Home
›
Articles
›

Moving off the server under the desk

Moving off the server under the desk

By Ibi Hasanli

·

·

3 min read

Moving off the server under the desk is overdue when that box still runs quotes, files, or the line-of-business app your London SMB cannot work without. If you need to move office server to cloud (or at least off the office floor), the goal is not a rewrite. It is a staged move onto hosting you own, with backups that actually restore and a runbook someone else can follow when the office is empty.

How the under-desk box fails

Power fails, the UPS was never tested, and the office goes dark on a Monday. Backups sit on the same shelf as the server, or the “backup” is a USB stick nobody has restored in two years. Holidays mean the one person who knows the passwords is away. Disk dies, RAID was never monitored, remote access is a brittle VPN or a port forward that IT would not sign off today. I’ve watched owners discover the business stopped the moment a silent PC under reception stopped humming, and that remote “access” was a RDP session over a home broadband link nobody documented.

That is not a home-lab hobby problem. It is operational risk: payroll, job sheets, and customer files tied to one piece of hardware in a room that gets warm in summer.

Do not rewrite first

The under-desk box is usually running something that already works for the team. Replacing the whole application before you move is how projects swell and miss the point. Lift what you have onto durable hosting, prove restores, then modernise if you still need to. A big-bang “rebuild in the cloud” is how you spend six months explaining why the old machine is still under the desk.

Stage the move

Survey what actually runs: apps, databases, file shares, printers, who logs in, and when downtime is tolerable. Stage a twin environment on managed hosting or a small cloud footprint. Cut over with a written rollback. Keep the old box offline but intact until the first week of real use has passed. Then retire it. Client keeps the accounts, the DNS, and the runbooks. Eight Mile’s infrastructure and cloud work is built for that path: survey, fixed-price move plan, owned accounts.

Managed hosting is enough for many office apps and sites that need a stable VM, backups, and someone reachable when the box is sick. Full AWS (or similar) is overkill when you do not need elastic scale or a dozen managed services; we already wrote about that trade-off for small SaaS in AWS vs managed hosting, and the same honesty applies here. This piece is about leaving the floor, not picking a cloud certification path.

Ownership after the move

After cutover you should be able to name who owns the hosting account, where restores live, and how a second person restarts the service. Secrets are not on the old PC. Monitoring and backups are not “we’ll set that up later.” If only one person can still touch production, you moved the cable, not the risk. Replace on-premise server habits with company-owned logins and a restore test you have actually run.

Get the box off the floor

If moving off the server under the desk is on your list, start with what is running and a fixed-price plan to host it somewhere durable — not with a Kubernetes fantasy or a rewrite nobody asked for.

If you want a short survey of what’s under the desk and a fixed-price move plan, contact Eight Mile and say what the box runs and how bad an outage would be this week.

Backups
Disaster recovery

Backup restore test

A backup restore test proves copies work on a clock you can live with. Untested backups are hope with a green tick.

·

3 min read

Web design
Care plan

Website care plan vs rebuild

Website care plan vs rebuild: when maintenance is enough for a London SMB site, and when a full rebuild earns the spend.

·

3 min read

Networking
Wi-Fi

Separate guest Wi-Fi from office network

Separate guest Wi-Fi from office network traffic so visitors never sit one hop from finance, payments, or file shares.

·

3 min read

Infrastructure
Cloud

Cloud bill went up and nobody can say why

Cloud bill went up and nobody can say why: usually orphaned resources, missing tags, or one shared account. Fix ownership first.

·

3 min read

Infrastructure
Cloud

One person knows how to deploy it

If one person knows how to deploy it, you have a person, not a process. Fix ownership, access, and a path the team can run.

·

3 min read

AWS
Infrastructure

Leaving a freelancer with your AWS root login

Leaving a freelancer with your AWS root login is a risk even after a good build. How to take back control with named IAM, MFA and a handoff.

·

4 min read

AI
Decision models

Jev vs Laya

Jev vs Laya for UK tech leads: when to call TypeSafe's hosted decision API and when to self-host the open-weight Laya model on your own infrastructure.

·

4 min read

Networking
Office move

Office move IT checklist for London SMEs

Office move IT checklist for London SMEs: internet lead times, surveys, cabling, network design, servers and security before and after moving.

·

4 min read

Networking
VPN

Multi-site office network

Multi-site office network design for London SMEs: site-to-site links, consistent segmentation, accounts you keep.

·

3 min read

Security
Audit

Security audit vs penetration test

Security audit vs penetration test: when to review config first, when a buyer needs a pentest, and usual order for SaaS.

·

3 min read

Security
Git

Secrets in git repository

Secrets in git repository history outlive a deleted file. Rotate first, then stop the next temporary commit.

·

3 min read

Networking
London

Office network for a growing London team

Office network for a growing London team: when the ISP router fails, what a proper office network includes, and why survey first.

·

3 min read

AWS
Infrastructure

AWS vs managed hosting for a small SaaS

AWS vs managed hosting for a small SaaS: when simple platforms fit, and when owned cloud, IAM and infra in git win.

·

3 min read