Skip to content
Fona

Service

DevOps and systems administration

We take care of the servers, the databases and the deployment pipeline so they stop being the company’s blind spot. Audit, hardening, automation and ongoing reinforcement when there is no dedicated systems team.

Let's talkSee case studies
Abstract illustration of a continuous deployment loop above a server panel

The problem

Infrastructure only gets attention once it has already gone down

The server has been running for years and nobody dares touch it. Backups are scheduled but have never actually been restored. Deployments happen over SSH by whoever holds the password, with no record of what changed.

Signs this is you

  • A server nobody knows how to update without breaking something
  • Backups that have never been tested with a real restore
  • Manual deployments over FTP or SSH, with no record of what was uploaded
  • Passwords and access shared in an uncontrolled document
  • Nobody watches the server until a client reports the site is down
  • The database keeps getting slower and nobody has audited why

What we do

What we do

Systems administration

Linux servers, access management, updates and hardening. With a clear inventory of what runs where and who can touch it.

Databases

PostgreSQL and MySQL: performance tuning, indexing, replication and backups with a genuinely rehearsed restore, not just a scheduled copy.

CI/CD and infrastructure as code

Automated deployments with Docker and GitHub Actions, reproducible environments and infrastructure changes versioned like the rest of the code.

Monitoring and agile delivery

Alerts before the client notices the outage, runbooks for what can go wrong, and the work organised on a Scrum board instead of a chat nobody rereads.

IaC
infrastructure as code, versioned like everything else
100%
automated deployments, nothing uploaded by hand over FTP
Backups
with a tested restore, not just a scheduled copy

How we do it

How we do it

  1. Infrastructure audit

    Inventory of servers, access, backups and failure points. We end with a risk list ranked by impact, not a generic report.

    DeliverableInventory and prioritised risks

  2. Hardening and automation

    We close the most urgent risks first: access control, tested backups, automated deployment. Every change is tested before it touches production.

    DeliverableTested backups and automated deployment

  3. Monitoring and runbooks

    Alerts wired to what actually matters, and a written procedure for every common incident, so the response doesn’t depend on who happens to be reachable that night.

    DeliverableLive alerts and runbooks

  4. Ongoing operation or handover

    We can stay on with monthly support, or hand it over fully documented and trained to your team. It’s your call, not a dependency.

    DeliverableMonthly support or documented handover

Technologies

  • Linux
  • Docker
  • GitHub Actions
  • AWS
  • PostgreSQL
  • MySQL
  • Git
  • Agile / Scrum
  • Project management

FAQ

What people ask us

Do you replace our hosting provider?

No. We work on top of the infrastructure you already have, migrate it if needed, or set it up from scratch if it doesn’t exist. We’re not a hosting company: we manage what runs on top.

Is this only for companies with no technical department?

We also work as reinforcement for a small development team that has nobody dedicated to systems. We don’t replace the internal team — we cover the part nobody currently has time to do properly.

How is it billed?

The initial audit is fixed price. From there, a fixed monthly fee for ongoing support, or a fixed price per phase for a one-off hardening or migration project.

Are you tied to a specific cloud provider?

No. We usually work with AWS, but the choice depends on where the client’s infrastructure already sits, or on what makes sense for the project — not on a commercial alliance of ours.

Next step

Half an hour well spent

Walk us through the problem on a short call. You leave with a first read on how we'd approach it and what it would involve — no commitment, no sales deck.

Let's talkHow we work