To know about an infrastructure problem before users do, you need monitoring. There are many tools and each suits something different. Let us go through the best-known ones and explain what they are for and which to choose when.

What monitoring is for

A monitoring tool continuously watches servers, network, services and devices and alerts when something stops working or a problem is approaching (an outage, a full disk, high temperature, slow response). Thanks to this you react before users feel the outage.

Well-known self-hosted tools

These tools you deploy on your own server or VPS. They give full control, but they need to be set up and maintained.

Zabbix

A comprehensive and powerful open source tool for monitoring servers, network and applications. It handles large environments, has elaborate alerts and overviews. It is versatile, but more demanding to set up.

  • For whom: medium and larger companies that want one robust system for everything.

Nagios

A classic and pioneer of monitoring. It watches the availability of services and devices and works on the principle of plugins, so it can be extended to almost anything. It is proven, but its interface feels older.

  • For whom: environments where Nagios is established, and teams that want maximum extensibility.

Icinga

A more modern successor to Nagios with a clearer interface and better handling of configuration. It keeps compatibility with Nagios plugins, but feels fresher.

  • For whom: those who want the power of Nagios, but in a more modern guise.

LibreNMS

A tool focused on network monitoring via SNMP with automatic discovery of devices. It finds switches, routers and servers on the network itself and draws clear graphs. It is pleasant to use.

  • For whom: companies with network infrastructure (switches, routers) that want an overview of the network.

Observium

A similar focus to LibreNMS, that is network monitoring via SNMP with an emphasis on clear graphs and simplicity. LibreNMS once originated from it.

  • For whom: those who want simple and clear network monitoring.

Smokeping

A specialist in measuring latency and connection quality (ping). It draws detailed graphs of response and outages over time, so it nicely reveals fluctuations and line problems.

  • For whom: as a complement for watching connection quality and latency.

SaaS monitoring: ePulz.io

Self-hosted tools are powerful, but they need to be installed, set up, secured and maintained, and they themselves must run somewhere and be available. An alternative is monitoring as a service (SaaS), which runs outside your infrastructure and does not need operating.

One such solution is ePulz.io, a monitoring platform hosted in the EU that watches website availability, SSL certificates and domains and alerts you immediately on an outage, without you having to install and maintain anything. Since it runs outside your network, it also detects an outage that internal monitoring would miss (because it would go down with it). You can try it for free during a 7-day trial.

Which tool to choose when

  • A large and complex environment, all in one: Zabbix.
  • Network infrastructure (SNMP), an overview of the network: LibreNMS or Observium.
  • Classic service monitoring with extensibility: Nagios or the more modern Icinga.
  • Latency and connection quality: Smokeping as a complement.
  • Website availability without operating your own system: SaaS like ePulz.io.

In practice they are often combined: a self-hosted tool for internal infrastructure and an external SaaS for website availability from the outside.

Conclusion

There is no single best monitoring tool, there is a suitable one for a given need. Zabbix for comprehensive monitoring, LibreNMS and Observium for the network, Nagios and Icinga for services, Smokeping for latency and SaaS like ePulz.io for website availability without worries. The important thing is to monitor at all and to know about outages right away.

Want to set up monitoring of your network, servers or website availability? Get in touch, we will design and deploy a tailored solution.

This article is part of our Computer networks overview.