For server monitoring, two open source solutions come up most often: the Prometheus and Grafana pair, and Zabbix. With either of them you can watch server load and service availability, and get an alert if something breaks. Their underlying approach differs, however, and that matters more when choosing than the feature lists do. In this article we compare them and describe when we recommend which.
Two different approaches
Zabbix is a unified monitoring system: a single server, a database and a web interface in which you configure everything, from adding hosts to alerts. The Zabbix agent installed on the machines collects the data, but Zabbix can also take measurements over SNMP, IPMI, HTTP and in many other ways. It works with templates: when you add a Linux server or a switch, the template brings the measured values, the graphs and the alert thresholds.
Prometheus is a time series database and data collector that queries (scrapes) the metrics from the monitored targets over HTTP. The metrics are made available by so-called exporters: node_exporter for the Linux machine’s data, mysqld_exporter for MySQL’s, and blackbox_exporter for the availability of websites and ports. Prometheus itself only collects, stores and evaluates rules; sending out alerts is done by Alertmanager, and visualisation by Grafana. That is three or four separate components, which you connect with configuration files.
The current versions
- Zabbix: the current LTS is 7.0, with full support until 30 June 2027, and the latest standard release is 7.4. According to the vendor’s announcement, the next LTS, Zabbix 8.0, is expected later in 2026. Since 7.0, Zabbix has been licensed under AGPLv3.
- Prometheus: a new minor version comes out every six weeks, with one LTS release a year. The current LTS is 3.13, released in July, which gets fixes until 31 July 2027. Its licence is Apache 2.0.
- Grafana: major version 13 was released in April 2026, at GrafanaCON. The free Grafana OSS is licensed under AGPLv3.
Comparison
| Prometheus + Grafana | Zabbix | |
|---|---|---|
| Data collection | the server queries the exporters (pull) | agent (active or passive), SNMP, IPMI, HTTP, SSH |
| Data storage | its own time series database (TSDB), 15 days by default | MySQL/MariaDB or PostgreSQL (with TimescaleDB) |
| Configuration | YAML files, easy to keep under version control | web interface and templates, can be automated via the API |
| Query language | PromQL, very flexible | predefined items and trigger expressions |
| Alerting | Alertmanager: grouping, silencing, routing | built-in triggers, actions, escalation |
| Visualisation | Grafana, very rich | built-in dashboards; there is also a Grafana plugin |
| Discovery | Kubernetes, cloud and DNS-based service discovery | network discovery, low-level discovery (LLD) |
| Network devices | snmp_exporter, with more manual work |
strong, with many ready-made templates |
| Containers, Kubernetes | its strength; this is the de facto standard | supported, but not what it was designed for |
| Long-term storage | needs a separate tool (Thanos, Mimir, VictoriaMetrics) | built in, with trend data |
| Learning curve | steeper, more components | slower start in the interface, simple after that |
When should you choose Zabbix?
- Mixed environment: Linux and Windows servers, switches, routers, printers and UPSs, where SNMP devices make up a significant share.
- Classic infrastructure: virtual machines, physical servers, a few dozen or a few hundred machines, in a setup that rarely changes.
- Few operations staff: one web interface, one database, one package source; with the ready-made templates, adding a new machine takes minutes.
- Multiple sites: a Zabbix proxy collects the data at a remote site and forwards it in one go to the central server.
We wrote about the new features of Zabbix 7.0 in detail when it was released.
When should you choose Prometheus and Grafana?
- Containers and Kubernetes: targets come and go, and Prometheus’s service discovery keeps track of this automatically.
- Application metrics: your own application exposes its own metrics (number of requests, response time, error rate) directly on a
/metricsendpoint. Most modern software – from Proxmox through GitLab to Traefik – provides Prometheus-format metrics out of the box, or has an exporter for it. - Infrastructure as code: every setting is a text file that can be stored in Git and deployed from CI/CD.
- Flexible analysis: with PromQL you can express almost any question, for example CPU utilisation per machine, averaged over the last five minutes:
100 - (avg by (instance) (rate(node_cpu_seconds_total{mode="idle"}[5m])) * 100)
An alerting rule that fires if free space on a file system drops below 10 per cent:
groups:
- name: disk
rules:
- alert: LowDiskSpace
expr: node_filesystem_avail_bytes{fstype!~"tmpfs|overlay"} / node_filesystem_size_bytes < 0.10
for: 15m
labels:
severity: warning
annotations:
summary: "Low disk space: {{ $labels.instance }} {{ $labels.mountpoint }}"
Trying them out in a few minutes
Prometheus and node_exporter on Debian 13, from the distribution’s packages:
sudo apt install prometheus prometheus-node-exporter prometheus-alertmanager
# web interface: http://server:9090, node_exporter: http://server:9100/metrics
It’s best to install Grafana from Grafana Labs’ own APT repository so that you get the current version; then you add Prometheus as a data source, and you can import a ready-made Node Exporter dashboard.
For Zabbix, once you select the operating system, the database and the web server, the zabbix.com download page gives you the exact commands for installing the package repository and setting up the server. The zabbix-agent2 package goes on the machines; the server uses TCP port 10051 and the agent port 10050.
They can be used together
The two don’t exclude each other. A common setup is for Zabbix to monitor the network devices and classic servers, Prometheus the containerised applications, and Grafana to show the data from both in one place (there is an official Grafana plugin for Zabbix). In a small team, however, maintaining two systems is twice the work, so we recommend starting with one and only adding the other if you really need it.
Whichever you choose, don’t forget two things. The monitoring system shouldn’t run on the same machine or in the same data centre as what it monitors, because in a bigger outage it becomes unreachable too. We also wrote about this in connection with the Cloudflare outage. The other is to install the agents, exporters and templates from code, for example with Ansible, as we showed in our Ansible for beginners article, so that a new machine is automatically added to monitoring as well.
Our recommendation in brief
- A few dozen servers, network devices, few operations staff: Zabbix.
- Containers, Kubernetes, in-house applications: Prometheus and Grafana.
- A mix of both: Zabbix for the infrastructure, Prometheus for the applications, Grafana as a shared interface.
We also help with building monitoring systems, setting up alerts and running them; you can read more on our Infrastructure and operations page.