Clockwork Control is the self-hosted fleet control panel and monitoring brain for our agency’s WordPress fleet. It watches every server we host, every site we run, and every signal that tells us something is about to break — so we get a heads-up before clients do.
What it does
A few things, all at once:
- Watches the servers — every 5 minutes Clockwork Control pulls metrics from each server’s cloud provider — DigitalOcean droplets give CPU, memory, disk, and load; Azure VMs give CPU and memory; Hetzner Cloud servers give CPU only — and tags any server that’s running hot.
- Watches the sites — every monitored site gets an HTTP probe a few times an hour. If a site stops responding for two probes in a row, we get a Mattermost alert before the client emails us.
- Watches for trouble — nginx logs are tailed, Gatekeeper and Wordfence security data is ingested, and IPs that misbehave land in a review queue. Repeated abusers get banned via fail2ban over SSH.
- Runs the care-plan checks — nightly Lighthouse performance scans (GTmetrix), weekly Sucuri SiteCheck, daily WordPress core file integrity, daily blacklist checks against Spamhaus / URLHaus / Google Web Risk.
- Talks to clients via the Companion plugin — every Companion-equipped site has a Tools → Clockwork Control section in wp-admin showing the same data we see, formatted for them.
What it isn’t
Clockwork Control is not a SaaS product. It runs on your own infrastructure (a local workstation, a dedicated Linux server / mini-PC, or a home lab VM on your office network). A Mac is NOT required — Clockwork Control runs natively and identically on Linux (Ubuntu, Debian, Fedora, Arch) and macOS. There’s no public URL, no client login here. Every account on Clockwork Control is an agency account.
It’s also not a replacement for your hosting providers or server control panels — it sits side-by-side with them. Whether your fleet is on SpinupWP, Cloudways, Pressable, WP Engine, Kinsta, or custom cloud servers, those platforms provision and run the infrastructure; Clockwork Control watches over them, catches issues early, and serves as your agency’s unified monitoring brain.
Where to start
If you’re new to the team:
- Read Concepts → Server, Site, Care plan, Hosting tier so the vocabulary on every other page makes sense.
- Skim Features → Servers & health and Features → Uptime monitoring — those two cover 80% of what you’ll touch day-to-day.
- Bookmark Runbooks → A site is down — what now? for when the alert hits.
If you’ve been around a while and you’re just looking something up — sidebar, ctrl-F. The whole catalog is on this page’s left.
How these docs work
Markdown files in resources/docs/. Edit, save, refresh. The sidebar rebuilds itself from the filesystem — drop in a new .md file and it shows up. See _conventions.md in the docs root for the frontmatter shape.
When you ship a feature, add or update the matching page under Features. Bump the updated: date at the top. The byline is whoever’s editing — change it from mine to yours when you take over a page.
