Production servers since 2004, not a lab stack.
The company behind DirektSupport has been operating since 2004. The work has been websites, mail, databases and the Linux under them, on machines other people own.
Panels changed. The failure modes did not.
In 2004 the usual job was a Linux host, Apache, mail and a DNS zone. The names now are NGINX, PHP-FPM, MariaDB, HestiaCP, DirectAdmin and Cloudflare. The incidents are the same: a full disk, a queue of spam, a certificate that expired, a site down after an update, a backup that had never been restored.
The work we still get called for
SSH gone, the site 502, the database not accepting connections, mail deferred. The first hour is the log and the disk, not a reinstall.
VPS to VPS, VPS to dedicated, Vesta onto something still maintained. Mail and DNS are planned before the cut, not after the bounce.
Password SSH, no offsite copy, a panel and a hand-edited vhost that disagree. We reconcile it, or we say it should be rebuilt.
No client logos, no invented uptime
Servers we administer are under the customer’s name, and the incidents are theirs to publish. This page does not borrow logos. Twenty-two years is the age of the company, not a claim that every hour since 2004 was a server rescue.
The useful part of that time is knowing which fault is the provider, which is PHP, and which is a backup that cannot be restored.
Your server. Your provider. Our Linux expertise.
Have one of those incidents now?
Tell us the provider and what changed. We reply with the approach before any work starts.