Introduction
The best way to have visbility on what is going on in your homelab is to have monitoring over everything from Networking, Storage, and Compute hosts. To do this, I have decided to use Zabbix in HA on my Kubernetes cluster, while integrating it with my PostgreSQL HA cluster to ensure ultimate uptime of my monitoring solution.
Architecture
Below is a simple diagram showing the physical infrastructure backing this Zabbix deployment.
Kubernetes
To move this to Kubernetes, ideally we would want a PostgreSQL cluster to store everything in. We can achieve this by using a PostgreSQL operator in Kubernetes, which automates provisioning StatefulSets of databases, which automates the High Availability by switching the Kubernetes service to the master node.
- PostgreSQL HA cluster using Zalando PostgreSQL Operator
- Zabbix Server in HA with LINDS-Kubernetes/zabbix
- TrueNAS Host to host the PostgreSQL data.
Monitored objects
- 2x Core switches
- 5x Wireless AP’s
- Physical Host hardware (HPE and Dell)
- VMware ESXi Hypervisor OS
- 2x TrueNAS VM
- 2X OPNSense VM
Alerting
To receive alerting, I’ve set up Discord webhooks as well as ZBX Viewer to receive any notifications on my phone if there is any alerts.
Screenshot
Dashboard
Hosts
