Server administration and hosting
Linux hosting run properly: ISPConfig, Nginx and Apache, Redis, TLS, backups that have actually been restored.
I run production ISPConfig servers of my own, across dozens of domains and mailboxes. That is a different discipline from building sites on top of someone else’s hosting, and it is the part most WordPress developers quietly outsource.
What this covers
- Provisioning and hardening: Nginx or Apache, PHP-FPM pools, UFW, fail2ban, TLS.
- Redis and object caching configured per site, with the sockets and permissions right.
- Mail that works: Postfix and Dovecot, relay through a provider such as Amazon SES, and the SPF, DKIM and DMARC records aligned and verified.
- PHP major version upgrades planned against your real plugin set rather than hoped through.
- Backups that get restored on a schedule, because an untested backup is a hypothesis.
- Panel and origin hardening, including Cloudflare Tunnels and authenticated origin pulls so the server is not reachable directly.
Why it matters for a WordPress site
Most of what gets blamed on WordPress is a server problem: PHP-FPM workers exhausted by a runaway scheduler, an object cache pointed at the wrong socket, a mail queue stalled in a pre-queue scanner. If nobody is looking at the layer underneath, those get diagnosed as “the site is slow” and never actually fixed.
How it is bought
As a monthly care plan rather than by the hour, because this work is only worth anything when it is continuous. Three tiers, from a baseline of tested updates, verified backups and monitoring, up to a plan with sixteen hours of work a month, a four working hour response and the emergency call-out fee waived. The full breakdown is on the pricing page.