Shadowrocket Node Timeout and Connection Troubleshooting: Latency Tests, Protocol Settings, and System Time

Troubleshoot node timeouts and connection failures in order: check latency test results, verify protocol settings and encryption, then check your device’s system time.

At a glance

A Shadowrocket node timeout does not necessarily mean your subscription has expired, and turning on the connection does not guarantee that proxy requests are succeeding. This guide is for users who already have a service provider and subscription. First, distinguish a failed test from a failed request, then check protocol fields, device time, and routing settings. The troubleshooting approach applies to iPhone, iPad, Mac, Windows, Android, and Linux. The steps use Home, Config, and Settings on iPhone / iPad as a reference; on other platforms, check the corresponding features, as button locations may differ.

1. Establish the scope of the issue before changing settings

Before you begin, keep your current configuration and note the selected node, network, current Global Routing mode, and when the problem occurred. Change only one thing per test. After each change, disconnect and reconnect, then visit the same page. This makes it possible to identify which change affected the result instead of mixing network changes, rule edits, and subscription updates.

A “timeout” usually means an operation did not complete within the allotted time. By itself, it does not tell you whether the server is down, the protocol is misconfigured, or DNS failed. First identify where the failure occurs: the latency test, proxy handshake, or web request. The sequence below is a troubleshooting model; not every protocol uses exactly the same connection steps.

Network reachability Address resolution Proxy handshake Rule-based routing Target response
1 setting
Change one condition per test and keep a record of results before and after
2 networks
When possible, compare Wi-Fi and mobile data
4 steps
Global Routing:Proxy / Direct / Config / Scene

Goal: establish repeatable test conditions

Use the same node and target page for each test. Start with one test on your current network. If the result changes only when you switch networks, investigate the network path first instead of rebuilding the entire configuration.

2. Check the latency test, but don’t treat a low reading as proof of a working connection

In Home, run a latency test for the current node and first check which test method is being used. A TCP connection test mainly measures whether a connection to the server port can be established. An HTTP-style test through the proxy also involves the proxy handshake, test destination, and response. Results from different methods are not directly comparable, and one low reading does not prove that every website will load.

  1. Keep the test node fixed

    Go to Home, select the node you’re troubleshooting, and note the test method and result. Repeat the test several times to distinguish consistent timeouts from intermittent failures. Don’t switch nodes before each test.

  2. Check network access

    Make sure you’ve completed any required sign-in on the current Wi-Fi network, or that mobile data is working. Check a known directly accessible page without relying on the proxy to rule out a network connection issue.

  3. Compare networks

    When possible, test the same node on both Wi-Fi and mobile data. If it fails on only one network, note which network and test method were used, then check whether that network allows TCP and UDP traffic to reach the destination.

  4. Test a real request

    Return to Home, connect to the current node, visit the same target page, and check the request log. Troubleshoot “test succeeds but request fails” separately from “both test and request fail.”

A test can fail if its destination is temporarily unavailable. In that case, use an actual request to help assess the connection. Conversely, if the TCP port is reachable but webpages won’t load, check authentication, transport settings, TLS, or routing. Looking for an even lower latency reading is unlikely to help.

3. Verify protocol settings against your provider’s original details

Open the node’s edit screen in Home. Confirm the protocol type, then check the server address, port, and authentication details one by one. Use the port provided by your service provider. For example, if the server is configured for 443, enter 443 in the client; don’t change it to 80 just because the connection fails. The port number alone does not tell you whether TLS is enabled.

Protocol Fields to check first Source of truth and common pitfalls
Shadowsocks Server, port, Password, encryption method The encryption method and Password must match the server. Don’t substitute an algorithm just because its name looks similar.
VMess / VLESS UUID, transport, TLS, Host, Path Enter only transport settings enabled on the server. A WebSocket Path is not a free-form note that can be changed at will.
Trojan Password, server port, TLS Server Name The authentication password and certificate name are different fields. The connection address and certificate name may also differ.
Hysteria2 Authentication details, TLS name, UDP reachability Uses QUIC and depends on UDP. A successful TCP test does not prove this protocol path is available.
WireGuard Keys, Endpoint, interface address, AllowedIPs Uses UDP. The local private key and peer public key are not interchangeable; don’t share your private key when asking for help.

When copying fields, check for leading or trailing spaces, line breaks, and missing characters. The TLS Server Name is used to verify the server’s identity and may also affect how the server routes the connection; it is not always the same as the connection address. Enter these parameters only when your provider specifies them. Don’t try arbitrary reachable domains in Host or Server Name.

Handle subscription updates separately from local edits

Get your subscription link from your service provider. If the provider has changed the settings, update the relevant Subscribe entry and check the node fields again. Save a record before editing a node generated by a subscription, since a later update may overwrite your changes. Updating a subscription cannot bypass UDP restrictions on your current network or correct your device’s clock.

4. Check system time last, and keep certificate verification enabled

If you’ve verified the address, port, and authentication fields but still see TLS handshake or certificate validity errors, check the device’s date and time. A clock that is significantly ahead or behind can make a valid certificate appear not yet valid or expired. It can also affect some authentication methods that check timestamps. This does not mean every timeout is caused by an incorrect clock.

date -Is
timedatectl status

timedatectl is available on systemd systems that include the tool, but not on every distribution. Check the output for system clock synchronization status. Enabling a time synchronization service does not mean synchronization has succeeded. A different displayed time zone does not necessarily cause TLS failures; confirm that the actual time is correct.

After correcting the time, reconnect and revisit the same page. If the time is correct but the certificate is still reported as expired, the remote certificate may genuinely need renewal; contact your service provider. Don’t hide the issue by disabling certificate verification, and don’t install certificates from unknown sources.

5. Latency looks normal but pages won’t load? Check Global Routing and DNS

In Home → Global Routing, check the current mode. Proxy routes traffic handled by that mode through the proxy; Direct lets you test a direct connection; Config routes traffic according to the current configuration rules; and Scene applies the configuration for the selected scene conditions. Don’t rely on the VPN icon alone when troubleshooting—check which path the failed request actually took.

Temporarily switch to Proxy, reconnect, and visit the same page, then restore the original setting. If it works in Proxy but fails in Config, check the active configuration, matching rules, and selected policy first. This is a troubleshooting clue, not a reason to route all traffic through Proxy permanently. For local network addresses or special exclusions, also check the actual request log.

Don’t use “entering the target IP directly in a browser” as the only way to decide whether HTTPS is working. HTTPS usually depends on the correct domain, certificate name, and virtual host. Visiting an IP directly can trigger a certificate error and introduce another variable.

6. Use the exact error message to choose the next step—don’t keep reinstalling

The following snippets may appear in network logs, sometimes alongside a target address or additional context. Full error wording can vary by system. Follow the corresponding troubleshooting path only if the actual log contains that snippet; a page spinning indefinitely does not identify a specific error.

Error: i/o timeout

Cause and next step: A network read or write did not finish in time. First check whether the log points to the server connection, proxy handshake, or target request. Then test the same node on another network to narrow down the cause.

Error: connection refused

Cause and next step: The connection was explicitly refused, possibly because nothing is listening on the port or a network device rejected it. Verify the address and port provided by your service provider, and ask them to check that the service is listening.

Error: no such host

Cause and next step: The current resolution process could not find an address for the host. Check the spelling of the domain in the error, then determine whether it is the proxy server’s domain or the target website’s domain and check the corresponding DNS resolution path.

Error: x509: certificate has expired or is not yet valid

Cause and next step: The certificate validity period does not match the time of verification. First check automatic time synchronization. If the time is correct and the error persists, contact your service provider to check the remote certificate.

If only one node keeps failing on the same network while your other existing nodes work, focus on that node’s settings or remote service, but don’t assume this alone identifies the cause. If all nodes fail, check shared conditions first—such as network sign-in, device time, VPN permissions, or configuration—instead of changing passwords at random.

7. Share reproducible details and restore temporary changes

If the issue persists after these steps, organize the results for your service provider. A useful report answers “which network, which protocol, where did it fail, and what changed the result?”—not just a screenshot of the connection toggle. For device settings, see the Settings Reference.

Reinstalling the app may remove local settings, but it won’t fix a remote port, certificate, or network path issue, so it shouldn’t be your first step. Start with the main sequence—latency test, protocol settings, then system time—then use actual request logs to check routing and DNS. This usually narrows the issue down to a specific part of the connection.

Download Shadowrocket