docker-compose.yaml

networks:
  internal_network:
    internal: true
  external_network:
    driver: bridge

volumes:
  grafana_shared_config:
  shared_logs:

services:
  mc-observability-manager:
    image: cloudbaristaorg/mc-observability:edge
    container_name: mc-observability-manager
    restart: always
    networks:
      - internal_network
      - external_network
    ports:
      - 18080:18080
    depends_on:
      mc-observability-grafana:
        condition: service_healthy
      mc-observability-infra:
        condition: service_healthy
      mc-observability-influx:
        condition: service_healthy
      mc-observability-influx-2:
        condition: service_healthy
      mc-observability-loki:
        condition: service_healthy
      mc-observability-maria:
        condition: service_healthy
      mc-observability-rabbitmq:
        condition: service_healthy
      mc-observability-tempo:
        condition: service_healthy
    env_file:
      - ./config/manager/.env
    volumes:
      - shared_logs:/applog/mc-o11y-manager:rw
      - grafana_shared_config:/grafana_config:ro
    healthcheck:
      start_period: 120s
      test: 'curl -s http://127.0.0.1:18080/api/docs ; STATUS=`echo $$?` ; if [ $$STATUS != 0 ] && [ $$STATUS != 52 ]; then exit 1 ; fi'
      interval: 10s
      timeout: 10s
      retries: 30

  mc-observability-front:
    build:
      context: ./front
      dockerfile: Dockerfile
    container_name: mc-observability-front
    restart: always
    networks:
      - internal_network
      - external_network
    ports:
      - 18081:18081
    depends_on:
      mc-observability-manager:
        condition: service_healthy

  mc-observability-infra:
    image: cloudbaristaorg/mc-observability-infra:edge
    container_name: mc-observability-infra
    restart: always
    networks:
      - internal_network
      - external_network
    ports:
      - 33000:3000
    depends_on:
      mc-observability-maria:
        condition: service_healthy
    env_file:
      - ./config/mc-observability-infra/.env
    volumes:
      - ./config/mc-observability-infra/ansible:/ansible
    healthcheck:
      start_period: 120s
      test: 'curl -s http://127.0.0.1:3000/api/ping ; STATUS=`echo $$?` ; if [ $$STATUS != 0 ] && [ $$STATUS != 52 ]; then exit 1 ; fi'
      interval: 10s
      timeout: 10s
      retries: 30

  mc-observability-rabbitmq-init-volumes:
    image: busybox:stable
    container_name: mc-observability-rabbitmq-init-volumes
    restart: no
    networks:
      - internal_network
    command:
      - sh
      - -c
      - |
        set -eux && \
        mkdir -p /var/lib/rabbitmq /var/log/rabbitmq && \
        if [ ! -s /var/lib/rabbitmq/.erlang.cookie ] || \
           [ "$(wc -c < /var/lib/rabbitmq/.erlang.cookie)" -lt 20 ]; then
          umask 177
          tr -dc 'A-Za-z0-9' </dev/urandom | head -c 48 > /var/lib/rabbitmq/.erlang.cookie
        fi && \
        chmod 600 /var/lib/rabbitmq/.erlang.cookie && \
        chown -R 132:141 /var/lib/rabbitmq /var/log/rabbitmq
    volumes:
      - /docker/rabbitmq_data:/var/lib/rabbitmq
    user: root
    init: true

  mc-observability-rabbitmq:
    image: rabbitmq:4.1.4-management-alpine
    container_name: mc-observability-rabbitmq
    restart: always
    ports:
      - "5672:5672"    # AMQP
      - "1883:1883"    # MQTT
      - "15672:15672"  # Dashboard
      # - "8883:8883"    # MQTT/TLS (optional)
      # - "15675:15675"  # MQTT WebSocket (optional)
    depends_on:
      mc-observability-rabbitmq-init-volumes:
        condition: service_completed_successfully
    environment:
      - RABBITMQ_ENABLED_PLUGINS_FILE=/etc/rabbitmq/enabled_plugins
      - RABBITMQ_LOAD_DEFINITIONS=/etc/rabbitmq/definitions.json
    volumes:
      - ./config/rabbitmq:/etc/rabbitmq
      - /docker/rabbitmq_data:/var/lib/rabbitmq
    networks:
      - internal_network
      - external_network
    healthcheck:
      start_period: 120s
      test: rabbitmq-diagnostics -q ping
      interval: 10s
      timeout: 10s
      retries: 30

  mc-observability-maria:
    image: mariadb:10.11.11
    container_name: mc-observability-maria
    restart: always
    networks:
      - internal_network
      - external_network
    ports:
      - 3306:3306
    volumes:
      - ./config/mariadb/maria_init.sql:/docker-entrypoint-initdb.d/maria_init.sql
      - ./config/mariadb/99-max-connections.cnf:/etc/mysql/mariadb.conf.d/99-max-connections.cnf
      - /docker/mariadb/etc/mysql/conf.d:/etc/mysql/conf.d:ro
      - /docker/mariadb/var/lib/mysql:/var/lib/mysql
      - /docker/mariadb/var/log/maria:/var/log/maria
    environment:
      - TZ="Asia/Seoul"
      - ALLOW_EMPTY_PASSWORD=no
      - MARIADB_ROOT_PASSWORD=qwe1212!Q
      - MARIADB_USER=mc-agent
      - MARIADB_DATABASE=mc_observability
      - MARIADB_PASSWORD=mc-agent
    healthcheck:
      start_period: 120s
      test: 'mysql -uroot -p"$$MARIADB_ROOT_PASSWORD" -e"SELECT 1;" || exit 1'
      interval: 10s
      timeout: 10s
      retries: 30

  mc-observability-influx-init-volumes:
    image: busybox:stable
    container_name: mc-observability-influx-init-volumes
    restart: no
    command: [ "sh", "-c", "chown -R 1500:1500 /var/lib/influxdb" ]
    volumes:
      - /docker/influxdb_data:/var/lib/influxdb
    user: root
    init: true
    networks:
      - internal_network

  mc-observability-influx:
    image: cloudbaristaorg/mc-observability-influx:edge
    container_name: mc-observability-influx
    restart: always
    networks:
      - internal_network
      - external_network
    ports:
      - 8086:8086
    depends_on:
      mc-observability-influx-init-volumes:
        condition: service_completed_successfully
    environment:
      - INFLUXDB_HTTP_AUTH_ENABLED=true
      - INFLUXDB_ADMIN_USER=mc-agent
      - INFLUXDB_ADMIN_PASSWORD=mc-agent
      - INFLUXDB_DB="mc-observability"
    volumes:
      - ./config/influxdb/influxdb-init:/docker-entrypoint-initdb.d
      - /docker/influxdb_data:/var/lib/influxdb
    healthcheck:
      start_period: 120s
      test: 'curl -s http://127.0.0.1:8086/ping ; STATUS=`echo $$?` ; if [ $$STATUS != 0 ] && [ $$STATUS != 52 ]; then exit 1 ; fi'
      interval: 10s
      timeout: 10s
      retries: 30

  mc-observability-influx-2-init-volumes:
    image: busybox:stable
    container_name: mc-observability-influx-2-init-volumes
    restart: no
    command: [ "sh", "-c", "chown -R 1500:1500 /var/lib/influxdb" ]
    volumes:
      - /docker/influxdb2_data:/var/lib/influxdb
    user: root
    init: true
    networks:
      - internal_network

  mc-observability-influx-2:
    image: cloudbaristaorg/mc-observability-influx:edge
    container_name: mc-observability-influx-2
    restart: always
    networks:
      - internal_network
      - external_network
    ports:
      - 8087:8086
    depends_on:
      mc-observability-influx-2-init-volumes:
        condition: service_completed_successfully
    environment:
      - INFLUXDB_HTTP_AUTH_ENABLED=true
      - INFLUXDB_ADMIN_USER=mc-agent
      - INFLUXDB_ADMIN_PASSWORD=mc-agent
      - INFLUXDB_DB="mc-observability"
    volumes:
      - ./config/influxdb/influxdb-init:/docker-entrypoint-initdb.d
      - /docker/influxdb2_data:/var/lib/influxdb
    healthcheck:
      start_period: 120s
      test: 'curl -s http://127.0.0.1:8086/ping ; STATUS=`echo $$?` ; if [ $$STATUS != 0 ] && [ $$STATUS != 52 ]; then exit 1 ; fi'
      interval: 10s
      timeout: 10s
      retries: 30

  mc-observability-loki-init-volumes:
    image: busybox:stable
    container_name: mc-observability-loki-init-volumes
    restart: no
    command: [ "sh", "-c", "chown -R 10001:10001 /loki" ]
    volumes:
      - /docker/loki_data:/loki
    user: root
    init: true
    networks:
      - internal_network

  mc-observability-loki:
    image: grafana/loki:3.4.2
    container_name: mc-observability-loki
    restart: always
    networks:
      - internal_network
      - external_network
    ports:
      - "3100:3100"
    depends_on:
      mc-observability-loki-init-volumes:
        condition: service_completed_successfully
    environment:
      - TZ=Asia/Seoul
    volumes:
      - /etc/localtime:/etc/localtime:ro
      - ./config/loki/config.yml:/etc/loki/local-config.yaml
      - /docker/loki_data:/loki
    command: -config.file=/etc/loki/local-config.yaml
    healthcheck:
      start_period: 120s
      test: 'wget --quiet --spider http://127.0.0.1:3100/metrics || exit 1'
      interval: 10s
      timeout: 10s
      retries: 30

  mc-observability-tempo-init-volumes:
    image: busybox:stable
    container_name: mc-observability-tempo-init-volumes
    restart: no
    command: ["sh", "-c", "chown -R 10001:10001 /etc/tempo"]
    volumes:
      - /docker/tempo_data:/etc/tempo
    user: root
    init: true
    networks:
      - internal_network

  mc-observability-tempo:
    image: grafana/tempo:2.8.2
    container_name: mc-observability-tempo
    restart: always
    ports:
      - "3200:3200"
      - "4317:4317"
      - "4318:4318"
    depends_on:
      mc-observability-tempo-init-volumes:
        condition: service_completed_successfully
    environment:
      - TZ=Asia/Seoul
    command: [ "-config.file=/etc/tempo-config.yaml" ]
    volumes:
      - /etc/localtime:/etc/localtime:ro
      - ./config/tempo/config.yaml:/etc/tempo-config.yaml
      - /docker/tempo_data:/etc/tempo
    networks:
      - internal_network
      - external_network
    healthcheck:
      start_period: 120s
      test: [ "CMD", "wget", "--quiet", "--spider", "http://127.0.0.1:3200/ready" ]
      interval: 10s
      timeout: 10s
      retries: 30

  mc-observability-grafana-init-volumes:
    image: busybox:stable
    container_name: mc-observability-grafana-init-volumes
    restart: no
    networks:
      - internal_network
    command: [ "sh", "-c", "chown -R 472:472 /var/lib/grafana && chown -R 472:472 /var/log/grafana && chown -R 472:472 /grafana_config" ]
    volumes:
      - /docker/grafana_data/data:/var/lib/grafana
      - /docker/grafana_data/log:/var/log/grafana
      - grafana_shared_config:/grafana_config
    user: root
    init: true

  mc-observability-grafana:
    image: cloudbaristaorg/mc-observability-grafana:edge
    container_name: mc-observability-grafana
    restart: always
    networks:
      - internal_network
      - external_network
    ports:
      - "33001:3000"
    depends_on:
      mc-observability-grafana-init-volumes:
        condition: service_completed_successfully
    env_file:
      - ./config/grafana/.env
    volumes:
      - /etc/localtime:/etc/localtime:ro
      - ./config/grafana:/etc/grafana
      - /docker/grafana_data/data:/var/lib/grafana
      - /docker/grafana_data/log:/var/log/grafana
      - grafana_shared_config:/grafana_config
    healthcheck:
      start_period: 120s
      test: |
        curl -s http://127.0.0.1:3000
        STATUS=`echo $$?`
        if [ $$STATUS != 0 ] && [ $$STATUS != 52 ]; then
          exit 1
        fi
        curl -k -s https://127.0.0.1:8443
        STATUS=`echo $$?`
        if [ $$STATUS != 0 ] && [ $$STATUS != 52 ]; then
          exit 1
        fi
      interval: 10s
      timeout: 10s
      retries: 30

  mc-observability-chronograf:
    image: chronograf:1.9.4
    container_name: mc-observability-chronograf
    restart: always
    ports:
      - 8888:8888
    volumes:
      - /docker/chronograf_data:/var/lib/chronograf
    networks:
      - internal_network
      - external_network
    environment:
      - INFLUXDB_URL=http://mc-observability-influx:8086
      - INFLUXDB_USERNAME=mc-agent
      - INFLUXDB_PASSWORD=mc-agent

  mc-observability-insight:
    image: cloudbaristaorg/mc-observability-insight:edge
    container_name: mc-observability-insight
    restart: always
    networks:
      - internal_network
      - external_network
    ports:
      - 9001:9001
    environment:
      - TZ=Asia/Seoul
      - OLLAMA_BASE_URL=${OLLAMA_BASE_URL:-}
    volumes:
      - shared_logs:/mc-insight/log:rw
    depends_on:
      mc-observability-maria:
        condition: service_healthy
      mc-observability-influx:
        condition: service_healthy
    healthcheck:
      test: [ "CMD", "curl", "-f", "http://localhost:9001/readyz" ]
      interval: 1m
      timeout: 5s
      retries: 3
      start_period: 10s


  mc-observability-insight-scheduler:
    image: cloudbaristaorg/mc-observability-insight-scheduler:edge
    container_name: mc-observability-insight-scheduler
    restart: always
    ports:
      - 9002:9002
    depends_on:
      mc-observability-maria:
        condition: service_started
    environment:
      - TZ=Asia/Seoul
      - AIRFLOW__DATABASE__SQL_ALCHEMY_CONN=mysql+mysqldb://mc-agent:mc-agent@mc-observability-maria:3306/mc_airflow
    networks:
      - internal_network
      - external_network
    volumes:
      - /etc/localtime:/etc/localtime:ro
    command: >
      /bin/bash -c "
        # Wait for MySQL
        sleep 10

        # Clean up pid
        rm -f airflow-webserver.pid

        # Set up metadata database
        airflow db init

        # Create default user
        airflow users create --username admin --password admin --email admin@innogrid.com --firstname admin --lastname admin --role Admin

        # Import variables & Make connections
        airflow variables set --description 'O11Y Manger API BASE URL' API_BASE_URL http://mc-observability-manager:18080/api/o11y

        airflow connections add --conn-type http --conn-host mc-observability-insight --conn-schema http --conn-port 9001 api_base_url
        airflow connections add --conn-type mysql --conn-host mc-observability-maria --conn-schema mc_observability --conn-login mc-agent --conn-password mc-agent --conn-port 3306 mcmp_db
        # Reload & Run dags
        airflow dags reserialize
        airflow dags unpause anomaly_detection
        airflow dags unpause down_sampling

        # Start airflow
        airflow scheduler & airflow webserver -p 9002

        # Keep the server on no matter what
        sleep infinity
            "
    healthcheck:
      start_period: 120s
      test: 'curl -s http://127.0.0.1:9002 ; STATUS=`echo $?` ; if [ $$STATUS != 0 ] && [ $$STATUS != 52 ]; then exit 1 ; fi'
      interval: 10s
      timeout: 10s
      retries: 30

  mc-observability-mcp-grafana:
    image: mcp/grafana:latest
    container_name: mc-observability-mcp-grafana
    restart: always
    tty: true
    stdin_open: true
    ports:
      - "8000:8000"
    depends_on:
      mc-observability-grafana:
        condition: service_healthy
    environment:
      - GRAFANA_URL=http://mc-observability-grafana:3000
      - GRAFANA_API_KEY=${GRAFANA_API_KEY:-}
    entrypoint: ["/bin/bash", "/app/docker-entrypoint.sh"]
    command: ["/bin/bash", "-c", "chmod +x /app/docker-entrypoint.sh"]
    volumes:
      - ./config/mcp-grafana/docker-entrypoint.sh:/app/docker-entrypoint.sh:ro
      - grafana_shared_config:/grafana_config
    networks:
      - internal_network
      - external_network

  mc-observability-mcp-maria:
    image: cloudbaristaorg/mc-observability-mcp-mariadb:edge
    container_name: mc-observability-mcp-maria
    restart: always
    networks:
      - internal_network
      - external_network
    ports:
      - "8001:8001"
    depends_on:
      mc-observability-maria:
        condition: service_healthy
    environment:
      - TZ=Asia/Seoul
      - DB_HOST=mc-observability-maria
      - DB_PORT=3306
      - DB_USER=mc-agent
      - DB_PASSWORD=mc-agent
      - DB_NAME=mc_observability
      - MCP_READ_ONLY=true
      - MCP_MAX_POOL_SIZE=10
      - OPENAI_API_KEY=${OPENAI_API_KEY:-}

  mc-observability-mcp-influx:
    image: cloudbaristaorg/mc-observability-mcp-influxdb:edge
    container_name: mc-observability-mcp-influx
    restart: always
    ports:
      - "8002:8002"
    depends_on:
      mc-observability-influx:
        condition: service_healthy
    environment:
      - TZ=Asia/Seoul
      - INFLUXDB_URL=http://mc-observability-influx:8086
      - INFLUXDB_USER=mc-agent
      - INFLUXDB_PASSWORD=mc-agent
      - INFLUXDB_DATABASE=mc-observability
    networks:
      - internal_network
      - external_network

  mc-observability-log-collector:
    image: fluent/fluent-bit:3.2.4
    container_name: mc-observability-log-collector
    restart: always
    networks:
      - internal_network
      - external_network
    depends_on:
      mc-observability-loki:
        condition: service_healthy
    environment:
      - TZ=Asia/Seoul
      - LOKI_HOST=mc-observability-loki
      - LOKI_PORT=3100
    volumes:
      - ./config/log-collector:/fluent-bit/etc:ro
      - shared_logs:/logs:ro
    command:
      - /fluent-bit/bin/fluent-bit
      - -c
      - /fluent-bit/etc/fluent-bit.conf
    healthcheck:
      test: ["CMD-SHELL", "wget --quiet --spider http://127.0.0.1:2020/api/v1/health || exit 1"]
      interval: 10s
      timeout: 5s
      retries: 3
      start_period: 10s


  # CB-Tumblebug
  mc-infra-manager:
    image: cloudbaristaorg/cb-tumblebug:0.12.8
    container_name: mc-infra-manager
    restart: always
    networks:
      - internal_network
      - external_network
    ports:
      - 1323:1323
    depends_on:
      mc-infra-manager-etcd:
        condition: service_started
      mc-infra-connector:
        condition: service_started
      mc-infra-manager-postgres:
        condition: service_healthy
    volumes:
      - ./config/mc-infra-manager/setup.env:/app/conf/setup.env
      - ./config/mc-infra-manager/cloud_conf.yaml:/app/conf/cloud_conf.yaml
      - ~/.cloud-barista/credentials.yaml.enc:/app/conf/credentials.yaml.enc
      - /docker/mc-infra-manager-container/meta_db/:/app/meta_db/
      - /docker/mc-infra-manager-container/log/:/app/log/
    environment:
      # - TB_ROOT_PATH=/app
      # # Enable TB_SELF_ENDPOINT to specify an endpoint for CB-TB API (default: localhost:1323)
      # # Use public IP if you want to access the API Dashboard from outside of localhost
      # - TB_SELF_ENDPOINT=xxx.xxx.xxx.xxx:1323
      - TB_SPIDER_REST_URL=http://mc-infra-connector:1024/spider
      - TB_ETCD_ENDPOINTS=http://mc-infra-manager-etcd:2379
      - TB_TERRARIUM_REST_URL=http://mc-terrarium:8055/terrarium
      - TB_IAM_MANAGER_REST_URL=http://mc-iam-manager:5000
      # - TB_ETCD_AUTH_ENABLED=false
      # - TB_ETCD_USERNAME=default
      # - TB_ETCD_PASSWORD=default
      - TB_POSTGRES_ENDPOINT=mc-infra-manager-postgres:5432
      - TB_POSTGRES_DATABASE=tumblebug
      - TB_POSTGRES_USER=tumblebug
      - TB_POSTGRES_PASSWORD=tumblebug
    # - TB_TERRARIUM_API_USERNAME=default
    # - TB_TERRARIUM_API_PASSWORD=default
    # - TB_ALLOW_ORIGINS=*
    # - TB_AUTH_ENABLED=true
    # - TB_AUTH_MODE=jwt
    # - TB_API_USERNAME=default
    # - TB_API_PASSWORD=$$2a$$10$$4PKzCuJ6fPYsbCF.HR//ieLjaCzBAdwORchx62F2JRXQsuR3d9T0q
    # - TB_AUTOCONTROL_DURATION_MS=10000
    # - TB_DRAGONFLY_REST_URL=http://cb-dragonfly:9090/dragonfly
    # - TB_DEFAULT_NAMESPACE=default
    # - TB_DEFAULT_CREDENTIALHOLDER=admin
    # - TB_LOGFILE_PATH=/app/log/tumblebug.log
    # - TB_LOGFILE_MAXSIZE=1000
    # - TB_LOGFILE_MAXBACKUPS=3
    # - TB_LOGFILE_MAXAGE=30
    # - TB_LOGFILE_COMPRESS=false
    # - TB_LOGLEVEL=debug
    # - TB_LOGWRITER=both
    # - TB_NODE_ENV=development
    healthcheck: # for CB-Tumblebug
      test: [ "CMD", "curl", "-f", "http://localhost:1323/tumblebug/readyz" ]
      interval: 1m
      timeout: 5s
      retries: 3
      start_period: 10s

  # CB-Tumblebug ETCD
  # This is used for storing CB-Tumblebug metadata.
  mc-infra-manager-etcd:
    image: gcr.io/etcd-development/etcd:v3.5.21
    container_name: mc-infra-manager-etcd
    restart: always
    networks:
      - internal_network
    ports:
      - 2379:2379
      - 2380:2380
    volumes:
      - /docker/mc-infra-manager/etcd/data:/etcd-data
    entrypoint: /usr/local/bin/etcd
    command:
      - --name
      - s1
      - --data-dir
      - /etcd-data
      - --listen-client-urls
      - http://0.0.0.0:2379
      - --advertise-client-urls
      - http://0.0.0.0:2379
      - --listen-peer-urls
      - http://0.0.0.0:2380
      - --initial-advertise-peer-urls
      - http://0.0.0.0:2380
      - --initial-cluster
      - s1=http://0.0.0.0:2380
      - --initial-cluster-token
      - tkn
      - --initial-cluster-state
      - new
      - --log-level
      - info
      - --logger
      - zap
      - --log-outputs
      - stderr
      - --auth-token
      - simple
    healthcheck: # for etcd
      test: [ "CMD", "/usr/local/bin/etcd", "--version" ]
      interval: 1m
      timeout: 5s
      retries: 3
      start_period: 10s

  # CB-Tumblebug PostgreSQL
  # This is used for storing CB-Tumblebug Spec and Image.
  mc-infra-manager-postgres:
    image: postgres:16-alpine
    container_name: mc-infra-manager-postgres
    restart: always
    networks:
      - internal_network
      # Enable external network for outbound access (not ideal for security)
      # In production, comment this out to disable external access
      - external_network
    ports:
      - 5432:5432
    volumes:
      - /docker/mc-infra-manager-container/meta_db/postgres/:/var/lib/postgresql/data/
    environment:
      - POSTGRES_USER=tumblebug
      - POSTGRES_PASSWORD=tumblebug
      - POSTGRES_DB=tumblebug
    healthcheck:
      test: [ "CMD-SHELL", "pg_isready -U cb_tumblebug" ]
      interval: 10s
      timeout: 5s
      retries: 5
      start_period: 10s

  # CB-Spider
  mc-infra-connector:
    image: cloudbaristaorg/cb-spider:edge
    container_name: mc-infra-connector
    restart: always
    networks:
      - internal_network
      - external_network # for outbound access (not ideal for security)
    ports:
      - 1024:1024
    volumes:
      - /docker/cb-spider/meta_db/:/root/go/src/github.com/cloud-barista/cb-spider/meta_db/
      - /docker/cb-spider/log/:/root/go/src/github.com/cloud-barista/cb-spider/log/
    environment:
      - PLUGIN_SW=OFF
      - SERVER_ADDRESS=0.0.0.0:1024
      # if you leave these values empty, REST Auth will be disabled.
      - SPIDER_USERNAME=default
      - SPIDER_PASSWORD=default
      - SPIDER_LOG_LEVEL=error
      - SPIDER_HISCALL_LOG_LEVEL=error
      - ID_TRANSFORM_MODE=OFF
    healthcheck: # for CB-Spider
      test: [ "CMD", "curl", "-f", "http://localhost:1024/spider/readyz" ]
      interval: 1m
      timeout: 5s
      retries: 3
      start_period: 10s