docker-compose.yml

# Usage:
#     docker-compose run archivebox init --setup
#     docker-compose up
#     echo "https://example.com" | docker-compose run archivebox archivebox add
#     docker-compose run archivebox add --depth=1 https://example.com/some/feed.rss
#     docker-compose run archivebox config --set PUBLIC_INDEX=True
#     docker-compose run archivebox help
# Documentation:
#     https://github.com/ArchiveBox/ArchiveBox/wiki/Docker#docker-compose

version: '3.9'   # '3.9' or greater also works

services:
    archivebox:
        build: .                              # for developers working on archivebox
        image: ${DOCKER_IMAGE:-matiaszanolli/archivebox-redux:latest}
        restart: always
        # command: tail -f /dev/null
        command: server --quick-init 0.0.0.0:8000
        # command: gunicorn archivebox.wsgi:application --bind 0.0.0.0:8000
        ports:
            - 8000:8000
        environment:
            - ALLOWED_HOSTS=*                   # add any config options you want as env vars
            - MEDIA_MAX_SIZE=750m
            - PG_HOST=host.docker.internal
            - PG_DATABASE=archivebox
            - PG_USERNAME=archivebox
            - PG_PASSWORD=archivebox
            - DEBUG=1
            - SEARCH_BACKEND_ENGINE=sonic     # uncomment these if you enable sonic below
            - SEARCH_BACKEND_HOST_NAME=sonic
            - SEARCH_BACKEND_PASSWORD=SecretPassword
        # dns:                                  # uncomment to use pihole below for ad/tracker blocking during archiving
        #     - pihole
        # depends_on:
        #     - cache
        volumes:
            - /mnt/archivebox-new:/data
            - ./archivebox:/archivebox    # for developers working on archivebox
            - /home/archivebox/cache:/var/lib/cache
        extra_hosts:
            - "host.docker.internal:host-gateway"
        deploy:
          resources:
            reservations:
              devices:
                - driver: nvidia
                  count: 1
                  capabilities: [gpu]
        networks:
            - default
    db:
        image: postgres:14.1-alpine
        restart: always
        environment:
            - POSTGRES_USER=archivebox
            - POSTGRES_PASSWORD=archivebox
        ports:
            - '5432:5432'
        volumes: 
            - /home/archivebox/db-data:/var/lib/postgresql/data  
        networks:
            - default 
        # Example: Put Nginx in front of the ArchiveBox server for SSL termination
    nginx:
        build: ./etc/nginx
        ports:
            - 80:8000
        depends_on:
            - archivebox
        networks:
            - default 
    sonic:
       image: valeriansaliou/sonic:v1.3.1
       expose:
           - 1491
       environment:
           - SEARCH_BACKEND_PASSWORD=SecretPassword
       volumes:
           - ./sonic.cfg:/etc/sonic.cfg:ro
           - ./data/sonic:/var/lib/sonic/store

    
networks:
    default:
        driver: bridge
        ipam:
            config:
                 - subnet: 172.16.57.0/24
    # netdata:
    #     image: netdata/netdata
    #     container_name: netdata
    #     hostname: 192.168.1.8 # set to fqdn of host
    #     ports:
    #         - 19999:19999
    #     restart: unless-stopped
    #     cap_add:
    #         - SYS_PTRACE
    #     security_opt:
    #         - apparmor:unconfined
    #     volumes:
    #         - /mnt/archivebox-new/netdataconfig:/etc/netdata
    #         - /mnt/archivebox-new/netdatalib:/var/lib/netdata
    #         - /mnt/archivebox-new/netdatacache:/var/cache/netdata
    #         - /etc/passwd:/host/etc/passwd:ro
    #         - /etc/group:/host/etc/group:ro
    #         - /proc:/host/proc:ro
    #         - /sys:/host/sys:ro
    #         - /etc/os-release:/host/etc/os-release:ro
    #     environment:
    #         - NETDATA_CLAIM_TOKEN=Junyu-WQSQJJ2-h3qMxmemxSCnr_yDJXHYuo6GwYgeEBDzifuzCyMJRSy2ByIyzTK4TRJSRkZUbNmiBlYY3_yESDQjXIIXvLHjuNF-jZ2GNeQ4kIhUXXT7qa1w0Rom-5zczptrQ
    #         - NETDATA_CLAIM_URL=https://app.netdata.cloud
    #         - NETDATA_CLAIM_ROOMS=2fe4a14a-af33-4ade-a693-3ba086912772
    #     networks:
    #         - default

# volumes:
#     cache:
#         driver: local
# cache:
    #     image: 'redis:latest'
    #     ports:
    #         - '6379:6379'
    #     command: redis-server --save 20 1 --loglevel warning
    #     volumes: 
    #         - cache:/data
    #     networks:
    #         - default
    # netdata:
    #     image: netdata/netdata
    #     container_name: netdata
    #     hostname: 192.168.1.8 # set to fqdn of host
    #     ports:
    #         - 19999:19999
    #     restart: unless-stopped
    #     cap_add:
    #         - SYS_PTRACE
    #     security_opt:
    #         - apparmor:unconfined
    #     volumes:
    #         - /mnt/archivebox-new/netdataconfig:/etc/netdata
    #         - /mnt/archivebox-new/netdatalib:/var/lib/netdata
    #         - /mnt/archivebox-new/netdatacache:/var/cache/netdata
    #         - /etc/passwd:/host/etc/passwd:ro
    #         - /etc/group:/host/etc/group:ro
    #         - /proc:/host/proc:ro
    #         - /sys:/host/sys:ro
    #         - /etc/os-release:/host/etc/os-release:ro
    #     environment:
    #         - NETDATA_CLAIM_TOKEN=Junyu-WQSQJJ2-h3qMxmemxSCnr_yDJXHYuo6GwYgeEBDzifuzCyMJRSy2ByIyzTK4TRJSRkZUbNmiBlYY3_yESDQjXIIXvLHjuNF-jZ2GNeQ4kIhUXXT7qa1w0Rom-5zczptrQ
    #         - NETDATA_CLAIM_URL=https://app.netdata.cloud
    #         - NETDATA_CLAIM_ROOMS=2fe4a14a-af33-4ade-a693-3ba086912772
    #     networks:
    #         - default
    
    ### Example: To run pihole in order to block ad/tracker requests during archiving,
    # uncomment this block and set up pihole using its admin interface

    # pihole:
    #   image: pihole/pihole:latest
    #   ports:
    #     - 80:80       # uncomment to access the admin HTTP interface on http://localhost:80
    #   environment:
    #     WEBPASSWORD: 'set a secure password here or it will be random'
    #   volumes:
    #     - ./data/pihole:/etc/pihole
    #     - ./data/dnsmasq:/etc/dnsmasq.d

    # To run the Sonic full-text search backend, first download the config file to sonic.cfg
    # curl -O https://raw.githubusercontent.com/ArchiveBox/ArchiveBox/master/etc/sonic.cfg
    # after starting, backfill any existing Snapshots into the index: docker-compose run archivebox update --index-only
    # Example: Run scheduled imports in a docker instead of using cron on the
    # host machine, add tasks and see more info with archivebox schedule --help
#    scheduler:
#        image: matiaszanolli/archivebox:latest
#        command: schedule --foreground --every=day --depth=1 'https://getpocket.com/users/USERNAME/feed/all'
#        environment:
#            - USE_COLOR=True
#            - SHOW_PROGRESS=False
#        volumes:
#            - ./data:/data

    ### Example: run all your ArchiveBox traffic through a WireGuard VPN tunnel

    # wireguard:
    #   image: linuxserver/wireguard
    #   network_mode: 'service:archivebox'
    #   cap_add:
    #     - NET_ADMIN
    #     - SYS_MODULE
    #   sysctls:
    #     - net.ipv4.conf.all.rp_filter=2
    #     - net.ipv4.conf.all.src_valid_mark=1
    #   volumes:
    #     - /lib/modules:/lib/modules
    #     - ./wireguard.conf:/config/wg0.conf:ro


    ### Example: Run PYWB in parallel and auto-import WARCs from ArchiveBox

    # pywb:
    #     image: webrecorder/pywb:latest
    #     entrypoint: /bin/sh -c '(wb-manager init default || test $$? -eq 2) && wb-manager add default /archivebox/archive/*/warc/*.warc.gz; wayback;'
    #     environment:
    #         - INIT_COLLECTION=archivebox
    #     ports:
    #         - 8080:8080
    #     volumes:
    #         - ./data:/archivebox
    #         - ./data/wayback:/webarchive