Monitoring and maintenance

Monitoring is judged on alerts handled, not on alerts produced.

Divisions assembled: Software and systems Networks and communications Physical and cyber security Power and backup supply

The issue

Delegating operations is legitimate; losing the ability to take them back is not.

Monitoring always decays the same way. A tool is installed, the supplied templates are enabled, and the console produces hundreds of events a day. Nobody can handle them, the team learns to ignore the colour red, and the real failure passes unnoticed among ordinary alerts.

Maintenance raises a separate question, and it is contractual. A provider working under a shared account makes every action unattributable; a contract with no exit clause turns delegation into dependency, and dependency is only noticed at the moment of leaving.

The third point is scope. Monitoring limited to servers ignores the most frequent causes of outage, which are power, cooling and the carrier link. It then reports the effect and never the cause.

Delegating without losing control, schematic

01Taking over: Inventory, access, backups, documentation. Committing on an unknown estate means committing to something else.
02Chosen indicators: Few, kept because their degradation is certain and its effect immediate.
03Commitments by severity: Acknowledgement and restoration kept distinct, hours covered, timestamping written down.
04Operation: Incidents handled within committed times, interventions recorded, reviews about discrepancies.
05Named accounts: A shared account makes every action unattributable, and reconstruction impossible after an incident.
06Reversibility: Documentation handed over during the contract, backups accessible, a handback schedule written on day one.

Schematic. The accented path closes back onto the takeover: that is what separates delegation from dependency.Typical commitment figures. Recovery time is set per severity and per site; this one is a sizing target, not a contractual commitment.

Outcome

What you get

An estate whose state is known remotely, whose incidents are timestamped and attributable, whose restoration times are committed by severity, and whose operation you can take back with what has been handed to you.

Scope

What the solution includes

Metrics: Collection of indicators from systems, network, backups, security and power; retention long enough to tell a slow drift from a one-off incident.
Alerts and procedures: Few rules, each tied to a justified threshold, a written action and a person; defined escalation and planned silences during interventions.
Day-to-day operations: Scheduled updates, backup management with timed restoration checks, account management, and tracking of support contracts and end-of-maintenance dates.
Field maintenance: Preventive maintenance of technical and security installations, wear parts held in stock, interventions recorded, and periodic tests run to a programme.
Service commitments: Acknowledgement and restoration times kept distinct, by severity, hours covered, on-call arrangements, named contacts, and how an incident is recorded.
Reversibility: A maintained inventory, operating documentation handed over periodically, backups accessible to the client, and a handback procedure and schedule at contract end.

Sequence

How it is run

01
Taking over
Inventory, access, backups and documentation, before any response commitment. Committing on an unknown estate means committing to something other than what exists.
02
Tuning the monitoring
Few indicators, chosen because their degradation is certain and its effect immediate. An alert that repeats without anyone acting leaves the console.
03
Operation
Incidents handled under committed times, interventions recorded, and periodic reviews with the client about discrepancies rather than activity volume.
04
Handback
Prepared from day one, executed to a schedule. Without a schedule, handback stretches out to the date when nobody is available to perform it.

Commitments

What is verified at acceptance

These commitments can be held against us. Written into a tender, they filter out the responses that will not hold — including ours, if it does not hold them.

Every member of our staff works under a named account, and their remote access is logged.
Acknowledgement and restoration times are distinct, by severity, with the hours covered.
How an incident is timestamped and recorded is written down, failing which no time commitment is verifiable.
Restorations are tested periodically, timed and recorded.
Operating documentation is handed to the client at regular intervals, not only at contract end.
The client can access its own backups, in a format readable without our tools.
The handback clause states what is returned, in what format and to what schedule.

Before committing

Questions to settle internally

An organisation that cannot answer these questions is not ready to launch the programme. Saying so is better than selling it a study.

Which indicators genuinely justify waking someone at night?
Which interruption, and for how long, carries a real cost for the organisation?
Who records the incident, and when does the clock start?
What access must our team hold, and how is it traced?
What happens if a committed time is missed?
What must you get back in order to run operations without us?

Evidence

What we have delivered on this scope

International organisations2022
European diplomatic mission — multi-system security
Complete overhaul of access control, video surveillance and backup power for a chancery in service.
Systems merged
3
Service interruptions
0
Reference KP-2022-014
Enterprise and sensitive-site security2025
Industrial group — security supervision centre
A single control room for six sites whose security systems did not talk to each other.
Sites connected
6
Threat scenarios rehearsed before commissioning
3
Reference KP-2025-002