Clean up Docker Compose (#1368)

* Fix hatchet, dockerfile * Update compose * point to correct docker image
SciPhi-AI · Oct 10, 2024 · 87c1e9e · 87c1e9e
1 parent ca9acdf
commit 87c1e9e
Show file tree

Hide file tree

Showing 3 changed files with 81 additions and 129 deletions.
diff --git a/py/Dockerfile b/py/Dockerfile
@@ -34,9 +34,9 @@ COPY --from=builder /usr/local/lib/python3.10/site-packages /usr/local/lib/pytho
 COPY --from=builder /usr/local/bin /usr/local/bin
 
 # Expose the port and set environment variables
-ARG PORT=8000 HOST=0.0.0.0
-ENV PORT=$PORT HOST=$HOST
-EXPOSE $PORT
+ARG R2R_PORT=8000 HOST=0.0.0.0
+ENV R2R_PORT=$R2R_PORT HOST=$HOST
+EXPOSE $R2R_PORT
 
 COPY . /app
 # Copy the application and config
@@ -47,4 +47,4 @@ COPY r2r.toml /app/r2r.toml
 COPY pyproject.toml /app/pyproject.toml
 
 # Run the application
-CMD ["sh", "-c", "uvicorn core.main.app_entry:app --host $HOST --port $PORT"]
+CMD ["sh", "-c", "uvicorn core.main.app_entry:app --host $HOST --port $R2R_PORT"]
diff --git a/py/compose.full.yaml b/py/compose.full.yaml
@@ -1,22 +1,23 @@
 networks:
   r2r-network:
-    name: r2r-network
     driver: bridge
     attachable: true
-    ipam:
-      driver: default
-      config:
-        - subnet: 172.28.0.0/16
     labels:
       - "com.docker.compose.recreate=always"
 
 volumes:
   hatchet_certs:
+    name: ${VOLUME_HATCHET_CERTS:-hatchet_certs}
   hatchet_config:
+    name: ${VOLUME_HATCHET_CONFIG:-hatchet_config}
   hatchet_api_key:
+    name: ${VOLUME_HATCHET_API_KEY:-hatchet_api_key}
   postgres_data:
+    name: ${VOLUME_POSTGRES_DATA:-postgres_data}
   hatchet_rabbitmq_data:
-  hatchet_rabbitmq.conf:
+    name: ${VOLUME_HATCHET_RABBITMQ_DATA:-hatchet_rabbitmq_data}
+  hatchet_rabbitmq_conf:
+    name: ${VOLUME_HATCHET_RABBITMQ_CONF:-hatchet_rabbitmq_conf}
 
 services:
   postgres:
@@ -33,7 +34,7 @@ services:
     networks:
       - r2r-network
     ports:
-      - "5432:5432"
+      - "${R2R_POSTGRES_PORT:-5432}:5432"
     healthcheck:
       test: ["CMD-SHELL", "pg_isready -U ${POSTGRES_USER:-postgres}"]
       interval: 10s
@@ -48,14 +49,14 @@ services:
     image: "rabbitmq:3-management"
     hostname: "hatchet-rabbitmq"
     ports:
-      - "5673:5672"
-      - "15673:15672"
+      - "${R2R_RABBITMQ_PORT:-5673}:5672"
+      - "${R2R_RABBITMQ_MGMT_PORT:-15673}:15672"
     environment:
       RABBITMQ_DEFAULT_USER: "user"
       RABBITMQ_DEFAULT_PASS: "password"
     volumes:
-      - "hatchet_rabbitmq_data:/var/lib/rabbitmq"
-      - "hatchet_rabbitmq.conf:/etc/rabbitmq/rabbitmq.conf"
+      - hatchet_rabbitmq_data:/var/lib/rabbitmq
+      - hatchet_rabbitmq_conf:/etc/rabbitmq/rabbitmq.conf
     healthcheck:
       test: ["CMD", "rabbitmqctl", "status"]
       interval: 10s
@@ -85,7 +86,7 @@ services:
       POSTGRES_USER: "${POSTGRES_USER:-postgres}"
       POSTGRES_PASSWORD: "${POSTGRES_PASSWORD:-postgres}"
       POSTGRES_HOST: "${POSTGRES_HOST:-postgres}"
-      POSTGRES_PORT: "${POSTGRES_PORT:-5432}"
+      POSTGRES_PORT: "${R2R_POSTGRES_PORT:-5432}"
       HATCHET_POSTGRES_DBNAME: "${HATCHET_POSTGRES_DBNAME:-hatchet}"
     networks:
       - r2r-network
@@ -96,10 +97,10 @@ services:
       POSTGRES_USER: "${POSTGRES_USER:-postgres}"
       POSTGRES_PASSWORD: "${POSTGRES_PASSWORD:-postgres}"
       POSTGRES_HOST: "${POSTGRES_HOST:-postgres}"
-      POSTGRES_PORT: "${POSTGRES_PORT:-5432}"
+      POSTGRES_PORT: "${R2R_POSTGRES_PORT:-5432}"
       HATCHET_POSTGRES_DBNAME: "${HATCHET_POSTGRES_DBNAME:-hatchet}"
       DATABASE_URL: "postgres://${POSTGRES_USER:-postgres}:${POSTGRES_PASSWORD:-postgres}@${POSTGRES_HOST:-postgres}:${POSTGRES_PORT:-5432}/${HATCHET_POSTGRES_DBNAME:-hatchet}?sslmode=disable"
-      DATABASE_POSTGRES_PORT: "${POSTGRES_PORT:-5432}"
+      DATABASE_POSTGRES_PORT: "${R2R_POSTGRES_PORT:-5432}"
       DATABASE_POSTGRES_HOST: "${POSTGRES_HOST:-postgres}"
       DATABASE_POSTGRES_USERNAME: "${POSTGRES_USER:-postgres}"
       DATABASE_POSTGRES_PASSWORD: "${POSTGRES_PASSWORD:-postgres}"
@@ -128,12 +129,12 @@ services:
       DATABASE_POSTGRES_PASSWORD: "${POSTGRES_PASSWORD:-postgres}"
       HATCHET_DATABASE_POSTGRES_DB_NAME: "${HATCHET_POSTGRES_DBNAME:-hatchet}"
       SERVER_TASKQUEUE_RABBITMQ_URL: amqp://user:password@hatchet-rabbitmq:5672/
-      SERVER_AUTH_COOKIE_DOMAIN: "http://host.docker.internal:7274"
-      SERVER_URL: "http://host.docker.internal:7274"
+      SERVER_AUTH_COOKIE_DOMAIN: "http://host.docker.internal:${R2R_HATCHET_DASHBOARD_PORT:-7274}"
+      SERVER_URL: "http://host.docker.internal:${R2R_HATCHET_DASHBOARD_PORT:-7274}"
       SERVER_AUTH_COOKIE_INSECURE: "t"
       SERVER_GRPC_BIND_ADDRESS: "0.0.0.0"
       SERVER_GRPC_INSECURE: "t"
-      SERVER_GRPC_BROADCAST_ADDRESS: "host.docker.internal:7077"
+      SERVER_GRPC_BROADCAST_ADDRESS: "hatchet-engine:7077"
       SERVER_GRPC_MAX_MSG_SIZE: 13421772800
     volumes:
       - hatchet_certs:/hatchet/certs
@@ -154,17 +155,17 @@ services:
       hatchet-setup-config:
         condition: service_completed_successfully
     ports:
-      - "7077:7077"
+      - "${R2R_HATCHET_ENGINE_PORT:-7077}:7077"
     environment:
       POSTGRES_USER: "${POSTGRES_USER:-postgres}"
       POSTGRES_PASSWORD: "${POSTGRES_PASSWORD:-postgres}"
       POSTGRES_HOST: "${POSTGRES_HOST:-postgres}"
       POSTGRES_PORT: "${POSTGRES_PORT:-5432}"
       HATCHET_POSTGRES_DBNAME: "${HATCHET_POSTGRES_DBNAME:-hatchet}"
       DATABASE_URL: "postgres://${POSTGRES_USER:-postgres}:${POSTGRES_PASSWORD:-postgres}@${POSTGRES_HOST:-postgres}:${POSTGRES_PORT:-5432}/${HATCHET_POSTGRES_DBNAME:-hatchet}?sslmode=disable"
-      SERVER_GRPC_BROADCAST_ADDRESS: "host.docker.internal:7077"
+      SERVER_GRPC_BROADCAST_ADDRESS: "hatchet-engine:7077"
       SERVER_GRPC_BIND_ADDRESS: "0.0.0.0"
-      SERVER_GRPC_PORT: 7077
+      SERVER_GRPC_PORT: "7077"
       SERVER_GRPC_INSECURE: "t"
       SERVER_GRPC_MAX_MSG_SIZE: 13421772800
       SERVER_LOGGER_LEVEL: warn
@@ -175,9 +176,7 @@ services:
       - hatchet_certs:/hatchet/certs
       - hatchet_config:/hatchet/config
     networks:
-      r2r-network:
-        aliases:
-          - host.docker.internal
+      - r2r-network
     healthcheck:
       test: ["CMD", "wget", "-q", "-O", "-", "http://localhost:8733/live"]
       interval: 10s
@@ -197,17 +196,14 @@ services:
       POSTGRES_HOST: host.docker.internal
       POSTGRES_PORT: "${POSTGRES_PORT:-5432}"
       HATCHET_POSTGRES_DBNAME: "${HATCHET_POSTGRES_DBNAME:-hatchet}"
-      DATABASE_URL: "postgres://${POSTGRES_USER:-postgres}:${POSTGRES_PASSWORD:-postgres}@${POSTGRES_HOST:-host.docker.internal}:${POSTGRES_PORT:-5432}/${HATCHET_POSTGRES_DBNAME:-hatchet}?sslmode=disable"
+      DATABASE_URL: "postgres://${POSTGRES_USER:-postgres}:${POSTGRES_PASSWORD:-postgres}@${POSTGRES_HOST:-host.docker.internal}:${R2R_POSTGRES_PORT:-5432}/${HATCHET_POSTGRES_DBNAME:-hatchet}?sslmode=disable"
     volumes:
       - hatchet_certs:/hatchet/certs
       - hatchet_config:/hatchet/config
     networks:
       - r2r-network
-    labels:
-      - "traefik.enable=true"
-      - "traefik.http.routers.hatchet-dashboard.rule=PathPrefix(`/`)"
-      - "traefik.http.services.hatchet-frontend.loadbalancer.server.port=80"
-      - "traefik.http.routers.hatchet-dashboard.entrypoints=hatchet"
+    ports:
+      - "${R2R_HATCHET_DASHBOARD_PORT:-7274}:80"
 
   setup-token:
     image: ghcr.io/hatchet-dev/hatchet/hatchet-admin:v0.49.0-alpha.4
@@ -287,55 +283,84 @@ services:
   unstructured:
     image: ${UNSTRUCTURED_IMAGE:-ragtoriches/unst-prod}
     ports:
-      - "7275:7275"
+      - "${R2R_UNSTRUCTURED_PORT:-7275}:7275"
     networks:
       - r2r-network
     healthcheck:
       test: ["CMD", "curl", "-f", "http://localhost:7275/health"]
       interval: 10s
       timeout: 5s
       retries: 5
+
   r2r:
-    image: ${R2R_IMAGE:-ragtoriches/prod:latest-unstructured}
+    image: ${R2R_IMAGE:-ragtoriches/prod:latest}
     build:
       context: .
       args:
-        PORT: ${PORT:-7272}
+        PORT: ${R2R_PORT:-${PORT:-7272}} # Eventually get rid of PORT, but for now keep it for backwards compatibility
+        R2R_PORT: ${R2R_PORT:-${PORT:-7272}}
     ports:
-      - "${PORT:-7272}:${PORT:-7272}"
+      - "${R2R_PORT:-${PORT:-7272}}:${R2R_PORT:-${PORT:-7272}}"
     environment:
       - PYTHONUNBUFFERED=1
-      - PORT=${PORT:-7272}
+      - R2R_PORT=${R2R_PORT:-${PORT:-7272}} # Eventually get rid of PORT, but for now keep it for backwards compatibility
       - HOST=${HOST:-0.0.0.0}
+
+      # R2R
       - CONFIG_NAME=${CONFIG_NAME:-}
       - CONFIG_PATH=${CONFIG_PATH:-}
+
+      # Postgres
       - POSTGRES_USER=${POSTGRES_USER:-postgres}
       - POSTGRES_PASSWORD=${POSTGRES_PASSWORD:-postgres}
       - POSTGRES_HOST=${POSTGRES_HOST:-postgres}
       - POSTGRES_PORT=${POSTGRES_PORT:-5432}
       - POSTGRES_DBNAME=${POSTGRES_DBNAME:-postgres}
+      - POSTGRES_MAX_CONNECTIONS=${POSTGRES_MAX_CONNECTIONS:-}
+      - POSTGRES_PROJECT_NAME=${POSTGRES_PROJECT_NAME:-r2r_default}
       - R2R_PROJECT_NAME=${R2R_PROJECT_NAME:-r2r_default}
-      - POSTGRES_MAX_CONNECTIONS=${POSTGRES_MAX_CONNECTIONS:-1024}
+
+      # OpenAI
       - OPENAI_API_KEY=${OPENAI_API_KEY:-}
       - OPENAI_API_BASE=${OPENAI_API_BASE:-}
+
+      # Anthropic
       - ANTHROPIC_API_KEY=${ANTHROPIC_API_KEY:-}
+
+      # Azure
       - AZURE_API_KEY=${AZURE_API_KEY:-}
       - AZURE_API_BASE=${AZURE_API_BASE:-}
       - AZURE_API_VERSION=${AZURE_API_VERSION:-}
+
+      # Google Vertex AI
       - GOOGLE_APPLICATION_CREDENTIALS=${GOOGLE_APPLICATION_CREDENTIALS:-}
       - VERTEX_PROJECT=${VERTEX_PROJECT:-}
       - VERTEX_LOCATION=${VERTEX_LOCATION:-}
+
+      # AWS Bedrock
       - AWS_ACCESS_KEY_ID=${AWS_ACCESS_KEY_ID:-}
       - AWS_SECRET_ACCESS_KEY=${AWS_SECRET_ACCESS_KEY:-}
       - AWS_REGION_NAME=${AWS_REGION_NAME:-}
+
+      # Groq
       - GROQ_API_KEY=${GROQ_API_KEY:-}
+
+      # Cohere
       - COHERE_API_KEY=${COHERE_API_KEY:-}
+
+      # Anyscale
       - ANYSCALE_API_KEY=${ANYSCALE_API_KEY:-}
+
+      # Ollama
       - OLLAMA_API_BASE=${OLLAMA_API_BASE:-http://host.docker.internal:11434}
+
+      # Unstructured
       - UNSTRUCTURED_API_KEY=${UNSTRUCTURED_API_KEY:-}
       - UNSTRUCTURED_API_URL=${UNSTRUCTURED_API_URL:-https://api.unstructured.io/general/v0/general}
       - UNSTRUCTURED_LOCAL_URL=${UNSTRUCTURED_LOCAL_URL:-http://unstructured:7275}
       - UNSTRUCTURED_NUM_WORKERS=${UNSTRUCTURED_NUM_WORKERS:-10}
+
+      # Hatchet
       - HATCHET_CLIENT_TLS_STRATEGY=none
       - HATCHET_CLIENT_GRPC_MAX_RECV_MESSAGE_LENGTH=${HATCHET_CLIENT_GRPC_MAX_RECV_MESSAGE_LENGTH:-13421772800}
       - HATCHET_CLIENT_GRPC_MAX_SEND_MESSAGE_LENGTH=${HATCHET_CLIENT_GRPC_MAX_SEND_MESSAGE_LENGTH:-13421772800}
@@ -344,31 +369,19 @@ services:
         if [ -z "$${HATCHET_CLIENT_TOKEN}" ]; then
           export HATCHET_CLIENT_TOKEN=$$(cat /hatchet_api_key/api_key.txt)
         fi
-        exec uvicorn core.main.app_entry:app --host $${HOST} --port $${PORT}
+        exec uvicorn core.main.app_entry:app --host $${HOST} --port $${R2R_PORT}
       '
     networks:
       - r2r-network
     healthcheck:
-      test: ["CMD", "curl", "-f", "http://localhost:${PORT:-7272}/v2/health"]
+      test: ["CMD", "curl", "-f", "http://localhost:${R2R_PORT:-7272}/v2/health"]
       interval: 6s
       timeout: 5s
       retries: 5
     restart: on-failure
     volumes:
       - ${CONFIG_PATH:-/}:${CONFIG_PATH:-/app/config}
       - hatchet_api_key:/hatchet_api_key:ro
-    labels:
-      - "traefik.enable=true"
-      - "traefik.http.routers.r2r.rule=PathPrefix(`/r2r-api`)"
-      - "traefik.http.services.r2r.loadbalancer.server.port=${PORT:-7272}"
-      - "traefik.http.middlewares.r2r-strip-prefix.stripprefix.prefixes=/api"
-      - "traefik.http.middlewares.r2r-strip-prefix.stripprefix.prefixes=/r2r-api"
-      - "traefik.http.middlewares.r2r-add-v2.addprefix.prefix=/v2"
-      - "traefik.http.routers.r2r.middlewares=r2r-strip-prefix,r2r-add-v2,r2r-headers"
-      - "traefik.http.middlewares.r2r-headers.headers.customrequestheaders.Access-Control-Allow-Origin=*"
-      - "traefik.http.middlewares.r2r-headers.headers.customrequestheaders.Access-Control-Allow-Methods=GET,POST,OPTIONS"
-      - "traefik.http.middlewares.r2r-headers.headers.customrequestheaders.Access-Control-Allow-Headers=DNT,User-Agent,X-Requested-With,If-Modified-Since,Cache-Control,Content-Type,Range,Authorization"
-      - "traefik.http.middlewares.r2r-headers.headers.customresponseheaders.Access-Control-Expose-Headers=Content-Length,Content-Range"
     extra_hosts:
       - host.docker.internal:host-gateway
     depends_on:
@@ -380,32 +393,8 @@ services:
   r2r-dashboard:
     image: emrgntcmplxty/r2r-dashboard:latest
     environment:
-      - NEXT_PUBLIC_API_URL=http://localhost:${R2R_DASHBOARD_PORT:-8001}/r2r-api
+      - NEXT_PUBLIC_API_URL=http://localhost:${R2R_PORT:-7272}/r2r-api
     networks:
       - r2r-network
-    labels:
-      - "traefik.enable=true"
-      - "traefik.http.routers.r2r-dashboard.rule=PathPrefix(`/`)"
-      - "traefik.http.services.r2r-dashboard.loadbalancer.server.port=3000"
-      - "traefik.http.routers.r2r-dashboard.entrypoints=r2r"
-
-  traefik:
-    image: traefik:v2.9
-    command:
-      - "--api.insecure=true"
-      - "--providers.docker=true"
-      - "--providers.docker.exposedbydefault=false"
-      - "--entrypoints.r2r.address=:${R2R_DASHBOARD_PORT:-8001}"
-      - "--entrypoints.hatchet.address=:${HATCHET_DASHBOARD_PORT:-8002}"
-      - "--accesslog=true"
-      - "--accesslog.filepath=/var/log/traefik/access.log"
     ports:
-      - "${R2R_DASHBOARD_PORT:-8001}:${R2R_DASHBOARD_PORT:-8001}"
-      - "${HATCHET_DASHBOARD_PORT:-8002}:${HATCHET_DASHBOARD_PORT:-8002}"
-      - "${TRAEFIK_DASHBOARD_PORT:-8081}:${TRAEFIK_DASHBOARD_PORT:-8081}"
-    volumes:
-      - /var/run/docker.sock:/var/run/docker.sock:ro
-    networks:
-      - r2r-network
-    labels:
-      - "traefik.http.middlewares.no-cache-headers.headers.customResponseHeaders.Cache-Control=no-cache, no-store, must-revalidate"
+      - "${R2R_DASHBOARD_PORT:-7273}:3000"