ConnectOS Docker

ConnectOS ships with Docker and Docker Compose, so the gateway can run containerized applications alongside its radio workloads, a Node-RED flow, a time-series database, a dashboard, or your own application image. Everything is managed from Services → Docker, without needing a shell on the device.

The Docker pages are organized as stacks: each stack is a directory holding a docker-compose.yml and an optional .env file, managed as a unit.

This guide deploys Node-RED from the bundled templates.

Before You Start

The gateway needs outbound internet access to pull images from Docker Hub.
Check you have room on the filesystem. Container images are large, the Node-RED image alone is around 670 MB.

The Docker Pages

Tab	Contents
Stacks	All Compose stacks, their status, and start/stop/restart controls
Containers	Every container on the device, running or not
Images	Images pulled to local storage
Volumes	Docker volumes, where container data persists
Networks	Docker networks
Settings	Docker daemon state and the Compose template library

Check the Daemon

Open Services → Docker → Settings. The Docker daemon toggle must read Running; the state is preserved across reboots.

This page also manages the template library, a set of ready-made Compose files that ConnectOS can download from a ZIP archive. The stock library includes grafana, influxdb1, influxdb2, nodered and victoria-metrics. Update re-downloads the library from the configured URL and replaces what is installed.

Figure 1: Docker Settings and Template Library

Create a Stack

Go to the Stacks tab and press New Stack.

Figure 1: Docker Stacks

In the dialog, pick a Source, a template, in this case nodered, and give the stack a name. Names may use alphanumeric characters, hyphens and underscores, up to 64 characters. Press Create.

Figure 1: New Stack Dialog

The stack is created on disk but nothing is running yet: it is listed as not_deployed.

Review the Compose File

Click the stack name to open it. The Stack Files section is an editor over the stack's files, select docker-compose.yml or .env from the dropdown, edit, and press Save.

Figure 1: Docker Stack Details

The Node-RED template looks like this:

services:

  nodered:
    image: nodered/node-red:latest
    restart: unless-stopped
    ports:
      - "${PORT:-1880}:1880"
    volumes:
      - node-red-data:/data
    environment:
      - TZ=${TZ:-Europe/London}
      - NODE_RED_ENABLE_PROJECTS=${NODE_RED_ENABLE_PROJECTS:-false}
    extra_hosts:
      - host.docker.internal=host-gateway

volumes:
  node-red-data:

A few things worth noting, because they are conventions the bundled templates follow:

${PORT:-1880} The published port comes from a variable with a default. Set PORT in the .env file to move the service to another port without touching the Compose file.
node-red-data:/data Flow data lives in a named volume, so it survives the container being recreated or the image being updated.
host.docker.internal Lets the container reach services running on the gateway itself, for example the built-in MQTT broker or ChirpStack LNS.

NOTE

restart: unless-stopped makes the container come back automatically after a gateway reboot. Keep it for anything you want running permanently.

Deploy

Press Deploy. This pulls the images and starts the stack, which is the slow step, a first pull over a typical connection takes several minutes. When it finishes, the badge at the top right turns running and the Containers section lists the container with its published ports.

Figure 1: Deployed Docker Stack

Back on the Stacks tab, the stack now reports running with its service count and published ports.

Figure 1: Running Docker Stacks

The action buttons above the editor are:

Button	Effect
Deploy	Pull images if needed and bring the stack up
Start	Start a stack that was created or stopped
Stop	Stop the containers, keeping them and their volumes
Restart	Stop and start again
Pull Images	Fetch newer images without changing the running containers
Destroy	Remove the stack's containers

To update a stack to a newer image: Pull Images, then Deploy to recreate the containers on the new image.

Read the Logs

The Logs section at the bottom of the stack page streams container output. Choose a service (or All services), press Show Logs, and leave Follow ticked to keep the output live. This is the first place to look when a container starts and immediately exits.

Figure 1: Docker Stack Logs

Use the Application

The Node-RED template publishes port 1880, so the editor is reachable at:

http://<gateway-ip>:1880

NOTE

Published ports are reachable from whichever firewall zone allows them. If you cannot reach the service from your network, check Network → Firewall, a container publishing a port does not by itself open the gateway's firewall to the outside.

Containers, Images, Volumes and Networks

The remaining tabs show Docker's state across all stacks.

Containers lists everything on the device with its image, status and ports.

Figure 1: Docker Containers

Images shows what has been pulled and how much space each image occupies, useful when reclaiming storage.

Figure 1: Docker Images

Volumes lists the named volumes holding persistent data. Destroying a stack does not delete its volumes, so data survives a redeploy, and equally, removing a volume here is irreversible.