YG-Soft/YG_FT
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

第一次提交

Browse Source
This commit is contained in:
wangjiming
2026-07-27 09:12:47 +08:00
commit b4ff5db17b
579 changed files with 48768 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

267
docker/README.md Normal file
View File

@@ -0,0 +1,267 @@
# Docker 部署说明
本目录按应用服务器和算力服务器拆分 Dockerfile 与 Docker Compose 文件。Compose 文件不包含 `build:`，不会在 `docker compose up` 时自动构建业务镜像。所有业务镜像需要先通过手动 `docker build` 构建，再由 Compose 启动。
## 基础镜像
| 镜像 | 用途 |
| --- | --- |
| `python:3.12-slim` | 应用后端基础镜像，后端运行环境要求 Python 3.12 及以上 |
| `nginx:1.27-alpine` | 前端静态资源与 `/modelTF` 反向代理运行镜像 |
| `hiyouga/llamafactory:latest` | 算力服务基础镜像，基于 LLaMA-Factory 官方镜像扩展 Compute API |
| `postgres:16-alpine` | 开发阶段内置 PostgreSQL |
| `redis:7-alpine` | 开发阶段内置 Redis |
一键拉取基础镜像：
```bash
docker pull python:3.12-slim && \
docker pull nginx:1.27-alpine && \
docker pull hiyouga/llamafactory:latest && \
docker pull postgres:16-alpine && \
docker pull redis:7-alpine
```
Windows PowerShell：
```powershell
$images = @(
"python:3.12-slim",
"nginx:1.27-alpine",
"hiyouga/llamafactory:latest",
"postgres:16-alpine",
"redis:7-alpine"
)
$images | ForEach-Object { docker pull $_ }
```
如果部署环境不能访问外网，需要提前在可联网环境执行上述拉取命令，再用 `docker save` / `docker load` 导出导入。
## 业务镜像
| 镜像 | Dockerfile | 构建命令 |
| --- | --- | --- |
| `yg-ft-backend-api:latest` | `docker/app/Dockerfile.backend` | `docker build -f docker/app/Dockerfile.backend -t yg-ft-backend-api:latest .` |
| `yg-ft-frontend-runtime:latest` | `docker/app/Dockerfile.frontend` | `docker build -f docker/app/Dockerfile.frontend -t yg-ft-frontend-runtime:latest .` |
| `yg-ft-compute-api:latest` | `docker/compute/Dockerfile.compute` | `docker build -f docker/compute/Dockerfile.compute -t yg-ft-compute-api:latest .` |
## 对外端口
所有宿主机对外端口统一使用 5 位端口。容器内部端口保持镜像默认端口，便于容器内服务和健康检查稳定。
| 服务 | 宿主机对外端口 | 容器内部端口 | 说明 |
| --- | --- | --- | --- |
| 前端 Nginx | `16801` | `80` | 前端页面入口 |
| 后端 API | `17861` | `8000` | FastAPI 服务 |
| PostgreSQL | `15432` | `5432` | 开发阶段内置数据库 |
| Redis | `16379` | `6379` | 开发阶段内置缓存 |
| Compute API | `19100` | `9100` | 算力服务器 API |
| File Gateway | `19101` | 后续服务端口 | 当前预留，后续拆出文件网关服务时使用 |
注意：`8000` 是后端容器内部端口，不作为宿主机对外访问端口。宿主机或浏览器应访问 `http://<app-server-ip>:17861/modelTF/health`；前端 Nginx 容器在 Docker 网络内部访问 `http://backend-api:8000/modelTF/...`。
对应配置文件：
- `docker/app/.env.example`
- `FRONTEND_PORT=16801`
- `BACKEND_API_PORT=17861`
- `POSTGRES_PORT=15432`
- `REDIS_PORT=16379`
- `docker/compute/.env.example`
- `COMPUTE_API_PORT=19100`
- `FILE_GATEWAY_PORT=19101`
## 运行模式
- 应用侧默认 `COMPUTE_MODE=real`，任务状态必须由真实算力同步逻辑更新。
- 算力侧默认 `COMPUTE_EXECUTION_MODE=real`，真实执行器未完成前不会伪造训练作业。
- 仅隔离联调时可显式设置 `COMPUTE_MODE=simulator` 或 `COMPUTE_EXECUTION_MODE=simulator`，该模式不得用于测试环境、生产环境或生产升级基线。
## 应用服务器部署
应用服务器包含前端 Nginx、Backend API、PostgreSQL、Redis。
当前 Compose 内置 PostgreSQL 使用 `backend/app/db/sql/001_platform_runtime.sql` 初始化运行库。`docs/postgres-schema.sql` 是完整目标架构设计，不应直接挂载为当前运行库初始化脚本，否则会与当前后端代码的运行表结构不兼容。
首次部署：
```bash
cd <repo-root>
# 1. 使用当前 Windows/宿主机 npm 构建前端静态产物
cd frontend
npm ci
npm run build
cd ..
# 2. 手动构建业务镜像
docker build -f docker/app/Dockerfile.backend -t yg-ft-backend-api:latest .
docker build -f docker/app/Dockerfile.frontend -t yg-ft-frontend-runtime:latest .
# 3. 启动应用服务
cd docker/app
cp .env.example .env
docker compose up -d
```
后端镜像构建过程中会执行依赖导入自检，确认 `fastapi`、`uvicorn`、`psycopg`、`sqlalchemy`、`redis` 等运行依赖已安装。构建后也可以手动检查：
```bash
docker run --rm yg-ft-backend-api:latest python -c "import psycopg; print(psycopg.__version__)"
```
默认访问地址：
```text
http://<app-server-ip>:16801
```
应用侧代码和数据外挂：
```text
../../backend -> /app
../../frontend/dist -> /usr/share/nginx/html
../../runtime/app/logs/backend -> /opt/yg-ft/logs/backend
../../runtime/app/data -> /data/yg-ft
```
前端容器启动前必须确保 `../../frontend/dist/index.html` 已存在。若前端 Nginx 日志出现 `directory index of "/usr/share/nginx/html/" is forbidden` 或 `rewrite or internal redirection cycle while internally redirecting to "/index.html"`，通常表示当前执行 `docker compose` 的项目目录下没有构建好的 `frontend/dist`，或挂载路径不是同一份代码目录。
```bash
# 在执行 docker compose 的同一份代码目录中检查
cd <repo-root>/frontend
npm run build
test -f dist/index.html && ls -lh dist/index.html
cd ../docker/app
docker compose up -d --force-recreate frontend
docker compose logs --tail=80 frontend
```
如果使用 Windows npm 构建、WSL 中运行 Docker Compose，需要确认 Windows 路径和 WSL 路径指向同一份仓库。例如在 `D:\...\YG_FT\frontend` 构建不会自动生成 `/mnt/d/wuyongtao/Code/YG_FT/frontend/dist` 下的产物，除非二者本就是同一个目录。
如果使用企业统一 PostgreSQL/Redis，修改 `docker/app/.env`：
```env
DATABASE_URL=postgresql+psycopg://<user>:<password>@<postgres-host>:15432/<db>
REDIS_URL=redis://<redis-host>:16379/0
USE_BUILTIN_POSTGRES=false
USE_BUILTIN_REDIS=false
```
生产环境如完全使用外部基础设施，可以删除或注释 Compose 中的 `postgres`、`redis` 服务及 `backend-api.depends_on` 中对应依赖。
## 算力服务器部署
算力服务器包含 Compute API、后续 Compute Agent、File Gateway、GPU runtime、本地训练数据目录和 LLaMA-Factory。`Dockerfile.compute` 基于 LLaMA-Factory 官方镜像：
```dockerfile
FROM hiyouga/llamafactory:latest
```
部署前需要安装：
- NVIDIA Driver
- NVIDIA Container Toolkit
- Docker Engine 和 Docker Compose Plugin
- 本地训练数据目录，默认 `/data/yg-ft`
首次部署：
```bash
cd <repo-root>
# 手动构建算力业务镜像
docker build -f docker/compute/Dockerfile.compute -t yg-ft-compute-api:latest .
# 启动算力服务
cd docker/compute
cp .env.example .env
docker compose up -d
```
健康检查：
```text
GET http://<compute-server-ip>:19100/modelTF/health
GET http://<compute-server-ip>:19100/modelTF/v1/compute/health
```
算力侧代码和数据外挂：
```text
../../compute -> /app/compute
${YG_FT_DATA_ROOT_HOST} -> /data/yg-ft
../../runtime/compute/logs -> /opt/yg-ft/logs/compute
../../runtime/compute/training-logs -> /opt/yg-ft/logs/training
```
## 应用与算力分离部署
应用服务器只需要主动访问算力服务器，不要求算力服务器回调应用服务器。
在 `docker/app/.env` 中配置：
```env
COMPUTE_API_BASE_URL=http://<compute-server-ip>:19100
FILE_GATEWAY_BASE_URL=http://<compute-server-ip>:19101
COMPUTE_SERVICE_TOKEN=change_me
COMPUTE_STATUS_SYNC_MODE=polling
COMPUTE_POLL_INTERVAL_SECONDS=10
COMPUTE_POLL_BATCH_SIZE=100
```
交互链路：
```text
Frontend
-> Backend API
-> Compute API
-> Compute Agent / LLaMA-Factory
-> 本地数据目录 / 模型目录 / 训练产物
<- Backend Worker 定时轮询 Compute API
```
## 多算力节点部署
多算力节点仍按“单机多 GPU 节点”部署。每台 GPU 服务器都独立部署一套 `docker/compute`：
```text
gpu-node-01: docker/compute + /data/yg-ft + 19100/19101
gpu-node-02: docker/compute + /data/yg-ft + 19100/19101
gpu-node-03: docker/compute + /data/yg-ft + 19100/19101
```
节点之间默认不互访。应用平台主动访问每个节点的 Compute API/File Gateway，并通过 `compute_nodes`、`resource_replicas`、`resource_sync_jobs` 统一调度和同步。
## 常用命令
重新构建应用镜像：
```bash
docker build -f docker/app/Dockerfile.backend -t yg-ft-backend-api:latest .
docker build -f docker/app/Dockerfile.frontend -t yg-ft-frontend-runtime:latest .
```
重新构建算力镜像：
```bash
docker build -f docker/compute/Dockerfile.compute -t yg-ft-compute-api:latest .
```
启动服务：
```bash
cd docker/app
docker compose up -d
cd ../compute
docker compose up -d
```
查看服务：
```bash
docker compose ps
docker compose logs -f
```

45
docker/app/.env.example Normal file
View File

@@ -0,0 +1,45 @@
APP_ENV=prod
APP_NAME=YG Fine-Tune Platform API
MODELTF_ROUTE_PREFIX=/modelTF
CORS_ALLOW_ORIGINS=http://localhost:16801,http://127.0.0.1:16801
FRONTEND_IMAGE=yg-ft-frontend-runtime:latest
BACKEND_API_IMAGE=yg-ft-backend-api:latest
# Five-digit host ports exposed outside the application server.
FRONTEND_PORT=16801
BACKEND_API_PORT=17861
POSTGRES_PORT=15432
REDIS_PORT=16379
POSTGRES_DB=yg_ft
POSTGRES_USER=yg_ft
POSTGRES_PASSWORD=change_me
DATABASE_URL=postgresql+psycopg://yg_ft:change_me@postgres:5432/yg_ft
REDIS_URL=redis://redis:6379/0
# Development uses the built-in PostgreSQL/Redis services in docker-compose.yml.
# For enterprise infrastructure, replace DATABASE_URL/REDIS_URL and remove or disable those services.
USE_BUILTIN_POSTGRES=true
USE_BUILTIN_REDIS=true
LOG_LEVEL=INFO
LOG_DIR=/opt/yg-ft/logs/backend
LOG_FILE_PREFIX=backend
LOG_ERROR_FILE_PREFIX=error
LOG_MAX_BYTES=20971520
LOG_RETENTION_DAYS=10
BACKEND_PROXY_PASS=http://backend-api:8000
# Split deployment: set these to the compute server address, for example http://10.10.20.31:19100.
COMPUTE_API_BASE_URL=http://compute-api:9100
COMPUTE_SERVICE_TOKEN=change_me
FILE_GATEWAY_BASE_URL=http://compute-api:9101
# The application side polls Compute API for job state to avoid opening reverse network access.
COMPUTE_MODE=real
COMPUTE_STATUS_SYNC_MODE=polling
COMPUTE_POLL_INTERVAL_SECONDS=10
COMPUTE_POLL_BATCH_SIZE=100

21
docker/app/Dockerfile.backend Normal file
View File

@@ -0,0 +1,21 @@
FROM python:3.12-slim
ENV PYTHONDONTWRITEBYTECODE=1 \
PYTHONUNBUFFERED=1 \
PIP_NO_CACHE_DIR=1
WORKDIR /app
COPY backend/requirements.txt /tmp/requirements.txt
RUN pip install --upgrade pip -i https://pypi.tuna.tsinghua.edu.cn/simple \
&& pip install -r /tmp/requirements.txt -i https://pypi.tuna.tsinghua.edu.cn/simple \
&& rm -f /tmp/requirements.txt
RUN python -c "import fastapi, uvicorn, psycopg, sqlalchemy, redis, jwt, passlib, httpx, alembic; print('backend dependency check ok')"
RUN mkdir -p /opt/yg-ft/logs/backend /data/yg-ft \
&& chmod -R 0775 /opt/yg-ft /data/yg-ft
EXPOSE 8000
CMD ["uvicorn", "app.main:app", "--host", "0.0.0.0", "--port", "8000"]

9
docker/app/Dockerfile.frontend Normal file
View File

@@ -0,0 +1,9 @@
FROM nginx:1.27-alpine
RUN mkdir -p /usr/share/nginx/html
EXPOSE 80
HEALTHCHECK --interval=30s --timeout=3s --start-period=10s --retries=3 \
CMD test -f /usr/share/nginx/html/index.html && wget -qO- http://127.0.0.1/index.html >/dev/null || exit 1

123
docker/app/docker-compose.yml Normal file
View File

@@ -0,0 +1,123 @@
services:
frontend:
image: ${FRONTEND_IMAGE:-yg-ft-frontend-runtime:latest}
container_name: yg-ft-frontend
depends_on:
backend-api:
condition: service_started
ports:
- "${FRONTEND_PORT:-16801}:80"
environment:
BACKEND_PROXY_PASS: ${BACKEND_PROXY_PASS:-http://backend-api:8000}
volumes:
- ../../frontend/dist:/usr/share/nginx/html:ro
- ../nginx.conf.template:/etc/nginx/templates/default.conf.template:ro
command:
- /bin/sh
- -c
- |
if [ ! -f /usr/share/nginx/html/index.html ]; then
echo "frontend dist is missing: build frontend first and ensure ../../frontend/dist is mounted";
ls -la /usr/share/nginx/html;
exit 1;
fi;
nginx -g 'daemon off;'
networks:
- yg-ft-app
restart: unless-stopped
backend-api:
image: ${BACKEND_API_IMAGE:-yg-ft-backend-api:latest}
container_name: yg-ft-backend-api
depends_on:
postgres:
condition: service_healthy
redis:
condition: service_healthy
expose:
- "8000"
ports:
- "${BACKEND_API_PORT:-17861}:8000"
environment:
APP_ENV: ${APP_ENV:-prod}
APP_NAME: ${APP_NAME:-YG Fine-Tune Platform API}
MODELTF_ROUTE_PREFIX: ${MODELTF_ROUTE_PREFIX:-/modelTF}
CORS_ALLOW_ORIGINS: ${CORS_ALLOW_ORIGINS:-http://localhost:16801,http://127.0.0.1:16801}
DATABASE_URL: ${DATABASE_URL:-postgresql+psycopg://yg_ft:change_me@postgres:5432/yg_ft}
REDIS_URL: ${REDIS_URL:-redis://redis:6379/0}
USE_BUILTIN_POSTGRES: ${USE_BUILTIN_POSTGRES:-true}
USE_BUILTIN_REDIS: ${USE_BUILTIN_REDIS:-true}
LOG_LEVEL: ${LOG_LEVEL:-INFO}
LOG_DIR: ${LOG_DIR:-/opt/yg-ft/logs/backend}
LOG_FILE_PREFIX: ${LOG_FILE_PREFIX:-backend}
LOG_ERROR_FILE_PREFIX: ${LOG_ERROR_FILE_PREFIX:-error}
LOG_MAX_BYTES: ${LOG_MAX_BYTES:-20971520}
LOG_RETENTION_DAYS: ${LOG_RETENTION_DAYS:-10}
COMPUTE_API_BASE_URL: ${COMPUTE_API_BASE_URL:-http://compute-api:9100}
COMPUTE_SERVICE_TOKEN: ${COMPUTE_SERVICE_TOKEN:-change_me}
FILE_GATEWAY_BASE_URL: ${FILE_GATEWAY_BASE_URL:-http://compute-api:9101}
COMPUTE_MODE: ${COMPUTE_MODE:-real}
COMPUTE_STATUS_SYNC_MODE: ${COMPUTE_STATUS_SYNC_MODE:-polling}
COMPUTE_POLL_INTERVAL_SECONDS: ${COMPUTE_POLL_INTERVAL_SECONDS:-10}
COMPUTE_POLL_BATCH_SIZE: ${COMPUTE_POLL_BATCH_SIZE:-100}
PYTHONPATH: /app
volumes:
- ../../backend:/app:ro
- ../../runtime/app/logs/backend:/opt/yg-ft/logs/backend
- ../../runtime/app/data:/data/yg-ft
networks:
- yg-ft-app
healthcheck:
test: ["CMD-SHELL", "python -c \"import urllib.request; urllib.request.urlopen('http://127.0.0.1:8000/modelTF/health', timeout=3).read()\""]
interval: 30s
timeout: 5s
retries: 3
start_period: 20s
restart: unless-stopped
postgres:
image: postgres:16-alpine
container_name: yg-ft-postgres
environment:
POSTGRES_DB: ${POSTGRES_DB:-yg_ft}
POSTGRES_USER: ${POSTGRES_USER:-yg_ft}
POSTGRES_PASSWORD: ${POSTGRES_PASSWORD:-change_me}
PGDATA: /var/lib/postgresql/data/pgdata
volumes:
- postgres_data:/var/lib/postgresql/data
- ../../backend/app/db/sql/001_platform_runtime.sql:/docker-entrypoint-initdb.d/001-platform-runtime.sql:ro
ports:
- "${POSTGRES_PORT:-15432}:5432"
networks:
- yg-ft-app
healthcheck:
test: ["CMD-SHELL", "pg_isready -U $${POSTGRES_USER} -d $${POSTGRES_DB}"]
interval: 10s
timeout: 5s
retries: 5
restart: unless-stopped
redis:
image: redis:7-alpine
container_name: yg-ft-redis
command: ["redis-server", "--appendonly", "yes"]
volumes:
- redis_data:/data
ports:
- "${REDIS_PORT:-16379}:6379"
networks:
- yg-ft-app
healthcheck:
test: ["CMD", "redis-cli", "ping"]
interval: 10s
timeout: 3s
retries: 5
restart: unless-stopped
networks:
yg-ft-app:
name: yg-ft-app
volumes:
postgres_data:
redis_data:

23
docker/compute/.env.example Normal file
View File

@@ -0,0 +1,23 @@
COMPUTE_ENV=prod
COMPUTE_HOST_ID=gpu-node-01
COMPUTE_EXECUTION_MODE=real
MODELTF_ROUTE_PREFIX=/modelTF
# Five-digit host ports exposed outside the compute server.
COMPUTE_API_PORT=19100
FILE_GATEWAY_PORT=19101
COMPUTE_API_IMAGE=yg-ft-compute-api:latest
# The application server actively polls Compute API; compute server does not need reverse access.
COMPUTE_SERVICE_TOKEN=change_me
ENABLE_APP_CALLBACK=false
# LLaMA-Factory is provided by the official hiyouga/llamafactory base image.
LLAMA_FACTORY_HOME=/app/LLaMA-Factory
YG_FT_DATA_ROOT=/data/yg-ft
YG_FT_DATA_ROOT_HOST=/data/yg-ft
LOG_DIR=/opt/yg-ft/logs/compute
CUDA_VISIBLE_DEVICES=all
NVIDIA_VISIBLE_DEVICES=all
NVIDIA_DRIVER_CAPABILITIES=compute,utility

27
docker/compute/Dockerfile.compute Normal file
View File

@@ -0,0 +1,27 @@
FROM hiyouga/llamafactory:latest
ENV DEBIAN_FRONTEND=noninteractive \
PYTHONDONTWRITEBYTECODE=1 \
PYTHONUNBUFFERED=1 \
PIP_NO_CACHE_DIR=1
WORKDIR /app
RUN apt-get update \
&& apt-get install -y --no-install-recommends tini \
&& rm -rf /var/lib/apt/lists/*
COPY compute/requirements.txt /tmp/requirements.txt
RUN pip install --upgrade pip -i https://pypi.tuna.tsinghua.edu.cn/simple \
&& pip install -r /tmp/requirements.txt -i https://pypi.tuna.tsinghua.edu.cn/simple \
&& rm -f /tmp/requirements.txt
RUN mkdir -p /opt/yg-ft/logs/compute /opt/yg-ft/logs/training /data/yg-ft /app/LLaMA-Factory \
&& chmod -R 0775 /opt/yg-ft /data/yg-ft /app/LLaMA-Factory
ENTRYPOINT ["/usr/bin/tini", "--"]
EXPOSE 9100
CMD ["uvicorn", "compute.api.main:app", "--host", "0.0.0.0", "--port", "9100"]

39
docker/compute/docker-compose.yml Normal file
View File

@@ -0,0 +1,39 @@
services:
compute-api:
image: ${COMPUTE_API_IMAGE:-yg-ft-compute-api:latest}
container_name: yg-ft-compute-api
gpus: all
ports:
- "${COMPUTE_API_PORT:-19100}:9100"
environment:
COMPUTE_ENV: ${COMPUTE_ENV:-prod}
COMPUTE_HOST_ID: ${COMPUTE_HOST_ID:-gpu-node-01}
COMPUTE_EXECUTION_MODE: ${COMPUTE_EXECUTION_MODE:-real}
MODELTF_ROUTE_PREFIX: ${MODELTF_ROUTE_PREFIX:-/modelTF}
COMPUTE_SERVICE_TOKEN: ${COMPUTE_SERVICE_TOKEN:-change_me}
ENABLE_APP_CALLBACK: ${ENABLE_APP_CALLBACK:-false}
LLAMA_FACTORY_HOME: ${LLAMA_FACTORY_HOME:-/app/LLaMA-Factory}
YG_FT_DATA_ROOT: ${YG_FT_DATA_ROOT:-/data/yg-ft}
LOG_DIR: ${LOG_DIR:-/opt/yg-ft/logs/compute}
CUDA_VISIBLE_DEVICES: ${CUDA_VISIBLE_DEVICES:-all}
NVIDIA_VISIBLE_DEVICES: ${NVIDIA_VISIBLE_DEVICES:-all}
NVIDIA_DRIVER_CAPABILITIES: ${NVIDIA_DRIVER_CAPABILITIES:-compute,utility}
PYTHONPATH: /app
volumes:
- ../../compute:/app/compute:ro
- ${YG_FT_DATA_ROOT_HOST:-/data/yg-ft}:${YG_FT_DATA_ROOT:-/data/yg-ft}
- ../../runtime/compute/logs:/opt/yg-ft/logs/compute
- ../../runtime/compute/training-logs:/opt/yg-ft/logs/training
networks:
- yg-ft-compute
healthcheck:
test: ["CMD-SHELL", "python -c \"import urllib.request; urllib.request.urlopen('http://127.0.0.1:9100/modelTF/health', timeout=3).read()\""]
interval: 30s
timeout: 5s
retries: 3
start_period: 20s
restart: unless-stopped
networks:
yg-ft-compute:
name: yg-ft-compute

30
docker/nginx.conf.template Normal file
View File

@@ -0,0 +1,30 @@
server {
listen 80;
server_name _;
root /usr/share/nginx/html;
index index.html;
client_max_body_size 200m;
location / {
try_files $uri $uri/ /index.html;
}
location /modelTF {
proxy_pass ${BACKEND_PROXY_PASS};
proxy_http_version 1.1;
proxy_set_header Host $host;
proxy_set_header X-Real-IP $remote_addr;
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
proxy_set_header X-Forwarded-Proto $scheme;
proxy_read_timeout 300s;
proxy_send_timeout 300s;
}
location ~* \.(?:js|css|png|jpg|jpeg|gif|ico|svg|woff|woff2|ttf)$ {
try_files $uri =404;
expires 30d;
add_header Cache-Control "public, immutable";
}
}