[ 08 — HOSTING ]
Managed servers, deployments, backups and monitoring on infrastructure you own.
[ 01 — THE PROBLEM ]
What this usually looks like
Before
- The site went down at the weekend and you found out from a customer on Monday.
- The certificate expired because the renewal was somebody's calendar reminder.
- Deploying means somebody copying files over SSH and hoping.
- The server is on an account belonging to a developer who left.
After
- Monitoring that pages a human, with a documented response time rather than best effort.
- Certificates issued and renewed automatically, and alerted on well before expiry.
- One pipeline: tests, build, deploy, and a rollback that takes a minute.
- Everything on accounts in your company's name, with the access list written down.
[ 02 — WHAT YOU GET ]
Deliverables, not deliverable-sounding words
Provisioned infrastructure
Servers, networking, firewall and DNS, on your cloud account or hardware, documented as code.
Deployment pipeline
Push to main, tests run, image builds, deploy happens. Rollback is one command.
TLS and domains
Certificates issued and renewed automatically, HTTPS enforced, security headers set.
Backups
Automated, offsite, encrypted, with a tested restore and a documented recovery time.
Monitoring and alerting
Uptime, certificate expiry, disk, memory and error rates, with alerts that reach a person.
[ 03 — HOW IT WORKS ]
The steps, and how long each one takes
- 01
Review what you run
2–3 daysWhat the application needs, what it talks to, and what the real traffic looks like.
- 02
Size and provision
3–5 daysRight-sized infrastructure in your name, built from code rather than clicked together.
- 03
Migrate
1–2 weeksMove the application and data, run both in parallel, cut over DNS when it is proven.
- 04
Harden and monitor
2–3 daysFirewall, fail2ban, unattended security updates, monitoring and alert routing.
- 05
Run it
ongoingPatching, backups, certificate renewals and incident response on an agreed retainer.
[ 04 — EXAMPLES ]
What people use this for
Moving off expensive managed platforms
The same application on your own servers, usually at a fraction of the monthly cost, still fully managed.
Taking over an orphaned server
A box nobody has logged into for a year: audited, patched, documented and brought under monitoring.
Containerised application hosting
Docker or Kubernetes, with a real pipeline, staging environment and rollback.
Hosting is not the hard part. Noticing at 3am that it has stopped is.
Owned by you, run by us
Everything runs on accounts and hardware in your name. We set up the servers, deployments, certificates and backups, then monitor them and fix what breaks. If you ever want to take it over or move it elsewhere, you already own everything needed to do that.
Boring on purpose
Infrastructure defined as code, deployments through one pipeline, rollback in a minute, and alerts that reach a person. None of it is clever. Clever infrastructure is how you end up with an outage nobody can explain.
[ 05 — STACK ]
What we build it with
Chosen because they are boring, well supported and easy to hire for. Nothing here will be abandoned next year.
- Docker
- nginx
- Kubernetes
- GitHub Actions
- Terraform
- Hetzner
- Cloudflare
- Grafana
[ 08 — FAQ ]
Hosting: questions people ask
[ 09 — ALSO IN THIS PILLAR ]
Related services
Let's talk about hosting.
Tell us the problem. We will tell you what we would build, what it costs and how long it takes — before you commit to anything.