text2sql-agent/docker-compose.prod.yml at main · Sakeeb91/text2sql-agent · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
# Arctic Text2SQL Agent - Production Environment Override
#
# Usage: docker-compose -f docker-compose.yml -f docker-compose.prod.yml up -d
#
# This file overrides the base docker-compose.yml for production deployments.

services:
  api:
    image: ghcr.io/sakeeb91/arctic-text2sql-agent:${IMAGE_TAG:-latest}
    build: !reset null
    environment:
      # Database
      - DATABASE_URL=${DATABASE_URL}
      # HuggingFace
      - HUGGINGFACE_TOKEN=${HUGGINGFACE_TOKEN}
      - TEXT2SQL_MODEL=${TEXT2SQL_MODEL:-Snowflake/Arctic-Text2SQL-R1-7B}
      - MODEL_DEVICE=${MODEL_DEVICE:-cuda}
      # API - Production settings
      - API_HOST=0.0.0.0
      - API_PORT=8000
      - API_DEBUG=false
      - CORS_ORIGINS=${CORS_ORIGINS}
      # Agent
      - AGENT_MAX_STEPS=${AGENT_MAX_STEPS:-5}
      - AGENT_MIN_CONFIDENCE=${AGENT_MIN_CONFIDENCE:-0.8}
      - AGENT_VERBOSITY=0
      # Logging
      - LOG_LEVEL=INFO
      - LOG_FORMAT=json
      # Cache
      - REDIS_URL=redis://redis:6379/0
      # Security
      - SECRET_KEY=${SECRET_KEY}
      # Rate Limiting
      - RATE_LIMIT_REQUESTS=${RATE_LIMIT_REQUESTS:-100}
      - RATE_LIMIT_WINDOW=${RATE_LIMIT_WINDOW:-60}
    volumes:
      # Production: no source mounts, only data volumes
      - app-data:/app/data
      - huggingface-cache:/home/appuser/.cache/huggingface
    deploy:
      mode: replicated
      replicas: ${REPLICAS:-2}
      resources:
        limits:
          memory: 32G
        reservations:
          memory: 16G
      update_config:
        parallelism: 1
        delay: 30s
        failure_action: rollback
        order: start-first
      rollback_config:
        parallelism: 1
        delay: 10s
      restart_policy:
        condition: any
        delay: 5s
        max_attempts: 3
        window: 120s
    labels:
      - "environment=production"
      - "version=${IMAGE_TAG:-latest}"
    logging:
      driver: json-file
      options:
        max-size: "100m"
        max-file: "5"

  db:
    environment:
      POSTGRES_USER: ${POSTGRES_USER}
      POSTGRES_PASSWORD: ${POSTGRES_PASSWORD}
      POSTGRES_DB: ${POSTGRES_DB:-text2sql_prod}
    deploy:
      resources:
        limits:
          memory: 4G
        reservations:
          memory: 2G
    labels:
      - "environment=production"
    logging:
      driver: json-file
      options:
        max-size: "50m"
        max-file: "3"

  redis:
    deploy:
      resources:
        limits:
          memory: 1G
        reservations:
          memory: 512M
    labels:
      - "environment=production"
    logging:
      driver: json-file
      options:
        max-size: "20m"
        max-file: "3"