docker-compose.failover-rehearsal.yml

name: durable-workflow-single-region-failover

x-server-image: &server-image ${DW_FAILOVER_SERVER_IMAGE:?Run through scripts/conformance/single-region-failover-published-artifacts.sh}

x-server-environment: &server-environment
  APP_NAME: "Durable Workflow Server"
  APP_ENV: testing
  APP_VERSION: ${DW_FAILOVER_SERVER_VERSION:?The runner records the concrete server version}
  APP_DEBUG: "false"
  DW_SERVER_KEY: ${DW_SERVER_KEY:-base64:5Zt4nUhlCm3DD0nLXZJQdHiwPfb56yGo9gNV/g3jYbY=}
  DB_CONNECTION: mysql
  DB_HOST: mysql
  DB_PORT: 3306
  DB_DATABASE: ${DB_DATABASE:-durable_workflow}
  DB_USERNAME: ${DB_USERNAME:-workflow}
  DB_PASSWORD: ${DB_PASSWORD:-workflow}
  REDIS_HOST: redis
  REDIS_PORT: 6379
  QUEUE_CONNECTION: redis
  CACHE_STORE: redis
  DW_AUTH_DRIVER: token
  DW_AUTH_TOKEN: ${DW_AUTH_TOKEN:-failover-rehearsal-token}
  DW_AUTH_BACKWARD_COMPATIBLE: "true"
  DW_SERVER_TOPOLOGY_SHAPE: standalone_server
  DW_V2_MULTI_NODE: "true"
  DW_TASK_DISPATCH_MODE: poll
  DW_WORKER_POLL_TIMEOUT: "2"
  DW_WORKER_POLL_INTERVAL_MS: "200"
  DW_WORKFLOW_TASK_TIMEOUT: ${DW_FAILOVER_WORKFLOW_TASK_LEASE_SECONDS:-8}
  DW_V2_TASK_REPAIR_REDISPATCH_AFTER_SECONDS: "1"
  DW_V2_TASK_REPAIR_LOOP_THROTTLE_SECONDS: "1"
  DW_ENV_AUDIT_STRICT: "1"

services:
  bootstrap:
    image: *server-image
    command: ["server-bootstrap"]
    healthcheck:
      disable: true
    environment:
      <<: *server-environment
      DW_SERVER_ID: failover-bootstrap
    depends_on:
      mysql:
        condition: service_healthy
      redis:
        condition: service_healthy

  server-a:
    image: *server-image
    ports:
      - "${DW_FAILOVER_SERVER_A_PORT:-18084}:8080"
    environment:
      <<: *server-environment
      DW_SERVER_ID: failover-server-a
      DW_SERVER_PROCESS_CLASS: server_http_node
    depends_on:
      bootstrap:
        condition: service_completed_successfully
    healthcheck:
      test: ["CMD", "curl", "-f", "http://localhost:8080/api/ready"]
      interval: 2s
      timeout: 2s
      retries: 30

  server-b:
    image: *server-image
    ports:
      - "${DW_FAILOVER_SERVER_B_PORT:-18085}:8080"
    environment:
      <<: *server-environment
      DW_SERVER_ID: failover-server-b
      DW_SERVER_PROCESS_CLASS: server_http_node
    depends_on:
      bootstrap:
        condition: service_completed_successfully
    healthcheck:
      test: ["CMD", "curl", "-f", "http://localhost:8080/api/ready"]
      interval: 2s
      timeout: 2s
      retries: 30

  queue-worker:
    image: *server-image
    command: ["php", "artisan", "queue:work", "redis", "--sleep=1", "--tries=3", "--max-time=3600"]
    healthcheck:
      test: ["CMD", "server-process-healthcheck", "worker"]
      interval: 5s
      timeout: 3s
      retries: 3
      start_period: 5s
    restart: unless-stopped
    environment:
      <<: *server-environment
      DW_SERVER_ID: failover-queue-worker
      DW_SERVER_PROCESS_CLASS: worker_node
    depends_on:
      bootstrap:
        condition: service_completed_successfully
      redis:
        condition: service_healthy

  scheduler:
    image: *server-image
    init: true
    # Include an unscoped repair pass for remote SDK task queues.
    command: >-
      sh -c 'while true; do php artisan schedule:evaluate --limit=100 --json; if [ "$$DW_V2_MATCHING_ROLE_QUEUE_WAKE" != "false" ] && [ "$$DW_V2_MATCHING_ROLE_QUEUE_WAKE" != "0" ]; then php artisan workflow:v2:repair-pass --json; fi; php artisan activity:timeout-enforce --limit=100; php artisan history:prune --limit=100; sleep 1; done'
    healthcheck:
      test: ["CMD", "server-process-healthcheck", "scheduler"]
      interval: 5s
      timeout: 3s
      retries: 3
      start_period: 5s
    environment:
      <<: *server-environment
      DW_SERVER_ID: failover-scheduler
      DW_SERVER_PROCESS_CLASS: scheduler_node
    depends_on:
      bootstrap:
        condition: service_completed_successfully

  load-balancer:
    image: ${DW_FAILOVER_NGINX_IMAGE:?The runner resolves the load-balancer image}
    ports:
      - "${DW_FAILOVER_LB_PORT:-18086}:8080"
    command:
      - /bin/sh
      - -ec
      - |
        cat >/etc/nginx/conf.d/default.conf <<'NGINX'
        upstream durable_workflow_api {
          zone durable_workflow_api 64k;
          server server-a:8080 max_fails=1 fail_timeout=2s;
          server server-b:8080 max_fails=1 fail_timeout=2s;
        }

        server {
          listen 8080;

          location / {
            proxy_http_version 1.1;
            proxy_connect_timeout 1s;
            proxy_read_timeout 65s;
            proxy_next_upstream error timeout http_502 http_503 http_504;
            proxy_next_upstream_tries 2;
            proxy_set_header Host $$host;
            proxy_set_header X-Forwarded-For $$proxy_add_x_forwarded_for;
            proxy_pass http://durable_workflow_api;
          }
        }
        NGINX
        exec nginx -g 'daemon off;'
    depends_on:
      server-a:
        condition: service_healthy
      server-b:
        condition: service_healthy

  mysql:
    image: ${DW_FAILOVER_MYSQL_IMAGE:?The runner resolves the database image}
    environment:
      MYSQL_DATABASE: ${DB_DATABASE:-durable_workflow}
      MYSQL_USER: ${DB_USERNAME:-workflow}
      MYSQL_PASSWORD: ${DB_PASSWORD:-workflow}
      MYSQL_ROOT_PASSWORD: ${DB_ROOT_PASSWORD:-root}
    volumes:
      - mysql_data:/var/lib/mysql
    healthcheck:
      test: ["CMD", "mysqladmin", "ping", "-h", "localhost"]
      interval: 2s
      timeout: 2s
      retries: 60

  redis:
    image: ${DW_FAILOVER_REDIS_IMAGE:?The runner resolves the acceleration image}
    healthcheck:
      test: ["CMD", "redis-cli", "ping"]
      interval: 2s
      timeout: 2s
      retries: 30

volumes:
  mysql_data: