act_runner/docker-compose.yml

services:
  # Split DNS for the dind daemon and its job containers: ts.net -> MagicDNS,
  # everything else -> Cloudflare. See Corefile for why MagicDNS alone fails.
  dns:
    image: coredns/coredns@sha256:af8c8d35a5d184b386c4a6d1a012c8b218d40d1376474c7d071bb6c07201f47d
    container_name: gitea-runner-dns
    restart: unless-stopped
    command: ["-conf", "/Corefile"]
    volumes:
      - ./Corefile:/Corefile:ro
    networks:
      runner:
        ipv4_address: 172.28.0.53

  # Isolated Docker daemon for the runner's job containers.
  # Keeps the host Docker clean and gives reliable cache networking.
  dind:
    image: docker:29-dind
    container_name: gitea-runner-dind
    privileged: true
    restart: unless-stopped
    # The third --host exposes the daemon over a unix socket on a volume shared
    # with the runner (dind-sock), *in addition to* its own /var/run/docker.sock.
    # act_runner's job-container DooD wiring only understands unix:// (or bare)
    # socket paths -- it bind-mounts that path from ITS OWN filesystem into every
    # job container as /var/run/docker.sock, it does not pass DOCKER_HOST through
    # as an env var. A bare "tcp://..." isn't recognized and silently falls back
    # to /var/run/docker.sock, which doesn't exist in the runner container -- so
    # this shared socket is what makes `docker build` etc. work inside jobs.
    # Two DNS settings, for two different resolvers, and both are load-bearing.
    #
    #   dns: (below)   the dind CONTAINER's resolv.conf, which is what dockerd
    #                  itself uses to pull images. Without an explicit list it
    #                  inherits Docker's embedded resolver, which forwards to
    #                  the host's systemd-resolved; with Tailscale DNS enabled
    #                  there, public names come back SERVFAIL and every job
    #                  dies at "docker pull" of the runner image itself. This
    #                  is what took the runner down on 2026-09-05.
    #
    #   --dns (here)   what dockerd hands to the JOB containers it creates.
    #                  The tailnet resolver first, so NOMAD_ADDR and the
    #                  staging Postgres resolve inside a job (they are
    #                  MagicDNS names); a public resolver second, because
    #                  100.100.100.100 answers only for the tailnet and
    #                  SERVFAILs everything else -- getaddrinfo falls through.
    #
    # Allocate 256 /24 subnets instead of Docker's large default /16 and /20
    # subnets. Jobs and their sibling test stacks share this daemon's pool.
    # The default bridge uses one slot. Interrupted cleanup can retain others;
    # a larger pool provides headroom but does not replace network cleanup.
    command: ["dockerd", "--host=tcp://0.0.0.0:2375", "--host=unix:///var/run/docker.sock", "--host=unix:///dind-sock/docker.sock", "--tls=false", "--dns=172.28.0.53", "--dns=1.1.1.1", "--default-address-pool", "base=10.210.0.0/16,size=24"]
    environment:
      DOCKER_TLS_CERTDIR: ""
    dns:
      - 172.28.0.53
      - 1.1.1.1
    depends_on:
      - dns
    networks:
      - runner
    volumes:
      # Persist pulled job images + docker layer cache => fast Rust rebuilds.
      - dind-storage:/var/lib/docker
      # Shared with the runner container; see the --host comment above.
      - dind-sock:/dind-sock
    healthcheck:
      test: ["CMD", "docker", "info"]
      interval: 5s
      timeout: 3s
      retries: 10
      start_period: 10s

  runner:
    # Ghim theo digest, KHÔNG dùng :nightly. Tag tự cập nhật + cache nằm trong
    # writable layer = một lệnh 'compose pull' là mất sạch cache.
    # Nâng cấp có chủ đích: đổi digest ở đây rồi commit.
    image: docker.io/gitea/act_runner@sha256:7940221bcfc9df16f3401ddafb2231bf98bba5517bc381224c7a2c504b558974
    container_name: gitea-runner-wsl
    restart: unless-stopped
    # Longer than runner.shutdown_timeout so cancellation can clean up too.
    stop_grace_period: 6m
    depends_on:
      dind:
        condition: service_healthy
    networks:
      - runner
    environment:
      GITEA_INSTANCE_URL: ${GITEA_INSTANCE_URL}
      GITEA_RUNNER_REGISTRATION_TOKEN: ${GITEA_RUNNER_REGISTRATION_TOKEN}
      GITEA_RUNNER_NAME: ${GITEA_RUNNER_NAME}
      GITEA_RUNNER_LABELS: ""
      # Point the runner at the DinD daemon. Overridden at startup by act_runner
      # itself to config.yaml's container.docker_host (the unix socket below),
      # since our labels require docker; kept here as a documented fallback.
      DOCKER_HOST: tcp://dind:2375
      CONFIG_FILE: /config.yaml
    volumes:
      - ./config.yaml:/config.yaml:ro
      # Persist runner registration so it survives restarts.
      - runner-data:/data
      # Shared with dind; see the --host comment on the dind service.
      - dind-sock:/dind-sock
      # Cache của actions/cache. Mặc định act_runner ghi vào
      # $HOME/.cache/actcache = WRITABLE LAYER của container -> mất sạch mỗi
      # lần container bị tạo lại. Phải khớp với cache.dir trong config.yaml.
      - runner-cache:/cache

networks:
  runner:
    ipam:
      config:
        - subnet: 172.28.0.0/24

volumes:
  dind-storage:
  runner-data:
  dind-sock:
  runner-cache: