
Summary: When the site goes down, first confirm the problem isn't just your network. Then follow the order: does the domain resolve in DNS? Is the certificate valid? Does the server respond, and with what HTTP code? Each answer points to a different cause: expired domain, changed DNS, expired certificate, suspended or overloaded hosting, or an error in the site itself.
A site being down is always urgent: every hour offline is a customer who can't request a quote. Instead of restarting everything and hoping, follow a checklist. It either solves the problem or, at least, lets you give support precise information.
1. Is it just you?
- Open the site on your phone using mobile data instead of Wi-Fi.
- Ask someone on another network to test.
- Use an uptime checking service that tests from several locations.
If it works for others and not for you, the problem may be your local DNS cache, the office network or the browser. On Windows, ipconfig /flushdns clears the DNS cache.
2. Does the domain resolve?
# Windows
nslookup www.yourcompany.com
# Linux/macOS
dig +short www.yourcompany.com
dig +short NS yourcompany.com
If no IP address comes back, the problem is the domain or DNS: expired domain, changed name servers or a deleted record. For .br domains, check the status at Registro.br; for .com and others, a WHOIS/RDAP lookup shows the registrar and expiry date. See expired domain.
3. Does the server respond? With which code?
curl -I https://www.yourcompany.com
The first line shows the HTTP status code. The most common during outages (MDN: status codes):
| Error | Meaning | Where to look |
|---|---|---|
DNS_PROBE_FINISHED_NXDOMAIN | The domain doesn't resolve | Expired domain, DNS |
ERR_CONNECTION_TIMED_OUT / REFUSED | The server doesn't accept connections | Hosting, firewall, wrong IP in DNS |
NET::ERR_CERT_DATE_INVALID | Expired certificate | SSL; see site not secure |
| 500 | Internal application error | PHP/server error logs; recent change |
| 502 | Gateway/proxy got an invalid response | Application server (PHP-FPM, Node) stopped |
| 503 | Service unavailable | Overload, resource limit, maintenance |
| 504 | Gateway timeout | Slow queries, overloaded server |
| 403 | Forbidden | File permissions, firewall rule, .htaccess |
| "Account suspended" | Provider's page | Pending payment or usage above limit |
4. The most common causes
Expired domain
Site and email stop together. Renew immediately. At Registro.br, an unpaid domain is frozen and, after a period, enters the release process, when anyone can register it.
A recent change
Did someone change DNS, update a plugin, switch the PHP version or edit .htaccess? Recent changes are suspect number one. If there's a backup, restoring the previous version is usually the fastest route. In WordPress, recovery mode and the debug log (WP_DEBUG_LOG) help find the culprit plugin.
Hosting
Instability, suspended account or CPU/memory limit reached. Open a ticket with the time, the error message and the curl -I output.
Hacking
Redirects to other sites, strange pages or a Google warning about a dangerous site. See the checklist in website security.
Planned maintenance? Return 503
If the site will be offline on purpose, Google recommends responding with status 503 (and, if possible, a Retry-After header) instead of 200 or 404. That signals a temporary outage and keeps pages from being dropped from the index over a short stop (Google: HTTP status codes and network errors). Long outages, of several days, can still affect indexing.
How to prevent the next outage
- Automatic monitoring of uptime and of SSL and domain expiry, with email or WhatsApp alerts.
- Daily off-server backups with tested restores.
- Domain auto-renewal and up-to-date contacts.
- Updates tested first on a copy of the site.
- A change log: who changed what, and when.
Our website maintenance includes monitoring, backups and fast fixes, and our hosting comes with support from people who know your site.
Sources and references
- MDN: HTTP status codes
- Google: how HTTP codes and network errors affect Search
- MDN: Retry-After
- Registro.br: what happens if I don't pay
- WordPress: debugging
Frequently asked questions
My site is down but email works. What does that indicate?
That the domain is probably active. The problem likely lies with the site's server, the site's DNS record (A/CNAME) or the application.
What is a 500 error?
An internal server error while running the application, often after an update, a code error or permissions. The error log shows the exact cause.
Does a day offline hurt Google rankings?
Short outages usually have little effect, especially if the server returns 503. Long outages can lead Google to reduce crawling and drop pages from the index.
How long until the site is back after renewing the domain?
Usually a few hours, the time for DNS to be republished and caches to expire.
Is your website down?
We diagnose the cause, bring the website back and set up monitoring so it doesn't happen again.
Read also
- Expired domain: what happens and how to recover it
- "Not secure" website: what it means and how to fix it
- Website security: the basics every business should have
Written by Mário Moreno, founder of Agência hetsi, which has been building websites and handling digital marketing for businesses in Brazil and abroad since 2000. Mário Moreno on LinkedIn