- Checked
- 2026-11-14
Your server is down. Check these first.
Most outages on a small server come from a short list: a full disk, memory running out, a service that stopped, a certificate, DNS, or a change made the day before. Work through it in order before you reboot or reinstall anything.
First, is it the server at all?
Open the site from another network, a phone on mobile data for example. Then run the free server health check on the address: it reads the certificate, the redirect to https, the security headers and how long the server takes to answer, from outside your network.
If the address does not resolve, look at DNS before the server. If the certificate has expired, the server may be fine and only browsers refuse it. If the server answers with an error, the machine is up and something on it is failing.
If you can still log in
These commands are read-only. They change nothing, so they are safe to run while you think.
df -hshows whether a disk is full. A full disk stops databases, logs and uploads, and is the most common cause on a small VPS.free -manddmesg -T | grep -i -E 'killed process|out of memory'show whether the kernel has been killing processes for lack of memory.systemctl --failedlists services that stopped.systemctl status nginx(or apache2, php-fpm, mysql) shows the last lines of the one you care about.journalctl -p err -bshows errors since the last boot.- Ask yourself what changed: an update, a new plugin, a config edit, a renewed certificate. The change from yesterday is the first suspect.
If you cannot log in
Open your provider's console, the browser terminal most VPS panels have. It reaches the machine even when SSH does not. Check the provider's status page too: a failed host or a network incident in their data centre looks exactly like your server being down.
Look at the graphs in the provider's panel. CPU pinned at 100 percent or memory full points at a runaway process; a flat line at zero points at a machine that is powered off or stuck at boot.
What not to do
- Do not reboot again and again. A reboot can clear the symptom and the evidence with it, and a server that fails at boot can stop answering altogether.
- Do not reinstall or restore over the top before you know the cause. A restore brings back the same problem if the cause was a full disk or a bad config.
- Do not delete files to free space until you know what they are. Database files and active logs look like clutter and are not.
- Do not paste passwords into chat tools or tickets while you ask for help.
When to hand it over
If the checks above did not name the cause within half an hour, or the server holds something you cannot afford to lose, hand it over. Our Server Rescue is one incident at a fixed $69, with a same-day start listed at $109. If we cannot fix it, you pay nothing.
You describe what you see and how the machine can be reached. Diagnosis comes first: likely causes ranked, read-only checks before any change, no destructive step without a backup in front of it. The fix is verified, and you get the steps in the order they were taken. Access is handed over separately, never pasted into the order form. People answer in support hours, every day 13:00 to 21:00 CET, which is 7am to 3pm Eastern.
Questions people ask
- What does an emergency server rescue cost?
- Server Rescue is $69 for one incident, with a same-day start listed at $109. The price is fixed and confirmed with the scope before work starts. If we cannot fix the problem, you pay nothing.
- Should I reboot a server that is down?
- Once, after you have looked at disk space, memory and the failed services, is reasonable. Rebooting repeatedly hides the cause and can leave a server that does not come back.
- What is the most common reason a small VPS goes down?
- A full disk, followed by memory running out and a service that stopped after an update or a config change.
df -h,free -mandsystemctl --failedanswer all three. - Do you need my password?
- No password goes into the order form. Access is arranged separately, for example with a key you add for the job and remove afterwards.
More server help
- 502 Bad Gateway on nginx
- Disk full: no space left on device
- Error establishing a database connection
- DNS not resolving or still on the old server
- 504 Gateway Timeout
- WordPress white screen or critical error
- Too many redirects
- 403 Forbidden on nginx or Apache
- SSL certificate expired: the fix
- Renewing a Let's Encrypt certificate
- Email from the server going to spam
- Website hacked: the first steps
- Moving a site to a new server
- Linux server hardening checklist
- All server help guides
