Skip to content
Centoculus

Fixes

Server down? We step in and fix it.

Most monitoring tools stop at the alert. When your server or website is down, Centoculus fixes what broke: a quick fix at a fixed price for the common causes, and a deep fix for everything else, around the clock, nights and weekends included. On your own server, or in your webspace on shared hosting with SSH access.

Opens the order form in the app. You see the price before you order.

Not sure whether it is down for everyone? Run the free website down checker first.

Want to look yourself first? Our guide Server down: what to check first walks through the usual causes in order.

Quick fix

Fixed price

Runs automatically. Restarts a stopped service after its configuration test passes, frees disk space, renews certificates and makes small configuration fixes with a backup copy first. If it cannot solve the problem, it tells you what it found and what to do next.

  • Never installs packages, never reboots the server
  • Never deletes website files or databases
  • Credited in full against a deep fix ordered within 24 hours

Deep fix

Billed in 15-minute blocks

We take over the rest: updates, repairs, complex problems too, around the clock. Each started block counts; the estimate you see before you order is the upper limit. If we could not carry it out, you pay nothing.

  • Price and estimated minutes before you order
  • Every session on your server logged and shown to you
  • Free follow-up check if something stopped working because of our work

In the app the two are called Quick Fix and Deep Fix.

What we need

Monitoring works without any access. For a fix we need a way onto the server, in one of three ways:

  1. The Responder

    A small program installed with one command as root from our signed package source. It fetches its work over outbound HTTPS, so it works behind a firewall or NAT without SSH, and you can limit it to an allowlist of commands.

  2. SSH as root on your own server

    VPS, dedicated or cloud server. Run one command as root, or give us a password once; we delete it right after the key is in place.

  3. SSH access to your webspace

    Shared hosting with a restricted SSH user. SFTP alone is not enough. Run one command as that user, or add our public key in your host's control panel.

Set up in a minute. The server agent for monitoring is separate: it only reports metrics and is never a way in for fixes.

Access expires when you decide, and you withdraw it with one click.

In your webspace at a hosting provider

With SSH access we fix what lies inside your webspace: logs, .htaccess, PHP settings, permissions, caches. If the cause is at your host, we tell you and you get a ready-to-send message for them.

Without SSH access, only your host can fix it. Monitoring and the status page work either way, and the website down checker gives you the message for your host.

Safety

  • Configuration test before every restart
  • Backup copy before every change
  • Every session logged and visible to you
  • Our key only works from our operations servers
  • Withdraw access with one click

In your language

What we found and did, and the text for your hoster, reach you in your language, as do the order and every mail about it. The app speaks more than 20 languages.

  • App: every user picks their language
  • Alerts by e-mail, push and Telegram, and the all-clear, in the recipient's language
  • Status pages in the visitor's language
  • Repair reports and the text for your hoster in your language
  • Invoices and account mails in your language
  • Write to support in your language

Self-healing servers, without building it yourself

Self-healing servers and auto-remediation usually mean runbooks, scripts and an orchestration tool that someone has to write and maintain. For a single server or a handful of websites, automatic fixes do the same job: a check fails, a waiting time rules out a short glitch, and a quick fix restarts the stopped service after its configuration test passes, frees the full disk or renews the certificate.

Every automatic fix follows the same safety rules as one you order, is logged and shown to you, and is paid from your prepaid cover with a monthly limit you set. When the quick fix cannot solve it, you get its findings and can order a deep fix in one step. How to switch it on: automatic fixes below, and the server monitoring that triggers them.

Automatic fixes when a server goes down

Switch them on per server or monitor, and a fix starts when a check fails, paid from prepaid cover: businesses with a valid VAT or tax ID from their balance, everyone else from the included minutes of a 12-month plan "Automatic fixes", like a phone contract. Unused included minutes lapse at the end of the term; when they are used up or the term ends, you get a new plan offer to accept or decline. With safeguards:

  • A waiting time against short glitches
  • A monthly spending limit you set
  • Only with enough balance or included minutes
  • No second fix while one is running
  • Optionally only in a time window you choose

Automatic fixes need the Responder or SSH access, like every fix. The monitoring that triggers them: server monitoring.

How you pay

Businesses with a valid VAT or tax ID pay from their prepaid balance: the amount is reserved when you order, charged when we are done, and released if we could not carry it out.

Everyone else prepays the estimated minutes at checkout, in full 15-minute blocks, with tax shown there. If we need fewer minutes, the rest stays with you as a minute voucher for your next fix; it is not paid out and expires after 12 months without an order. For automatic fixes you take a 12-month plan "Automatic fixes" with included minutes; unused included minutes lapse at the end of the term.

Fixes can be ordered on every plan, the free plan included. If we could not carry out the fix at all, you get your money back.

Follow-up check

Has something stopped working after a deep fix? For 30 days we check again for free and fix it when our work caused it. If the logs show the problem existed before, the work is billed in 15-minute blocks, and you decide first.

Questions about fixes

Is it safe?

A quick fix never installs packages, never reboots and never deletes website files or databases. Every restart is preceded by a configuration test, every change by a backup copy. Every session is logged and shown to you, our key only works from our operations servers, and you withdraw access with one click.

What if the fix does not work?

If a quick fix cannot solve the problem, it tells you what it found and what to do next, and its price is credited in full if you order the deep fix within 24 hours. If we could not carry out a fix at all, nothing is charged.

Do I pay if you cannot fix it?

No. You pay for what we could carry out. Businesses get the reserved amount released; everyone else gets the prepaid minutes refunded.

Do I need a paid plan?

No. Fixes can be ordered on the free plan too: you prepay the estimated minutes at checkout.

Do you replace my host's support?

On your own server, we do most of what a host's support would do, around the clock. In a webspace we fix what lies inside it; what lies at the host (the web server itself, the network, the hardware) stays with the host, and you get a ready-to-send message for them.

Is there emergency server support at night and on weekends?

Yes. Quick fixes and deep fixes run around the clock, nights and weekends included. When your server is down, order a fix in the app; with automatic fixes switched on, it starts without you. We promise no response time; you follow the status in the app.

Can it run without me?

Yes, with automatic fixes: switch them on per server or monitor, set a waiting time and a monthly limit, and a fix starts when a check fails. Businesses pay from their prepaid balance; everyone else takes a 12-month plan "Automatic fixes" with included minutes, like a phone contract. Unused included minutes lapse at the end of the term; when they are used up or the term ends, we offer a new plan that you accept or decline.