Is a Service Down or Is It You? The Complete Method to Find Out

Your favorite social network won’t load, your online bank shows an error, a messaging app stays stuck… The first question to ask is simple: is the service really down, or is the problem on your end (connection, device, cache)? Answering it in a few minutes saves you from wasting time needlessly restarting your gear — or, conversely, passively waiting on an outage that only exists on your device. This guide gives you a clear diagnostic method, valid for any site or app, so you know which side to look at and how to react.

Who this guide is for: any user facing a website, app or online service that has stopped working, who wants to quickly determine the source of the problem.

Prerequisites: your device (computer or smartphone), an internet connection (even partial) and five minutes. No technical skills required.


The diagnostic principle in one minute

A malfunction always has one of two origins: either the service itself is down (its servers, a global outage, maintenance), or the problem is local (your internet connection, your device, the app, a corrupted cache). The method is to rule out a general outage first — verifiable in a minute — then, if the service works for others, to go up the chain on your side: network, device, app.

This logic, from the broadest to the most specific, saves precious time and spares you needless fiddling.

Step 1: check whether the outage is general

Always start here. Several free sources show in real time whether a service is having a massive outage.

• Outage detection sites (like Downdetector) show a curve of reports. A sudden, high spike confirms a general outage; a flat curve suggests a problem on your end.
• The service’s official account on social media (X, Facebook) often posts a message during an incident.
• The official status page: many services have a « status » page (often status.servicename.com) showing the live state of their systems.
• A simple search for « [service name] down » surfaces ongoing discussions if many users are affected.

Tip: cross-check at least two sources. A single isolated report proves nothing; a simultaneous spike on an outage detector and messages from other users confirm a real outage.

Step 2: test from another device or another network

If the sources aren’t clear, do a quick cross-test. Open the service on another device (the phone if the problem is on the computer, or vice versa) and, ideally, on another network (switch from Wi-Fi to mobile data, or the reverse).

• If the service works on the other device or network, the outage isn’t general: the problem is your first device or your connection.
• If the service works nowhere, the outage is probably at the service, or further upstream at your provider.

Good to know: switching to mobile data (4G/5G) while Wi-Fi has trouble is the fastest test to isolate a home internet-box outage.

Step 3: tell a specific-site problem from an internet outage

Sometimes it’s not the service that’s down but your entire internet access. Check by opening two or three other well-known, reliable sites.

• If all sites are unreachable, the problem is your internet connection: see our dedicated guide [Wi-Fi connected but no internet](internal link to a Wanda-techs article on: « Wi-Fi connected but no internet: the complete checklist »).
• If only the service in question fails while others open, the problem is indeed tied to that specific service (a service outage or a local issue with that site).

Tip: a test like a « ping » or simply reaching a search engine immediately confirms whether the internet works overall on your end.

Step 4: if the outage is general, what to do?

If you’ve confirmed an outage on the service side, the best attitude is… patience. You can’t fix anything at an external provider, and fiddling with your gear will only add confusion.

• Don’t change your settings (password, network parameters) during an outage: you’d risk creating a real local problem that persists after recovery.
• Follow the status page or official account for progress and the estimated recovery time.
• Note the time if the incident affects you professionally; services often publish a report afterward.

Warning: beware of fake « tech support » that pops up during big outages. No serious service will ask for your credentials or a payment to « restore your access. »

Step 5: if the problem is on your end, the chain of checks

The service works for others but not for you? Go up the chain, from simplest to deepest.

1. Reload the page or close and reopen the app; the error is sometimes fleeting.

2. Check your connection: Wi-Fi active, mobile data available, Airplane Mode off.

3. Restart the device, then, if needed, your internet box.

4. Clear the cache of the browser or app (a corrupted cache often blocks a single service).

5. Update the app or browser: an outdated version can be rejected by the service.

6. Test in private browsing or with another browser: this rules out a faulty extension or local setting.

Good to know: in a browser, combining a forced refresh (ignoring the cache) with a private-browsing test resolves a large share of cases where a site works everywhere but on your end.

Step 6: the less obvious local causes

If the service stays unreachable for you alone after the basic checks, a few subtler causes deserve a look.

• A faulty DNS: your device translates the site’s address poorly. Trying a well-known public DNS can unblock access.
• An active VPN or proxy: it may route your traffic through a blocked or badly seen server; disable it to test.
• A browser extension (ad blocker, security) breaking the site: disable them one by one.
• The device’s date and time set wrong: this prevents security (certificates) from working and blocks many sites. Check they’re automatic.
• A suspended account or an expired session: log out then back in to the service.

Tip: a precise error message (a code, « not secure connection, » « 403, » « 503 ») is a valuable lead. Searching the exact message often points directly to the cause.

Step 7: build your own outage dashboard

To never be caught off guard again, keep a few reference points handy. For the services you use most, note the address of their official status page and their support account. Bookmark a general-purpose outage detector.

That way, at the next alert, you’ll know within thirty seconds whether it’s a general outage or a local problem, without guesswork.

Good to know: some services let you receive an email alert as soon as an incident is declared or resolved on their status page. A useful subscription for critical tools (bank, work email).

Understanding the most common error messages

The message shown is often the best clue to which side the problem is on. Here are the most frequent.

• « 503 Service Unavailable » / « 502 Bad Gateway » / « 504 Gateway Timeout »: these codes mean the service’s servers are overloaded, in maintenance or unreachable. The problem is almost always on the service side: wait it out.
• « 500 Internal Server Error »: an internal error at the service, also on its side. Nothing to fix on yours.
• « 404 Not Found »: the specific page doesn’t exist or was moved; the service itself works. Check the address.
• « 403 Forbidden »: access is denied — often a matter of rights, an expired session, or a geographic block (sometimes tied to an active VPN).
• « Not secure connection » / certificate error: frequently caused by a wrong date/time on your device, or a network intercepting traffic. It’s a local problem to fix.
• « DNS_PROBE » / « site can’t be reached »: your device can’t translate the address. A connection or DNS problem on your side (Step 6).

Tip: note the exact code and search for it. A 5xx code tells you to wait; a 4xx code or a DNS error sends you back to a check on your end.

Special case: a mobile app that no longer works

On a smartphone, an app can fail where the same service’s website works, and vice versa. That’s information in itself: if the site works in the browser but not the app, the problem is the app (cache, version, permissions), not the service.

In that case, the approach is targeted: force-close the app then reopen it, clear its cache in the settings, check it’s up to date on the store, and review its permissions (network, background). If nothing changes, uninstall then reinstall it: this starts from a clean version without touching your account.

Good to know: some apps refuse to work on a too-old version of the system or of the app itself. After a service-side outage, the developer sometimes ships a corrective update: check the store before concluding there’s a lasting problem.

Frequently asked questions

Is an outage detection site reliable? It reflects user reports: a clear spike is a good indicator of a real outage, but the absence of a spike doesn’t guarantee all is well. Always cross-check with another source.

The service works for my friends but not me: what should I do? The problem is local. Apply the Step 5 chain (reload, connection, cache, update, private browsing).

Should I restart my box for every problem? Only if several sites fail at the same time. For a single service in a general outage, it’s pointless.

How do I know if it’s my internet provider that has an issue? If all sites are unreachable and mobile data works, the origin is probably your fixed connection or your provider.

Can a general outage come from my antivirus or firewall? Yes, security software can block a specific service. Disable it temporarily to test, then re-enable it.

Staying calm during major global outages

Some outages are bigger than a single app. Large parts of the internet rely on a handful of shared infrastructures — cloud providers, DNS services, content delivery networks. When one of them stumbles, dozens of unrelated services fail at the same time: a social network, a bank, a game and a smart doorbell can all break together. Paradoxically, that simultaneity is reassuring: it’s a strong sign the problem isn’t yours.

In these moments, the right reflexes are simple. Don’t panic-change your settings or reset your passwords — you’d only add a local problem to a global one. Check a general outage detector, which will light up across many services at once. And wait: these large-scale incidents are usually fixed within a few hours, and there’s nothing an end user can do to speed them up.

Good to know: for truly critical needs (payments, urgent communication), it helps to keep an alternative ready in advance — a second messaging app, another payment method — so a big outage never leaves you completely stuck.

In summary

Facing a service that no longer responds, go from the broadest to the most specific: check for a general outage first via an outage detector, the status page and social media (Step 1), test on another device and network (Step 2), then tell a site issue from an internet outage (Step 3). If the outage is service-side, wait and follow the status page (Step 4). If it’s on your end, go up the chain: reload, connection, restart, cache, update, private browsing (Step 5), then the subtler local causes (Step 6). By keeping your reference points handy (Step 7), you’ll settle the question in seconds next time.

Which method let you settle it fastest? Share your tip in the comments on Wanda-techs.com.