How to Install Zabbix on Ubuntu 24.04 — Complete Monitoring Stack Guide

Infrastructure monitoring is no longer optional. If you are running production servers, web apps, databases, or network gear, you need a unified view of health, performance, and availability — and you need alerts the moment something drifts from normal. This tutorial walks you through installing Zabbix 7.0 LTS on Ubuntu 24.04, from a fresh VPS to a hardened, TLS-secured monitoring server complete with agents, auto-registration, Grafana dashboards, and Slack/email notifications.

Zabbix is the most widely deployed open-source enterprise monitoring platform in the world. Unlike SaaS tools that charge per host or per metric, a self-hosted Zabbix instance on a single VPS can comfortably monitor hundreds of hosts and tens of thousands of metrics — for the flat cost of the server itself.

Skip the manual work? CloudCore VPS plans give you the headroom Zabbix needs to collect, store, and graph metrics at scale. Launch a Professional VPS and follow this guide end to end in under an hour.

Table of Contents

What is Zabbix?

Zabbix is an open-source, enterprise-grade monitoring platform that collects metrics, logs, and events from servers, network devices, applications, databases, containers, and cloud services. Originally released in 2001, it is now maintained by Zabbix LLC under the GPLv2 licence and has a massive global community of sysadmins, MSPs, and enterprise IT teams.

At its core, Zabbix is composed of four parts: a server that collects and processes data, a database (MySQL/MariaDB, PostgreSQL, or TimescaleDB) that stores metrics and configuration, a web frontend that provides dashboards and admin UI, and agents that run on monitored hosts to collect system-level data. Optional proxies let you scale collection across datacenters or monitor hosts behind firewalls.

Zabbix supports an unusually wide range of monitoring techniques out of the box: native agents for Linux, Windows, macOS, FreeBSD, AIX, and Solaris; agentless collection via SNMP v1/v2c/v3, IPMI, JMX, ODBC, SSH, and Telnet; HTTP/HTTPS web scenario checks; synthetic transaction monitoring; log file parsing; and over 300 pre-built integration templates covering everything from PostgreSQL and Redis to Cisco routers, VMware ESXi, AWS CloudWatch, and Kubernetes. The built-in auto-discovery engine can scan IP ranges, detect new hosts automatically, and apply monitoring templates without manual intervention.

The platform scales from a single-server install monitoring a dozen hosts to large clustered deployments tracking tens of thousands of hosts and millions of metrics per minute. Version 7.0, released in 2024, is the current long-term support (LTS) release with support through 2029, and it introduces proxy load balancing, a redesigned dashboard engine, and a substantially faster history syncer.

Why Self-Host Zabbix vs Datadog or New Relic?

Commercial observability platforms are excellent products — but they are priced for budgets that most small teams, MSPs, and individual operators do not have. Before you sign a six-figure annual contract, it is worth understanding what self-hosted Zabbix gives you in exchange for a few hours of setup time.

• Flat, predictable cost — A Professional VPS at EUR 19.99/month can monitor 200+ hosts. Datadog charges per host, per container, per custom metric, per log line ingested, per APM trace, and per synthetic check. Real-world bills routinely land between USD 15–40 per host per month once all modules are enabled.
• No metric or host caps — Zabbix does not count your interfaces, containers, or custom items. Collect every metric your agents can produce without a line item appearing on next month's invoice.
• Full data ownership — Every data point stays on your server. For GDPR, HIPAA, SOC 2, or ISO 27001 audits, this simplifies compliance dramatically compared to shipping production telemetry to a US-based SaaS vendor.
• No retention cliffs — SaaS platforms charge extra for anything beyond 15–30 days of raw metric history. Zabbix keeps history and trends for as long as your disk allows, with configurable per-item housekeeping and built-in trend aggregation.
• Custom everything — Write your own UserParameters, external scripts, webhooks, and trigger expressions. Extend the system with custom modules. No vendor gatekeeping on what you are allowed to monitor or alert on.
• Offline and air-gapped support — Zabbix works without any outbound internet connection. Critical for regulated environments and on-premises industrial monitoring.

Cost Comparison: Zabbix vs SaaS Observability

Prerequisites

Before you begin, make sure you have:

• A VPS running Ubuntu 24.04 LTS with root or sudo access
• SSH access to the server
• A fully qualified domain name pointing to the server (required for Let's Encrypt TLS)
• At least 4 GB of RAM (8 GB+ recommended for > 50 monitored hosts)
• At least 40 GB of disk space (metric history grows quickly — plan for 80 GB+ in production)
• Ports 80, 443, and 10051/tcp open (agent connections use 10051 outbound from agents to server)

Recommended Plan: CloudCore Professional

Zabbix is CPU-light but RAM-hungry and disk-I/O-sensitive once you cross a few dozen hosts. For a production monitoring server, we recommend the CloudCore Professional plan:

- 6 vCPU cores

- 12 GB RAM

- 100 GB NVMe SSD

- Unmetered bandwidth

- EUR 19.99/month

This comfortably handles 150–200 hosts with full history retention, the Nginx frontend, MariaDB, and a co-located Grafana instance on the same box.

Connect to your server via SSH to start:

ssh root@your-server-ip

Step 1: Update System Packages

Bring the system up to date before installing anything.

sudo apt update && sudo apt upgrade -y

Install a few utilities that the rest of this guide depends on:

sudo apt install -y curl wget gnupg2 ca-certificates lsb-release software-properties-common ufw

If the kernel was upgraded, reboot:

sudo reboot

Reconnect after a minute.

Step 2: Install MariaDB

Zabbix 7.0 supports MySQL 8.0+ and MariaDB 10.5+. MariaDB is the lighter, faster option for most single-server deployments, and it is what we will use here.

sudo apt install -y mariadb-server mariadb-client

Enable and start the service:

sudo systemctl enable --now mariadb

Run the interactive hardening script and accept the safe defaults (set a root password, remove anonymous users, disable remote root, drop the test database):

sudo mysql_secure_installation

Verify it is running:

sudo systemctl status mariadb --no-pager

Expected output:

● mariadb.service - MariaDB 10.11.x database server
     Loaded: loaded (/lib/systemd/system/mariadb.service; enabled; preset: enabled)
     Active: active (running) since Thu 2026-04-16 10:00:00 UTC; 30s ago

Step 3: Add the Zabbix 7.0 LTS Repository

Zabbix ships an official Ubuntu 24.04 repository package that configures APT sources and GPG keys in one step.

wget https://repo.zabbix.com/zabbix/7.0/ubuntu/pool/main/z/zabbix-release/zabbix-release_7.0-2+ubuntu24.04_all.deb
sudo dpkg -i zabbix-release_7.0-2+ubuntu24.04_all.deb
sudo apt update

Expected output (abbreviated):

Get:1 https://repo.zabbix.com/zabbix/7.0/ubuntu noble InRelease [7,502 B]
Get:2 https://repo.zabbix.com/zabbix/7.0/ubuntu noble/main amd64 Packages [...]
Reading package lists... Done

Step 4: Install Zabbix Server, Frontend, and Agent

Install the four key packages: the server (with MySQL driver), the PHP frontend, the Nginx config package, and the agent to monitor this host itself.

sudo apt install -y zabbix-server-mysql zabbix-frontend-php zabbix-nginx-conf zabbix-sql-scripts zabbix-agent2

Each package serves a specific role:

• zabbix-server-mysql — The Zabbix server daemon compiled with MySQL/MariaDB driver support.
• zabbix-frontend-php — The PHP 8.3 web frontend with all dashboards and admin UI.
• zabbix-nginx-conf — Pre-built Nginx server block for the frontend.
• zabbix-sql-scripts — The SQL schema and initial data dumps used to seed the database.
• zabbix-agent2 — The modern Go-based agent that replaces the legacy C agent with better plugin support.

Step 5: Create Database and Import Schema

Log in to MariaDB as root:

sudo mysql -uroot -p

Create the Zabbix database, a dedicated user, and grant privileges. Use a strong password of your own:

CREATE DATABASE zabbix CHARACTER SET utf8mb4 COLLATE utf8mb4_bin;
CREATE USER 'zabbix'@'localhost' IDENTIFIED BY 'ChangeThisStrongPassword123!';
GRANT ALL PRIVILEGES ON zabbix.* TO 'zabbix'@'localhost';
SET GLOBAL log_bin_trust_function_creators = 1;
FLUSH PRIVILEGES;
EXIT;

The log_bin_trust_function_creators = 1 flag is required so that the schema import can define stored functions. We will disable it again after the import.

Import the Zabbix schema and initial data. The SQL dump is shipped compressed — pipe it directly through zcat:

sudo zcat /usr/share/zabbix-sql-scripts/mysql/server.sql.gz | sudo mysql --default-character-set=utf8mb4 -uzabbix -p zabbix

Enter the zabbix user password you just created. The import takes 30–90 seconds depending on disk speed.

Disable the log_bin flag now that the import is done:

sudo mysql -uroot -p -e "SET GLOBAL log_bin_trust_function_creators = 0;"

Step 6: Configure Zabbix Server

Tell the Zabbix server daemon how to reach the database. Edit /etc/zabbix/zabbix_server.conf:

sudo nano /etc/zabbix/zabbix_server.conf

Find and set these lines (uncomment if needed):

DBHost=localhost
DBName=zabbix
DBUser=zabbix
DBPassword=ChangeThisStrongPassword123!
CacheSize=256M
HistoryCacheSize=128M
TrendCacheSize=32M
ValueCacheSize=128M
StartPollers=10
StartTrappers=5
StartPingers=5
LogFileSize=10

Save and exit. These defaults comfortably handle 100+ hosts on a 12 GB RAM server — tune them upward as your deployment grows (see Performance Tuning).

Step 7: Configure PHP and Nginx

The zabbix-nginx-conf package installs a server block at /etc/zabbix/nginx.conf. Edit it to set the listen port, server name, and PHP timezone.

sudo nano /etc/zabbix/nginx.conf

Uncomment and set:

listen          80;
server_name     zabbix.yourdomain.com;

Next, set the PHP timezone in /etc/zabbix/php-fpm.conf:

sudo nano /etc/zabbix/php-fpm.conf

Uncomment and set the timezone to match your region (full list: php.net/timezones):

php_value[date.timezone] = Europe/Berlin

The default Nginx config on Ubuntu 24.04 ships an enabled default server block on port 80 that will conflict with Zabbix. Remove it:

sudo rm -f /etc/nginx/sites-enabled/default

Test the Nginx config:

sudo nginx -t

Expected output:

nginx: the configuration file /etc/nginx/nginx.conf syntax is ok
nginx: configuration file /etc/nginx/nginx.conf test is successful

Step 8: Start Services and Complete the Web Installer

Enable and start all four daemons:

sudo systemctl restart zabbix-server zabbix-agent2 nginx php8.3-fpm
sudo systemctl enable zabbix-server zabbix-agent2 nginx php8.3-fpm

Confirm the Zabbix server is running:

sudo systemctl status zabbix-server --no-pager

Expected output:

● zabbix-server.service - Zabbix Server
     Loaded: loaded (/lib/systemd/system/zabbix-server.service; enabled; preset: enabled)
     Active: active (running) since Thu 2026-04-16 10:15:00 UTC; 5s ago

Open your browser to http://zabbix.yourdomain.com/ (or your server IP if DNS is not yet propagated). You should see the Zabbix welcome wizard.

Click through the installer:

Step 9: First Login and Initial Hardening

Log in with the default credentials:

• Username: Admin
• Password: zabbix

While you are in the admin area, also:

• Set a global URL under Administration > General > GUI: https://zabbix.yourdomain.com (we will add TLS in Step 14).
• Configure SMTP under Alerts > Media types > Email — fill in your SMTP host, port, credentials, and a From address.

Step 10: Install and Register zabbix-agent2 on Client Hosts

On each host you want to monitor, install the agent. Add the Zabbix repo just like you did on the server (Step 3), then install the agent:

sudo apt install -y zabbix-agent2

Edit /etc/zabbix/zabbix_agent2.conf:

sudo nano /etc/zabbix/zabbix_agent2.conf

Set:

Server=1.2.3.4              # IP of your Zabbix server
ServerActive=1.2.3.4        # Same IP, for active checks
Hostname=web01.example.com  # Unique hostname for this monitored host
HostMetadata=linux-webserver

Restart and enable:

sudo systemctl restart zabbix-agent2
sudo systemctl enable zabbix-agent2

Back in the Zabbix UI, go to Data collection > Hosts > Create host, enter the hostname exactly as in the agent config, add the Linux servers by Zabbix agent template, and save. Within 60 seconds the dashboard should show green for agent availability.

Step 11: Auto-Registration and Discovery Rules

Registering dozens or hundreds of hosts by hand is not scalable. Zabbix has two complementary mechanisms for this: active-agent auto-registration (agent calls home, server creates the host) and network discovery (server scans an IP range and creates hosts for responders).

Active Agent Auto-Registration

In the UI, go to Alerts > Actions > Autoregistration actions > Create action.

• Name: Auto-register Linux hosts
• Conditions: Host metadata contains linux-webserver
• Operations:

Now any new host you install with HostMetadata=linux-webserver in its agent config will auto-register, be added to the right group, and inherit the right template the first time it phones home. No manual UI clicks needed.

Network Discovery

For agentless devices (switches, printers, IPMI), use Data collection > Discovery > Create discovery rule:

• Name: Office LAN sweep
• IP range: 192.168.1.1-254
• Checks: ICMP ping, SNMPv2 agent sysName {.1.3.6.1.2.1.1.5.0}
• Update interval: 1h

Step 12: Triggers, Actions, and Notifications

Zabbix separates what is wrong (triggers) from what to do about it (actions).

Triggers

Triggers are boolean expressions over collected items. Most come pre-built in templates, but creating custom ones is straightforward. To alert when a host's 5-minute load average exceeds 4 on a 4-core box, go to Configuration > Hosts > [your host] > Triggers > Create trigger:

• Name: High CPU load on {HOST.NAME}
• Severity: High
• Expression: last(/web01/system.cpu.load[percpu,avg5]) > 1.0

Actions

Actions wire triggers to media types. Go to Alerts > Actions > Trigger actions > Create action:

• Name: Notify ops on any high+ severity
• Conditions: Trigger severity is greater than or equals High
• Operations: Send message to user group: Ops Team via Email and Slack
• Recovery operations: Send recovery message

Slack Notifications

Zabbix 7.0 ships a built-in Slack webhook media type. Go to Alerts > Media types > Slack > Update, paste a Slack incoming webhook URL (create one at api.slack.com/apps), and assign the media type to users via Users > Users > [your user] > Media.

Within seconds of saving, any high-severity trigger fires a rich Slack message with the host name, trigger, severity, and a direct link back to the Zabbix UI.

Email Notifications

Under Alerts > Media types > Email, set your SMTP server, credentials, and a From address. Assign this media type to users the same way you did for Slack. For Gmail/Google Workspace, use an app password — not your account password — and smtp.gmail.com:587 with TLS.

Step 13: Connect Grafana with the Zabbix Plugin

Zabbix has a solid built-in UI, but many teams prefer Grafana for polished dashboards and cross-datasource correlation. The Alexander Zobnin Zabbix plugin turns Grafana into a first-class Zabbix frontend.

If you do not yet have Grafana installed, follow our companion guide: how-to-install-grafana-ubuntu. Then install the Zabbix plugin:

sudo grafana-cli plugins install alexanderzobnin-zabbix-app
sudo systemctl restart grafana-server

In Grafana, go to Connections > Add new connection > Zabbix > Enable, then Add data source:

• URL: https://zabbix.yourdomain.com/api_jsonrpc.php
• Username: a dedicated read-only Zabbix user (create one in the Zabbix UI with role User)
• Password: the user's password
• Trends: Enabled

Step 14: Secure the Frontend with Let's Encrypt TLS

Running Zabbix over plain HTTP is fine for an internal network but unacceptable on the public internet. Certbot can provision and auto-renew a free TLS certificate from Let's Encrypt.

Install Certbot:

sudo apt install -y certbot python3-certbot-nginx

Make sure your firewall allows HTTP and HTTPS:

sudo ufw allow 80/tcp
sudo ufw allow 443/tcp
sudo ufw allow 10051/tcp
sudo ufw enable

Run Certbot against the Zabbix Nginx config:

sudo certbot --nginx -d zabbix.yourdomain.com

When prompted, enter your email, accept the ToS, and choose option 2 (Redirect) so that plain HTTP requests are automatically upgraded to HTTPS. Certbot rewrites the Nginx config in place, reloads the service, and registers a systemd timer that renews the cert automatically every 60 days.

Verify:

sudo certbot renew --dry-run

Expected output:

Congratulations, all simulated renewals succeeded.

Visit https://zabbix.yourdomain.com/ — the padlock should be green, and the TLS grade on ssllabs.com/ssltest should come back as A or A+ on the default Certbot config.

Performance Tuning

Once you cross 50 monitored hosts or 100,000 items, a few tweaks pay big dividends.

MariaDB

Edit /etc/mysql/mariadb.conf.d/50-server.cnf and set:

innodb_buffer_pool_size = 4G
innodb_log_file_size = 512M
innodb_flush_log_at_trx_commit = 2
innodb_flush_method = O_DIRECT
innodb_file_per_table = 1
max_connections = 200

Restart MariaDB:

sudo systemctl restart mariadb

The single biggest win is making innodb_buffer_pool_size roughly 50% of your server RAM — Zabbix's workload is extremely read-heavy on recent rows.

Zabbix Server

In /etc/zabbix/zabbix_server.conf, scale these based on host count:

sudo systemctl restart zabbix-server

Housekeeping

The built-in housekeeper deletes old history rows. For servers with more than a million items, switch to partitioning via TimescaleDB (PostgreSQL) or enable HousekeepingFrequency=24 and MaxHousekeeperDelete=20000 in the server config to spread the work overnight.

Troubleshooting

Viewing Logs

The two most useful log files:

sudo tail -f /var/log/zabbix/zabbix_server.log
sudo tail -f /var/log/nginx/error.log

FAQ

Is Zabbix free for commercial use?

Yes. Zabbix is released under the GPLv2 licence, which permits commercial use, modification, and redistribution with no per-host fees. Zabbix LLC sells optional paid support, training, and certified templates, but none of these are required to run a fully featured production deployment. The 7.0 LTS release is supported through 2029 with free security and bugfix updates from the official repositories.

How does Zabbix compare to Prometheus?

Zabbix and Prometheus solve overlapping problems with different philosophies. Prometheus is pull-based (scrapes metrics endpoints on a schedule), excels at ephemeral cloud-native workloads (Kubernetes, containers), uses PromQL as a query language, and pairs with Alertmanager for routing. Zabbix is push- and pull-capable, ships with a rich web UI out of the box, stores configuration and history in a relational database, and covers use cases Prometheus does not (agentless SNMP, IPMI, log file parsing, synthetic web monitoring). Many teams run both: Prometheus for containers and microservices, Zabbix for VMs, network gear, and business-level checks.

Can Zabbix monitor logs like Graylog or Loki?

Zabbix has native log-file item types that tail files on agent hosts and trigger alerts on regex matches — excellent for targeted monitoring of a handful of log patterns. For full-text search, aggregation, and retention of millions of log events, pair Zabbix with a dedicated log platform like Graylog. The two tools complement each other: Zabbix handles metrics and availability, Graylog handles log search and audit trails.

How many hosts can a single Zabbix server handle?

On the CloudCore Professional plan (6 vCPU, 12 GB RAM, 100 GB NVMe), a well-tuned Zabbix 7.0 server comfortably monitors 150–200 hosts with full history retention for 90 days. The practical ceiling on a single server is roughly 1,000 hosts and 500 new values per second, after which you should introduce Zabbix proxies to distribute collection, or move the database to TimescaleDB for better long-term scalability.

Does Zabbix replace a SIEM like Wazuh?

No — the tools overlap only slightly. Zabbix tracks infrastructure health (CPU, disk, services, network latency) and triggers on thresholds. Wazuh is a SIEM/XDR that focuses on security events: file integrity monitoring, rootkit detection, authentication log analysis, compliance reporting, and threat intelligence feeds. For a complete stack, run Zabbix for operational monitoring and Wazuh for security monitoring — they coexist happily on the same hosts with different agents.

How do I back up Zabbix configuration and data?

Two things need backing up: the MariaDB database (which contains both configuration and historical metrics) and the /etc/zabbix/ directory (config files and custom scripts). A simple nightly cron job does the trick:

mysqldump --single-transaction -uzabbix -p'password' zabbix | gzip > /backups/zabbix-$(date +%F).sql.gz
tar czf /backups/zabbix-conf-$(date +%F).tar.gz /etc/zabbix/

Ship the resulting files off-server (to S3, Backblaze B2, or another VPS) and you have a recoverable deployment.

Can I upgrade from Zabbix 6.0 LTS to 7.0 LTS in place?

Yes. Stop the server, update the zabbix-release package to the 7.0 variant, run apt update && apt upgrade, and restart the server. The daemon detects the old schema on first start and runs in-place migrations automatically (this can take a few minutes on larger databases). Always take a full database dump before upgrading, and read the official 7.0 upgrade notes for version-specific caveats.

Next Steps

Your Zabbix server is now live, secured, and collecting metrics. Recommended follow-ups:

• Install Grafana for richer dashboards — Follow our Grafana install guide and connect it to Zabbix via the plugin installed in Step 13.

• Add Prometheus for container metrics — If you are running Docker or Kubernetes, deploy Prometheus alongside Zabbix and use node-exporter + cadvisor for per-container visibility.

• Centralize logs with Graylog — Pipe all your systemd, Nginx, and application logs into Graylog for unified search, alerting, and retention.

• Harden with Wazuh SIEM — Install Wazuh agents alongside your Zabbix agents to add security event monitoring, FIM, and compliance reporting.

• Build a status page — Expose a public-facing status page by scraping Zabbix problems via the api_jsonrpc.php endpoint into a lightweight tool like Cachet or Statusfy.

• Read the official docs — The Zabbix documentation is extensive and kept current. The template reference and API reference are particularly worth bookmarking.

Run Zabbix on CloudCore Professional

Our Professional VPS plan is the sweet spot for self-hosted monitoring: enough RAM for the InnoDB buffer pool, enough NVMe for months of metric history, and enough vCPU to run Zabbix, MariaDB, Nginx, and Grafana side by side.

- 6 vCPU cores

- 12 GB RAM

- 100 GB NVMe SSD

- Unmetered bandwidth

- Deploys in under 60 seconds

Launch Your Monitoring VPS — Plans start at EUR 19.99/month.