Maintenance & Support
Maintenance for systems nobody wants to touch
Software does not age through use. It ages through standing still: dependencies go out of support, vulnerabilities become public, knowledge walks out of the door.
The most common case in our maintenance work: an application runs, but the person who built it is gone. Updates get postponed because nobody knows what will break. Two years later a routine task has turned into a project.
We take on systems like that even when we did not build them. The first step is always the same assessment: what runs where, which dependencies are out of date, which risks are urgent, and what has to happen for an update to become a normal event again.
What you get
Assessment and risk picture
A one-off review of dependencies, configuration and operations with a prioritised list of what is urgent and what can wait.
Ongoing updates
Regular updates of dependencies and runtimes, backed by automated tests rather than by hope.
Monitoring and incident response
Monitoring of availability and errors, with agreed response times and a defined route by which an incident reaches us.
Small improvements
A fixed allowance for change requests too small for a project and too important for a backlog nobody works through.
How we work
Take over
Access, documentation, operational knowledge. We write down what we find, including the uncomfortable parts, because that is the basis for everything after.
Stabilise
Urgent vulnerabilities, missing backups and blind spots in monitoring first. Convenience comes later.
Automate
Updates become routine once tests and deployment are automated. That is the actual investment, everything else is follow-up cost.
Keep the rhythm
A fixed cadence for updates and a short, readable report of what happened. Not an invoice nobody can place.
Technologies involved
- Automated dependency and vulnerability scanning
- Automated tests as a precondition for low-risk updates
- Monitoring of availability, errors and certificates
- Reproducible environments through Docker
- Documented backup and restore paths
Common questions about this service
Do you also take on systems you did not build?
Yes, that is the normal case. An assessment comes first and makes effort and risks visible. You then decide whether ongoing support makes sense.
What response times are possible?
Response times are agreed based on how critical the system is. Short response times require monitoring to be in place, otherwise we learn about an incident from your users.
What does maintenance cost?
Maintenance is agreed as a monthly allowance whose size follows from system size, criticality and the response time you need. The initial assessment provides the basis for that estimate.