of servers. Node names are usually labels; changing a name does not change the connection settings.

Subscription group

Node source & update scope

Subscription groups organize nodes by source or purpose and help show which entries an update applies to. A group label is not automatically a routing condition, and selecting a group does not switch the active connection. If entries seem to disappear after an update, clear any list filters first, then check the selected group and update log.

See subscription and connection FAQs →