Shadowrocket Official Manual · Troubleshooting by symptom
Troubleshooting Shadowrocket
First determine whether the issue is with device connectivity, establishing a connection, or access after connecting. Then go to the relevant section. Change one thing at a time and retest the same destination.
Setup guide walks you through importing, selecting Global Routing, connecting, and verifying the setup. This page helps diagnose problems you’ve already encountered. If you haven’t connected successfully yet, complete the setup steps first, then return to the step that failed.
The instructions below mainly use iPhone and iPad as examples. On both devices, Shadowrocket is a one-time $2.99 purchase from the App Store. Check the App Store listing for system requirements. The Mac (Apple silicon and Intel), Windows, Android, and Linux versions are free; get their installers from this site via the Get the app page at /dl/shadowrocket/. A one-time app purchase is not a service plan. This manual uses only subscription or server details you already have; it does not provide, sell, or recommend them.
Connection toggle won’t turn on or keeps switching off
First distinguish an authorization issue from a connection failure
Open Home and confirm that SERVER contains an entry from your existing server details and that the entry is selected. Tap the connection toggle once and wait for the status to settle. No visible change, a system authorization prompt, and a brief attempt followed by Not Connected are three different symptoms. If nothing changes, check that the interface is responding and the entry is selected. If authorization is requested, complete it. If the connection drops back, note the message from the app or system, then check the configuration and network. Repeated taps can cancel a connection that is still being established and make it harder to tell which action caused the final status.
When you start a connection for the first time, the system may ask to add a VPN configuration and then prompt you to authenticate as directed on your device. Confirm only when you’ve just started the connection from Shadowrocket and the prompt matches that action. If you canceled earlier, return to Home and try connecting again. If no prompt appears, search for “VPN” in system settings and check whether the configuration is already there. An existing configuration doesn’t mean a connection is active; don’t delete it just to make the authorization prompt appear again. See First launch and VPN configuration permissions for where to find these options.
Check system restrictions and automatic connection conditions
On a school- or work-managed device, VPN settings may be controlled by an administrator, or a network feature may be restricted to specific networks. Check whether the system says the device is managed, then ask your administrator. Don’t remove management profiles, reset managed settings, or try to replace an organization-required VPN. On a personal device, if another VPN you manage is connected, disconnect it temporarily if appropriate. Check that it doesn’t reconnect automatically, then try Shadowrocket. Restore any settings you need afterward.
If the connection comes back after you disconnect manually, or drops just after you switch Wi-Fi networks, note the network name and the sequence of actions, then check On Demand. Record its original enabled state and conditions first. On a personal device, you can temporarily test with manual connections only, keeping the network unchanged. If a manual connection stays stable and the issue returns when On Demand is restored, check whether its conditions match the current network. If manual connections are also unstable, investigate the server and underlying network rather than repeatedly changing network conditions.
Retest with the fewest possible changes
Disconnect Shadowrocket and open a page that normally works without a proxy to confirm that Wi-Fi is actually connected, not just joined. If the network requires a sign-in page, complete its usual authentication, then return to Home. Stay on the same network and test with server details that have worked before. If the toggle drops only with a newly imported entry, keep the old one and don’t replace the whole configuration. Go to Server timeout and check the fields. If no entry connects, restart the device and test once more, noting whether the system message changed.
Consider removing a VPN configuration and letting the app request a new one only if you’ve confirmed the issue is limited to a VPN configuration you manage and have saved what you need to restore it. After authorizing again, check two things separately: whether the toggle stays on and whether webpages open. If the first works but the second doesn’t, go to No internet access after connecting. Repeating authorization won’t usually provide new information. If the connection still fails, note whether it happened before or after authorization and keep a redacted copy of the message and your retest notes.
Connected, but webpages or apps won’t load
Compare two destinations
Choose one page that normally opens on the current network without a proxy and one page that’s actually failing. Note their full hostnames, error messages, and when the issue occurs. Disconnect and test the first page, then connect and test both pages again. Each time, load a page that makes a fresh request; don’t rely on content already displayed in the browser. If even the normally accessible page fails while disconnected, troubleshoot the network first—for example, the device may not have completed Wi-Fi sign-in, the router may be offline, or the app may not be allowed to use cellular data. Changing rules won’t fix an underlying connection problem.
Record which step failed instead of simply noting “no internet.” A hostname lookup failure, connection timeout, certificate warning, and server error point to different layers. Receiving an error page means a response arrived, but doesn’t prove that sign-in, resource loading, or later requests are working. Cached content also doesn’t confirm current connectivity. If you see a certificate warning, don’t enter account details. Check the domain, system time, and whether the network requires authentication, then retest on a network you know works.
Compare Global Routing under the same conditions
| Mode | What it helps you compare | What it doesn’t prove |
|---|---|---|
| Config | Handles requests according to the rules and policies in the currently enabled configuration. | Editing a file doesn’t mean it’s enabled. |
| Proxy | Check whether the same destination works through the proxy route. | One working destination doesn’t mean other sites or apps are working. |
| Direct | Compare a route that bypasses the proxy while the client is connected. | This isn’t the same as disconnecting the client; its network settings may still affect the device. |
Note the original mode before testing. If Direct opens normally accessible sites but Proxy fails for several destinations that worked before, check the selected server and connection route. If Proxy and Direct work but the issue occurs in Config, check the active configuration and the rule matched by that hostname. After changing modes, make a new request; existing connections may keep using the previous route. Don’t leave the client in another mode just because the comparison worked—later requests may no longer follow your intended routing.
Check rule order and the hostname that actually fails
Keep the configuration currently in use under Config. The rules below are a standalone example to illustrate ordering, not a template for replacing your entire configuration. It handles an example domain first, then matches applicable requests against IPs in a specified region, and sends unmatched requests to the proxy policy. If your configuration uses policy groups, check their names and current selections. Don’t blindly replace names in your own copy with the example values.
[Rule]
DOMAIN-SUFFIX,example.com,PROXY
GEOIP,CN,DIRECT
FINAL,PROXY
Before changing a rule, check that a broader rule above it isn’t already matching. A site may use multiple hosts: the main page might load while images, sign-in, or submissions fail, and the host handling those requests may differ from the one in the address bar. Use the known failing hostname for a narrowly scoped comparison rather than routing a whole group of related domains the same way. If the change doesn’t help, restore the original rule and check DNS or the service itself. For routing only selected sites through a proxy, see Routing selected sites with FINAL,DIRECT.
Keep one unchanged destination as a control during retesting. While fixing the failing page, confirm that a destination that should use a direct route still works as expected. If both outcomes can’t be achieved, the change may be too broad. Record the results in three columns: destination, mode, and success or failure. If only Config fails, start with the rules. If the destination can’t be resolved in any mode, go to the DNS section.
Connectivity Test times out or only some servers fail
Separate test results from real-world access
Connectivity Test checks whether the test route can complete its check at that moment. A timeout means no successful response arrived during the wait; by itself, it doesn’t prove the server is offline, the credentials are wrong, or the device is at fault. Note which server you tested, the network, and the time. Wait a few seconds, test once more, then use the same page you normally access to check a real request. If the test fails but access works, check whether the test destination alone is unreachable. If the test succeeds but a webpage fails, investigate the hostname, DNS, or rules. Test duration also can’t be directly converted into file download speed.
If only one entry fails, test another entry you already have and have used successfully, keeping the network and mode the same. If the other entry works, check the failing entry rather than changing DNS everywhere. If several entries fail, first confirm that a normal webpage opens when disconnected, then compare the same entry on two networks you know work. On iPhone, compare Wi-Fi and cellular data if your plan and data allowance permit. A Wi-Fi-only iPad can be tested on another network you’re allowed to use. Don’t change networks and protocols at the same time, or you won’t know which change mattered.
Check the protocol and its related fields
| Field | What should match your existing details |
|---|---|
| Address and port | Keep paths and ports out of the address field; use the port assigned to that service endpoint. |
| Shadowsocks | The encryption method and password must be the matching values from the same set of details. |
| VMess / VLESS | Check the protocol identifier, transport method, and required server parameters for the selected protocol. |
| Trojan | Check the password and related settings, including the server name used for TLS. |
A display name helps identify an entry, but it isn’t a connection address. When entering details manually, check for leading or trailing spaces, full-width punctuation, and accidentally using a service webpage address as the server address. Match the TLS server name and transport parameters to your existing details. Don’t replace a hostname with an IP just because it resolves, or disable certificate verification to dismiss an error. For certificate or time-related warnings, check the system time and the hostname in your details. If the settings are correct but the connection still fails, ask the service administrator to verify them; don’t dismiss a verification failure as ordinary network delay.
Choose the next step based on how widely the issue occurs
If one server entry fails on two networks while an older entry works on both, keep the working entry and check whether the failing details have changed. In particular, confirm the address, port, and transport method all come from the same set of updated details. Repeated imports can leave entries with similar names; return to Home and confirm the new entry is selected before testing. If the same details work on another device you manage, note the differences in network and settings rather than assuming the current device’s permissions are fine.
If all entries fail on just one network, check its usage requirements and whether sign-in is required. Contact the network administrator if needed. If all entries fail on two networks you know work, check whether a subscription, configuration, or system change happened recently. When reporting the issue to the service administrator, include the failure stage, approximate time, whether it reproduces on both networks, and whether other entries in the same group work. Don’t share the full subscription URL, passwords, identifier fields, or QR codes containing them.
Subscription won’t import or update, or is empty after updating
First identify the type of information you have
A single server entry, a Subscribe URL containing multiple entries, and a configuration file used under Config are different kinds of data. A QR code opened with Scan QR Code may also contain different kinds of information; recognizing its contents doesn’t mean they were added to the right list. Confirm what the original details are for, then choose the right option in the setup guide. Don’t paste a single server’s connection details into the subscription URL field or import a rules file as a set of server entries.
Copy the full URL from its original source. Check whether automatic line wrapping became an actual line break, whether punctuation at the end was copied, and whether anything after the question mark was cut off. Don’t remove query parameters to make the URL easier to inspect; they may determine what the service returns. Don’t paste the URL into a public conversion or diagnostic website, or display it in full while sharing your screen. If the details have been exposed, use the original service’s management tools to get new ones and replace the old details.
Separate request failures from content parsing failures
Note which step the app reaches before showing the message. If the request can’t be sent or times out, check basic connectivity, whether the subscription hostname resolves, and whether the current route is direct or uses an existing proxy. Retry once on a network you know works, rather than repeatedly fetching from a public network that requires sign-in. An incorrect system time can also affect HTTPS requests. If there’s a certificate warning, check the time and hostname; don’t bypass it. If the network route works but timeouts continue, check whether the existing service has announced maintenance and keep a note of the current conditions while waiting for confirmation.
If the app reports incompatible content or completes an update without finding any entries, don’t check only the network. The server may have returned a sign-in page, a generic error page, or data that doesn’t match the entry point you’re using. A successful response doesn’t necessarily contain a valid subscription. Use the original management portal to check that the subscription is active, the share URL hasn’t been regenerated, and the format provided is supported. If access is explicitly denied, check whether your existing authorization is still valid instead of retrying the same URL repeatedly.
Keep existing entries and verify the new content
If you have working entries, first save the necessary details using a recovery method provided by the app and keep the backup somewhere you control. If the list is empty after an update, keep the message and note when it happened; don’t immediately clear every list or delete the configuration. Return to Home and check subscription groups separately from manually added entries. Confirm whether that group returned empty or whether the selected list or view changed. If the new URL works, import it once where you can distinguish it from the old group. Then check the old and new entries’ names, protocols, and selection states so you don’t mistake an old entry with the same name for the update result.
When verifying the result, first check that a reasonable set of entries appears, then select one and test the connection and a webpage. Successful import, successful connection, and successful access are three separate checkpoints; none guarantees the others. If only some entries fail to connect, go to Server timeout instead of repeatedly updating the whole subscription. If all entries fail, compare what actually changed from before the update: the URL, entry details, or current selection.
Connected, but pages or downloads are slow
First determine whether the delay is before or during transfer
Waiting a long time for the first content to appear is different from a transfer that starts but remains slow. The first may be due to DNS lookup, connection setup, or a slow response from the site. The second can also depend on route quality, server speed, Wi-Fi conditions, and background traffic on the device. Choose a page or small file that normally works and can be tested repeatedly. Check whether each attempt stalls at roughly the same point. If only a particular view is slow after sign-in, also consider the view itself, external resources it loads, and the account service.
For each comparison, record the network, selected server, Global Routing mode, and time. Compare direct access with the current route for the same permitted destination. If the destination doesn’t work directly, don’t treat that failure as a valid speed comparison with the proxy route. Make a new request after changing conditions so you don’t rely on cached content. Run only a few tests per set and check whether the results are consistent. Avoid repeated large-file speed tests on cellular data. If two attempts differ greatly, record the variation rather than reporting only the fastest or slowest result.
Use two rounds of comparisons to narrow down the network and server
In the first round, keep the server unchanged and check whether performance differs when you’re near versus far from the access point, or compare with another network you know works. If Wi-Fi is slow only in one location, compare direct access with other devices on the same network to tell a device issue from network congestion. If it’s just as slow with Shadowrocket disconnected, troubleshoot that underlying condition first. A strong Wi-Fi icon only says something about the device-to-access-point link; it doesn’t show whether the route from the access point to the remote service is congested.
In the second round, keep the network and destination unchanged and compare only servers you already have and can use. If the slowdown is limited to one entry, note when it happens and which destination is affected; don’t rush to change protocol settings. If it occurs only at certain times, retest at another time and share your notes with the administrator of the existing service. If different servers are slow only for the same site while other sites and apps work, the destination itself may be responding slowly. Avoid changing servers, DNS, and rules together; even if performance improves, you won’t know which change to keep.
Use Data as a guide, not to infer latency
In Data, compare upload and download readings before and after testing. First check which statistics view is selected and whether the values include traffic accumulated from earlier use. If the device is syncing photos, updating apps, or streaming video, the increase may not come from the current test. Use a short time window with just one activity, then compare the increase with what you experienced. These figures show the data volume recorded by the app, not all traffic on the device, and don’t show how long a request waited before the connection was established. See How to read traffic statistics on the Data screen.
If each first page load is slow, check DNS and connection setup next. If transfers are slow throughout, keep the working route and test other networks and destinations separately. If the issue occurs only in Config, check whether the destination and its key resources use the intended policy. Don’t route all traffic through the proxy just to speed up one page. When done, restore the original mode and test both the slow destination and one that was already working to confirm that normal routing is unaffected.
DNS lookup failures and inconsistent results
Tell lookup failures apart from connections to resolved addresses that still fail
DNS looks up the records needed for a domain. Once an address is found, the client still has to establish a connection and may need to validate a certificate before it can receive a webpage response. If the browser explicitly says it can’t find the server, first check for a typo in the domain, then test another domain that normally works. If only one host fails, possible causes include that host’s records, how the current network resolves it, or the app making a request to a different host. Several unrelated domains failing at once points more toward a general DNS route issue, but first confirm that the device has basic connectivity.
Being able to access an IP address doesn’t prove that the website should work. Entering an IP directly doesn’t reproduce the certificate and site selection associated with the domain; it may show a different site or an error. Apps may not all use the same system resolver either. If only one browser or app has the issue, note whether it has its own DNS settings. When comparing the same host in two apps, make sure each sends a new request rather than using its own cache.
Restore a known baseline, then check custom DNS
Check whether the device is using system DNS, custom DNS, or DNS over HTTPS. Save the current settings, then change only one thing for a short comparison. If the problem started after adding a custom server, temporarily restore the previously working DNS method and retry on the same network with the same server and mode. If system DNS works but the custom method doesn’t, check whether the custom endpoint is reachable and its address is correct. Don’t add several fallback settings at once. See DNS settings compared for more on choosing an option.
DNS over HTTPS still depends on other conditions to establish the initial connection. If its endpoint is configured as a hostname, the client needs a working route to resolve that hostname and establish an HTTPS connection. Check that you haven’t entered the full endpoint as a plain server address or missed a path required by your existing configuration. If it fails only on one Wi-Fi network, first rule out sign-in requirements and access restrictions. For a certificate warning, check the time and endpoint name; disabling verification isn’t a fix.
Check rule-matching clues in Config
Domain rules and IP-based rules use different matching criteria. DOMAIN-SUFFIX matches a domain suffix; IP-CIDR matches an IPv4 range, and IP-CIDR6 matches an IPv6 range; GEOIP matches based on an IP’s geographic classification. A change in DNS results can change the outcome of IP-based rules; that doesn’t mean the domain rule itself has stopped working. If an issue occurs only in Config and changes when you switch DNS methods, restore the original settings, then test a minimal rule for that host in a copy of the configuration. Avoid expanding the change to an entire group of sites.
After making a change, send a new request. If needed, disconnect and reconnect the client on a device you manage, then reopen the affected app. Caches can exist at multiple layers; don’t immediately reset network settings because one result didn’t change. Test at least one domain that failed and one that worked. For internal services, also check the DNS method required by your organization or home network. A temporary fix for one domain doesn’t mean the new DNS settings will work on every network. If there’s no improvement, restore the known-working settings and investigate the connection or site based on the actual error.
Increased battery drain or repeated reconnects while the screen is locked
First compare battery level and activity under similar conditions
Check the system’s battery screen for app activity and battery changes during the affected period. The percentage is relative to the selected reporting window; it doesn’t show how much extra charge an app used just because its share is high. Note screen-on time, foreground tasks, cellular signal, and network changes, and compare windows of similar length. After charging, a system update, or media syncing, the device may have extra activity, so don’t compare it directly with an idle period from the previous day. System activity is a useful clue, but a single number can’t identify the cause.
Choose two short, similar observation windows. Keep the network, server, and screen use as consistent as possible. On a personal device, compare one period while connected with another while disconnected if your work allows it. Follow administrator requirements on managed devices. If battery use is similar while disconnected, check for syncing, video calls, hotspot use, or system tasks during that period. If it happens only while connected, check whether it occurs during stable idle time or around events such as leaving Wi-Fi or waking the screen.
Check On Demand and weak-signal network changes
If the issue tends to occur after you leave Wi-Fi coverage, keep the device on one stable network and observe it first. Before locking the screen, note the connection status in Home. After unlocking, check whether it’s still connected and whether webpages open. If idle time on a stable network is fine but the issue returns after repeatedly moving in and out of coverage, note roughly when each status change happens. A brief reconnect may be normal during a network switch. If the cycle continues without recovering, check the trigger conditions, server availability, and whether the device has network access after switching.
Before checking On Demand, note its current conditions. On a personal device, temporarily compare with a manual connection without changing the server or DNS at the same time. If manual connections stay stable but the issue returns when automatic triggers are restored, check whether a condition keeps matching on networks you regularly use. If manual connections also drop repeatedly, return to Connection toggle and Server timeout to find why the connection is unstable. Don’t keep a speed-test page open to monitor idle behavior; it keeps the network active and makes the comparison unreliable.
Check background traffic, then restore your usual settings
Use Data to compare changes between the two observation windows, and check system activity for tasks that may have needed a continuous connection. If traffic has increased noticeably, pause a task you started only if it’s safe to do so, then retest. Conversely, low traffic doesn’t rule out repeated connection attempts or unstable signal. If changing one trigger condition stabilizes the connection and battery use returns to its usual range, keep that adjustment and restore other test settings one at a time. Avoid leaving unrelated changes in place.
If the device gets unusually hot, shuts down unexpectedly, or continues to lose charge rapidly after disconnecting, stop extended testing. Check battery health and active tasks in system settings, and contact device support if needed. When reporting an app-related issue, include the observation window, charging status, network, whether On Demand was enabled, and whether the issue also occurred while disconnected. Don’t submit a full-page screenshot that exposes other apps’ private information. A reproducible sequence of steps is usually more useful than a single battery graph.
Connection issues after an update or settings that seem to have changed
First identify what was updated
An app update, system update, subscription refresh, and configuration replacement may all happen around the same time, but they affect different things. Note when each change actually occurred and when the connection last worked. If ordinary webpages stopped opening just after a system update, check system networking first. If only some entries fail after a subscription refresh, check those entries. If a site fails after switching Config, check the selected configuration and its rules. Choose the next step based on what changed, rather than blaming every issue on the most recent update.
If the app opens, first check in Home that the right server is selected and Global Routing is set as before, then confirm the active item in Config. An entry still having the same name after an update doesn’t mean its contents are unchanged. Likewise, opening a configuration file doesn’t mean it’s enabled. Test with a destination you know worked before and a server entry that’s still valid, and note the results. If interface labels changed after an update, see English interface terms to avoid mistaking a changed layout for a missing feature.
Restore from your own saved details; don’t overwrite blindly
Before comparing configurations, back up the current state and open a previously saved copy. Check rule order, DNS, policy names, and custom settings. Don’t overwrite a file with a full remote configuration if it contains local changes you still need. If only one setting differs from the known-good state, restore just that setting in a copy and compare the original failing destination with one that works. If Config fails but other modes work, focus on Config. If all three modes and the disconnected state fail, restoring the configuration probably isn’t the right first step.
If the app closes immediately on launch, stop trying to open it, restart the device, then try once more on a network you know works. On iPhone or iPad, check the App Store product page and system requirements. Installers and system prompts for Mac, Windows, Android, and Linux are covered in the installation instructions on the Get the app page. For restrictions on a managed device, check with the administrator first. Before reinstalling, find out where the data you need to restore is stored. An eligible App Store purchase can let you download the app again, but it doesn’t save your subscriptions, server details, or local rules.
Write down the steps to reproduce the issue
Start by checking basic connectivity while disconnected, then record whether the app opens, an entry can be selected, a connection can be established, and the original destination can be reached. Stop at the first failed step; repeatedly checking later rules won’t help if an earlier condition isn’t met. If the app and system were both updated on the same day, note which came first and whether the issue started before the subscription was refreshed. If you can test on another device, say whether it used the same network, configuration, and server details.
When the issue is resolved, note the specific change that worked, such as restoring the original configuration or selecting the original server again. Then test another route you use regularly. Until the cause is clear, don’t make a temporary comparison setting your new default. When reporting the issue, include system messages, reproducible steps, affected features, and conditions you’ve ruled out; don’t submit a full backup. Check version information as shown on the device or download page at the time rather than filling in a version from memory.
iPad-specific checks: device changes, Split View, and network switching
The same purchase record doesn’t mean connection details are transferred too
If Shadowrocket works on iPhone but not iPad, troubleshoot the iPad separately first. App Store purchase history covers getting the app; it doesn’t transfer another device’s servers, rules, or system VPN permissions. On the iPad, open Home and check whether SERVER contains your details, the same entry is selected, and Global Routing matches. For the first connection, follow the system authorization prompts on that device. If the app opens but the list is empty, return to the import steps. Don’t delete the original entry from your iPhone or assume all settings have synced.
When downloading the app again, use the genuine app verification guide to compare the App Store listing’s developer, Shadow Launch Technology Limited, app ID 932747118, and product icon. Check the store listing for system requirements. To restore your details, use a configuration or subscription you saved yourself and can verify. If the iPad is shared between home, school, or work, don’t put backups containing personal credentials in a shared folder. After restoring, check which configuration is active rather than just confirming that a file appears in Config.
Compare both devices on the same network first
A Wi-Fi-only iPad can’t switch directly to cellular data like a device with a cellular connection. Even if both devices show a Wi-Fi icon, check that they’re on the same network, have both completed any sign-in, and aren’t subject to device-specific access restrictions. Disconnect the client and open the same basic website on both devices. Then retest using the same server and mode. If the iPhone’s successful test relied on cellular data, put both devices on the same network before deciding the issue is specific to the iPad.
If possible, retest on another network you’re allowed to use. Before using a personal hotspot, check data limits and permissions, and record “switched to hotspot” as a separate change. If the connection works again after rejoining a network, investigate the network switch; that alone doesn’t prove the iPad hardware or app is at fault. For a school- or work-managed iPad, check whether network policies allow an additional VPN. If the organization manages VPN or DNS settings, contact the administrator rather than trying to remove them.
Isolate Split View, screen lock, and app-specific issues
If a webpage keeps loading in Split View, first confirm that Home in Shadowrocket shows an active connection. Then open the affected app by itself and retest with a fresh page request. Split View isn’t a routing rule; when you return to an app, its page may resume an earlier request. If only that app fails while normal destinations open in the browser, note the specific feature and destination instead of deleting the VPN configuration. If the issue occurs only after locking and unlocking the screen, check whether Wi-Fi reconnected and whether the client needs a moment to recover. Don’t assume the connection dropped just because the app still shows an old screen.
When troubleshooting is complete, test two types of requests on the iPad: a page that should normally go direct and a destination you’ve confirmed should follow the current rules. Then try an action you regularly perform, such as locking and unlocking the screen or switching between networks you’re allowed to use, and confirm the result is consistent. Keep a note for future reference: whether the issue occurs only on iPad, only on one network, or only in one app, and which mode was active. Continue from the relevant section based on which condition best explains the change; there’s no need to redo the entire setup.