I use Icinga to monitor my entire infrastructure – from Proxmox and vSphere to SAN and the Internet, down to small services. I appreciate the flexibility and simplicity. Icinga is continuously developed and has an active community. I like the further developments, such as IcingaDB and IcingaX509. Icinga Director is one of the best things about Icinga for me and offers many automation features. I also appreciate that I can automatically provision hosts and agents. A new Proxmox host is automatically assigned the appropriate service sets based on its note and receives both Proxmox and Linux checks via the standardized Linux agent. With Ansible modules, I can implement the installation easily and fully automatically. I don't have to work in config files, and Icinga is not as memory-hungry as Nagios. Moreover, I can rely on Icinga to automatically notify me when a problem occurs – it lets me sleep peacefully.

The main features of Icinga include the exceptional ability to customize checks, allowing targeted monitoring of virtually any service through custom-developed scripts. Icinga's interface is quite complex and requires thorough study to fully exploit its various functionalities; at the same time, however, it is very flexible and versatile. Icinga easily integrates with external systems for sending alerts, for example via email or Slack. The product is open-source and has a low adoption cost, with an excellent cost/utility ratio. Moreover, it does not require significant computing resources: except in cases of monitoring complex systems, performance is excellent. There are no AI features.

We are a managed service provider and have been running Icinga 2 for more than five years. It monitors our own infrastructure and our customers’ environments, between 100 and 500 hosts in total. The distributed architecture is the main reason we stay with it. Each customer environment gets its own satellite zone, so we keep tenants cleanly separated and still see everything in one central place. Icinga Director is the second big win: host templates, service sets and baskets let us roll out a consistent monitoring standard instead of writing config by hand. The plugin ecosystem is very flexible. Because it is Nagios-compatible, we can combine community plugins such as the Linuxfabrik Monitoring Plugins with our own checks. And since it is open source with no per-host licensing, costs don’t grow every time we onboard a new customer.