How to Build a Full‑Featured Zabbix Monitoring Platform with Docker Compose

Core Architecture

Zabbix follows a client/server (C/S) model with four main components:

Zabbix Server : receives data, processes triggers, stores history, and sends alerts.

Zabbix Agent : installed on monitored hosts; supports active and passive modes.

Zabbix Proxy : optional distributed collector that forwards data to the server (useful for multi‑site deployments).

SNMP devices : monitored via SNMP without agents.

One‑Click Deployment with Docker Compose

Create a docker-compose.yml file that defines MySQL, Zabbix Server, Web UI, and Agent services. Example:

version: '3.8'
services:
  mysql:
    image: mysql:8.0
    container_name: zabbix-mysql
    command:
      - mysqld
      - --character-set-server=utf8mb4
      - --collation-server=utf8mb4_bin
      - --default-authentication-plugin=mysql_native_password
    environment:
      MYSQL_DATABASE: zabbix
      MYSQL_USER: zabbix
      MYSQL_PASSWORD: zabbix_password
      MYSQL_ROOT_PASSWORD: root_password
    volumes:
      - zabbix_mysql_data:/var/lib/mysql
    networks:
      - zabbix-net
    restart: unless-stopped

  zabbix-server:
    image: zabbix/zabbix-server-mysql:ubuntu-7.0-latest
    container_name: zabbix-server
    environment:
      DB_SERVER_HOST: mysql
      MYSQL_DATABASE: zabbix
      MYSQL_USER: zabbix
      MYSQL_PASSWORD: zabbix_password
      MYSQL_ROOT_PASSWORD: root_password
      ZBX_CACHESIZE: 128M
      ZBX_STARTPOLLERS: 10
      ZBX_STARTPINGERS: 5
    ports:
      - "10051:10051"
    volumes:
      - zabbix_server_alertscripts:/usr/lib/zabbix/alertscripts
      - zabbix_server_externalscripts:/usr/lib/zabbix/externalscripts
    networks:
      - zabbix-net
    depends_on:
      - mysql
    restart: unless-stopped

  zabbix-web:
    image: zabbix/zabbix-web-nginx-mysql:ubuntu-7.0-latest
    container_name: zabbix-web
    environment:
      ZBX_SERVER_HOST: zabbix-server
      DB_SERVER_HOST: mysql
      MYSQL_DATABASE: zabbix
      MYSQL_USER: zabbix
      MYSQL_PASSWORD: zabbix_password
      PHP_TZ: Asia/Shanghai
    ports:
      - "8080:8080"
    networks:
      - zabbix-net
    depends_on:
      - mysql
      - zabbix-server
    restart: unless-stopped

  zabbix-agent:
    image: zabbix/zabbix-agent:ubuntu-7.0-latest
    container_name: zabbix-agent
    environment:
      ZBX_HOSTNAME: "Zabbix server"
      ZBX_SERVER_HOST: zabbix-server
    ports:
      - "10050:10050"
    networks:
      - zabbix-net
    depends_on:
      - zabbix-server
    restart: unless-stopped

volumes:
  zabbix_mysql_data:
  zabbix_server_alertscripts:
  zabbix_server_externalscripts:

networks:
  zabbix-net:
    driver: bridge

Start all services:

# Run in background
docker-compose up -d
# View server logs
docker-compose logs -f zabbix-server

Access the web UI at http://SERVER_IP:8080 with default credentials Admin / zabbix. Change the password after first login.

Agent Deployment and Configuration

Linux

Install the official Zabbix repository and the latest Agent2 package:

# Ubuntu/Debian
wget https://repo.zabbix.com/zabbix/7.0/ubuntu/pool/main/z/zabbix-release/zabbix-release_7.0-1+ubuntu$(lsb_release -rs)_all.deb
dpkg -i zabbix-release_7.0-1+ubuntu$(lsb_release -rs)_all.deb
apt update
apt install zabbix-agent2 -y

# CentOS/RHEL
rpm -Uvh https://repo.zabbix.com/zabbix/7.0/rhel/8/x86_64/zabbix-release-7.0-1.el8.noarch.rpm
dnf clean all && dnf makecache
dnf install zabbix-agent2 -y

Edit /etc/zabbix/zabbix_agent2.conf (example values):

# Server IP(s)
Server=192.168.1.100
# Active server for active checks
ServerActive=192.168.1.100
# Hostname must match the name used in Zabbix UI
Hostname=web-server-01
# Debug level (1=error, 3=debug)
DebugLevel=3
# Allow custom user parameters
UnsafeUserParameters=1

Enable and start the agent:

systemctl enable zabbix-agent2 --now
systemctl status zabbix-agent2

Windows

Download the MSI installer from the Zabbix download page, run the installer, set the Server IP and Hostname, then start the Zabbix Agent 2 service. Configuration file is C:\Program Files\Zabbix Agent 2\zabbix_agent2.conf with the same parameters as Linux.

Monitoring Configuration in the Web UI

Add hosts, assign interfaces, link templates (e.g., "Linux by Zabbix agent"), and configure host groups. Use auto‑discovery rules for bulk resources such as disk partitions ( vfs.fs.discovery) or Docker containers (custom script docker_discovery.py).

Auto‑Discovery Example – Disk Partitions

Create a discovery rule with key vfs.fs.discovery and filter {#FSTYPE}=ext4|xfs|btrfs to automatically add items for usage and free space.

Auto‑Discovery Example – Docker Containers

#!/usr/bin/env python3
import subprocess, json

def discover_containers():
    result = subprocess.run(['docker','ps','--format','{{.Names}}'], capture_output=True, text=True)
    containers = result.stdout.strip().split('
')
    data = [{'{#CONTAINER}': c} for c in containers if c]
    return json.dumps({'data': data})

if __name__ == '__main__':
    print(discover_containers())

Make the script executable ( chmod +x /usr/lib/zabbix/externalscripts/docker_discovery.py) and use key docker.containers.discovery in a discovery rule.

Alerting Configuration

Email Alerts

Install mailutils on the server and create a script /usr/lib/zabbix/alertscripts/send_email.sh:

#!/bin/bash
TO="$1"
SUBJECT="$2"
BODY="$3"

echo "$BODY" | mail -s "$SUBJECT" "$TO"

Make it executable ( chmod +x /usr/lib/zabbix/alertscripts/send_email.sh) and create a media type of type Script named "Email" that calls the script with parameters {ALERT.SENDTO}, {ALERT.SUBJECT}, {ALERT.MESSAGE}. Assign the media to users.

WeChat (Enterprise WeChat) Alerts

Create /usr/lib/zabbix/alertscripts/wechat.py:

#!/usr/bin/env python3
import sys, requests, json

CORPID='your_corpid'
SECRET='your_secret'
AGENTID=1000002

def get_token():
    url=f'https://qyapi.weixin.qq.com/cgi-bin/gettoken?corpid={CORPID}&corpsecret={SECRET}'
    return requests.get(url).json()['access_token']

def send_message(user, subject, message):
    token=get_token()
    url=f'https://qyapi.weixin.qq.com/cgi-bin/message/send?access_token={token}'
    data={
        "touser": user,
        "msgtype": "text",
        "agentid": AGENTID,
        "text": {"content": f"【{subject}】
{message}"}
    }
    return requests.post(url, json=data).json()

if __name__=='__main__':
    user=sys.argv[1]
    subject=sys.argv[2]
    message=sys.argv[3]
    print(send_message(user, subject, message))

Register a script media type "WeChat" that calls wechat.py with the same three parameters.

Escalation (Alarm Upgrade)

Create an action with multiple steps: Step 1 (0‑30 min) notifies on‑call staff, Step 2 (30‑60 min) notifies a senior engineer, Step 3 (after 60 min) notifies the technical director and triggers a phone call.

Distributed Monitoring with Zabbix Proxy

Deploy a proxy using Docker Compose ( docker-compose-proxy.yml) and configure it to forward data to the central server:

version: '3.8'
services:
  zabbix-proxy:
    image: zabbix/zabbix-proxy-sqlite3:ubuntu-7.0-latest
    container_name: zabbix-proxy
    environment:
      ZBX_PROXYMODE: 0   # 0 = active mode (recommended)
      ZBX_HOSTNAME: proxy-beijing
      ZBX_SERVER_HOST: 192.168.1.100   # central server IP
      ZBX_CONFIGFREQUENCY: 300
    ports:
      - "10051:10051"
    volumes:
      - zabbix_proxy_data:/var/lib/zabbix
    restart: unless-stopped

volumes:
  zabbix_proxy_data:

Start it with docker-compose -f docker-compose-proxy.yml up -d. The proxy collects data from local agents and synchronizes with the server, enabling multi‑site monitoring without exposing the server directly.

Performance Tuning

Server Settings

# Example /etc/zabbix/zabbix_server.conf adjustments
CacheSize=256M
HistoryCacheSize=64M
TrendCacheSize=16M
StartPollers=20
StartPollersUnreachable=5
StartPingers=10
StartDiscoverers=5

Database Optimizations

Use partitioned tables for history and history_uint (daily or monthly).

Delete data older than 90 days and run OPTIMIZE TABLE regularly.

Prefer Unix socket connections when DB and server share a host.

Item Interval Recommendations

Critical metrics (CPU, memory): 30 s – 1 min.

Normal metrics (disk usage): 5 min.

Trend metrics (traffic): 30 min.

Keep per‑host items under 200.

Common Monitoring Templates

Import ready‑made templates from the Zabbix Template Marketplace for MySQL, Redis, Nginx, AWS, Azure, etc., or add custom UserParameter definitions such as:

# MySQL
UserParameter=mysql.status[*],mysqladmin -umonitor -ppassword extended-status | awk '/$1/{print $$4}'
UserParameter=mysql.ping,mysqladmin -umonitor -ppassword ping | grep -c alive

# Redis
UserParameter=redis.info[*],redis-cli -h localhost -p 6379 INFO $1 | grep -w $2 | cut -d: -f2
UserParameter=redis.ping,redis-cli -h localhost ping | grep -c PONG

# Nginx (requires status page)
UserParameter=nginx.active,curl -s http://127.0.0.1/nginx_status | awk '/Active/{print $3}'
UserParameter=nginx.accepts,curl -s http://127.0.0.1/nginx_status | awk 'NR==3{print $1}'

Dashboard Customization

Create a dashboard with widgets such as "Problems by severity", "Top hosts by CPU/Memory", trend graphs, geo‑map of hosts, and service status. Add SLA reports for critical services using Zabbix's SLA management.

Operational Recommendations

Start with a small set of critical hosts and expand gradually.

Set meaningful trigger thresholds to avoid alert fatigue; aim for <5 min response for high‑severity alerts.

Review and clean unused items quarterly.

Back up the Zabbix database and configuration files regularly.

Leverage official documentation and the Template Marketplace for quick, reliable integrations.