Files
NetAlertX/docs/DOCKER_SWARM.md
2025-11-09 17:03:25 +00:00

80 lines
2.2 KiB
Markdown
Executable File

# Docker Swarm Deployment Guide (IPvlan)
This guide describes how to deploy **NetAlertX** in a **Docker Swarm** environment using an `ipvlan` network. This enables the container to receive a LAN IP address directly, which is ideal for network monitoring.
---
## ⚙️ Step 1: Create an IPvlan Config-Only Network on All Nodes
> Run this command on **each node** in the Swarm.
```bash
docker network create -d ipvlan \
--subnet=192.168.1.0/24 \ # 🔧 Replace with your LAN subnet
--gateway=192.168.1.1 \ # 🔧 Replace with your LAN gateway
-o ipvlan_mode=l2 \
-o parent=eno1 \ # 🔧 Replace with your network interface (e.g., eth0, eno1)
--config-only \
ipvlan-swarm-config
```
---
## 🖥️ Step 2: Create the Swarm-Scoped IPvlan Network (One-Time Setup)
> Run this on **one Swarm manager node only**.
```bash
docker network create -d ipvlan \
--scope swarm \
--config-from ipvlan-swarm-config \
swarm-ipvlan
```
---
## 🧾 Step 3: Deploy NetAlertX with Docker Compose
Use the following Compose snippet to deploy NetAlertX with a **static LAN IP** assigned via the `swarm-ipvlan` network.
```yaml
services:
netalertx:
image: ghcr.io/jokob-sk/netalertx:latest
ports:
- 20211:20211
volumes:
- /mnt/YOUR_SERVER/netalertx/config:/data/config:rw
- /mnt/YOUR_SERVER/netalertx/db:/netalertx/data/db:rw
- /mnt/YOUR_SERVER/netalertx/logs:/netalertx/tmp/log:rw
environment:
- TZ=Europe/London
- PORT=20211
networks:
swarm-ipvlan:
ipv4_address: 192.168.1.240 # ⚠️ Choose a free IP from your LAN
deploy:
mode: replicated
replicas: 1
restart_policy:
condition: on-failure
placement:
constraints:
- node.role == manager # 🔄 Or use: node.labels.netalertx == true
networks:
swarm-ipvlan:
external: true
```
---
## ✅ Notes
* The `ipvlan` setup allows **NetAlertX** to have a direct IP on your LAN.
* Replace `eno1` with your interface, IP addresses, and volume paths to match your environment.
* Make sure the assigned IP (`192.168.1.240` above) is not in use or managed by DHCP.
* You may also use a node label constraint instead of `node.role == manager` for more control.