Modern Docker Management, Designed for Everyone
Go composecontainer-managementcontainersdockerdocker-composedocker-managementgoself-hostedsveltekittypescriptweb-ui
services:
# Docker Socket Proxy - see https://github.com/Tecnativa/docker-socket-proxy for official documentation
docker-socket-proxy:
image: tecnativa/docker-socket-proxy:latest
container_name: arcane-docker-proxy
environment:
- EVENTS=1
- PING=1
- VERSION=1
- POST=1
# Required by Arcane
- AUTH=1
- BUILD=1
- COMMIT=1
- CONTAINERS=1
- DISTRIBUTION=1
- EXEC=1
- GRPC=1
- IMAGES=1
- INFO=1
- NETWORKS=1
- SESSION=1
- SYSTEM=1
- VOLUMES=1
# Swarm management: set to 1 when the host is a Swarm manager
- CONFIGS=0
- NODES=0
- SECRETS=0
- SERVICES=0
- SWARM=0
- TASKS=0
# Not used by Arcane
- PLUGINS=0
volumes:
- /var/run/docker.sock:/var/run/docker.sock:ro # Read-only access to host socket
networks:
- arcane-internal
restart: unless-stopped
security_opt:
- no-new-privileges:true
arcane:
image: getarcaneapp/arcane:latest
container_name: arcane
ports:
- '3552:3552'
volumes:
- arcane-data:/app/data
environment:
- ENCRYPTION_KEY=replace_me_with_a_random_32_char_value
# Optional: run Arcane as a specific host UID/GID instead of the default runtime user.
# With DOCKER_HOST over TCP, no Unix socket group mapping is needed.
# - PUID=1000
# - PGID=1000
- DOCKER_HOST=tcp://docker-socket-proxy:2375 # Use the proxy instead of direct socket
networks:
- arcane-internal
depends_on:
- docker-socket-proxy
healthcheck:
test: ['CMD', './arcane', 'health', '--timeout', '2s']
interval: 10s
timeout: 3s
retries: 5
start_period: 15s
# Use host cgroup namespace for better container self-detection
# This allows Arcane to detect its own container ID more reliably
cgroup: host
restart: unless-stopped
stop_grace_period: 15s
networks:
arcane-internal:
driver: bridge
name: arcane-internal
volumes:
arcane-data:
name: arcane-data