High-availability Zabbix

Context

Monitoring needed to support production IT/OT infrastructure and provide a reliable foundation for detecting, prioritising and analysing incidents.

Responsibility

End-to-end ownership of Zabbix: administration, maintenance, platform evolution, monitoring and alert configuration, connected to support operations.

Architecture

A high-availability architecture for Zabbix and related services, with decoupled components and PostgreSQL access through a virtual IP. Published diagrams are conceptual and do not reproduce real topologies.

Implementation

Definition of templates, hosts, alerts and dependencies; platform maintenance; Grafana integration; and automation of operational processes linked to monitoring.

Challenges

Balancing sensitivity and noise, preserving availability of the monitoring platform itself and presenting useful information to different support levels.

Outcomes and lessons

A centrally managed observability platform with more consistent alerting processes and a technical foundation for analysis and operational improvement.

Monitoring is a production service: it needs availability, maintenance, data governance and a clear relationship with incident response.