From 7ae41a8d3dddc51fbba68b906cbe0c1213216e1d Mon Sep 17 00:00:00 2001
From: GuoQing Zhang <zhangguoqing@dataelem.com>
Date: Wed, 28 May 2025 19:17:40 +0800
Subject: [PATCH] feat: add celery worker container

---
 docker/bisheng/entrypoint.sh         | 17 +++++++++---
 docker/docker-compose.yml            | 40 +++++++++++++++++++++++++++-
 src/backend/bisheng/worker/config.py |  4 +++
 src/backend/entrypoint.sh            | 17 +++++++++---
 4 files changed, 71 insertions(+), 7 deletions(-)

diff --git a/docker/bisheng/entrypoint.sh b/docker/bisheng/entrypoint.sh
index c5a9affca..d754b4149 100644
--- a/docker/bisheng/entrypoint.sh
+++ b/docker/bisheng/entrypoint.sh
@@ -1,4 +1,15 @@
-nohup uvicorn bisheng.main:app --host 0.0.0.0 --port 7860 --no-access-log --workers 8 &
+#!/bin/bash
 
-# -c 是指定celery的并发数
-celery -A bisheng.worker.main worker -l info -c 16
+start_mode=${1:-api}
+
+if [ "$start_mode" == "api" ]; then
+    echo "Starting API server..."
+    uvicorn bisheng.main:app --host 0.0.0.0 --port 7860 --no-access-log --workers 8
+elif [ "$start_mode" == "celery" ]; then
+    echo "Starting Celery worker..."
+    # -c 是指定celery的并发数，线程数
+    celery -A bisheng.worker.main worker -l info -c 100 -P threads
+else
+    echo "Invalid start mode. Use 'api' or 'celery'."
+    exit 1
+fi
diff --git a/docker/docker-compose.yml b/docker/docker-compose.yml
index 33c21cd6f..9d287aa4a 100644
--- a/docker/docker-compose.yml
+++ b/docker/docker-compose.yml
@@ -75,7 +75,45 @@ services:
       - ${DOCKER_VOLUME_DIRECTORY:-.}/data/bisheng:/app/data
     security_opt:
       - seccomp:unconfined
-    command: sh entrypoint.sh # --workers 表示使用几个进程，提高并发度
+    command: sh entrypoint.sh api  # 启动api服务
+    restart: on-failure
+    healthcheck:
+      test: ["CMD", "curl", "-f", "http://localhost:7860/health"]
+      start_period: 30s
+      interval: 90s
+      timeout: 30s
+      retries: 3
+    depends_on:
+      mysql:
+        condition: service_healthy
+      redis:
+        condition: service_healthy
+      office:
+        condition: service_started
+  
+  backend_worker:
+    container_name: bisheng-backend-worker
+    image: dataelement/bisheng-backend:v1.3.0
+    environment:
+      TZ: Asia/Shanghai
+      BS_MILVUS_CONNECTION_ARGS: '{"host":"milvus","port":"19530","user":"","password":"","secure":false}'
+      BS_MILVUS_IS_PARTITION: 'true'
+      BS_MILVUS_PARTITION_SUFFIX: '1'
+      BS_ELASTICSEARCH_URL: 'http://elasticsearch:9200'
+      BS_ELASTICSEARCH_SSL_VERIFY: '{}'  # 可根据自己部署的密码进行配置 '{"basic_auth": ("elastic", "elastic")}'
+      BS_MINIO_SCHEMA: 'false'
+      BS_MINIO_CERT_CHECK: 'false'
+      BS_MINIO_ENDPOINT: 'minio:9000'
+      BS_MINIO_SHAREPOIN: 'minio:9000'
+      BS_MINIO_ACCESS_KEY: 'minioadmin'
+      BS_MINIO_SECRET_KEY: 'minioadmin'
+    volumes:
+      - ${DOCKER_VOLUME_DIRECTORY:-.}/bisheng/config/config.yaml:/app/bisheng/config.yaml
+      - ${DOCKER_VOLUME_DIRECTORY:-.}/bisheng/entrypoint.sh:/app/entrypoint.sh
+      - ${DOCKER_VOLUME_DIRECTORY:-.}/data/bisheng:/app/data
+    security_opt:
+      - seccomp:unconfined
+    command: sh entrypoint.sh worker  # 启动celery的异步worker服务，用来处理一些耗时的任务
     restart: on-failure
     healthcheck:
       test: ["CMD", "curl", "-f", "http://localhost:7860/health"]
diff --git a/src/backend/bisheng/worker/config.py b/src/backend/bisheng/worker/config.py
index 4c2dfd4fe..3ab50c665 100644
--- a/src/backend/bisheng/worker/config.py
+++ b/src/backend/bisheng/worker/config.py
@@ -7,3 +7,7 @@ result_serializer = 'json'
 accept_content = ['json']
 timezone = 'Asia/Shanghai'
 enable_utc = False
+task_routes = {
+    'bisheng.worker.workflow.*': {'queue': 'celery_workflow'},
+    'bisheng.worker.knowledge.*': {'queue': 'celery_knowledge'},
+}
diff --git a/src/backend/entrypoint.sh b/src/backend/entrypoint.sh
index 31357455f..d754b4149 100755
--- a/src/backend/entrypoint.sh
+++ b/src/backend/entrypoint.sh
@@ -1,4 +1,15 @@
-nohup uvicorn bisheng.main:app --host 0.0.0.0 --port 7860 --no-access-log --workers 8 &
+#!/bin/bash
 
-# -c 是指定celery的并发数，线程数
-celery -A bisheng.worker.main worker -l info -c 100 -P threads
+start_mode=${1:-api}
+
+if [ "$start_mode" == "api" ]; then
+    echo "Starting API server..."
+    uvicorn bisheng.main:app --host 0.0.0.0 --port 7860 --no-access-log --workers 8
+elif [ "$start_mode" == "celery" ]; then
+    echo "Starting Celery worker..."
+    # -c 是指定celery的并发数，线程数
+    celery -A bisheng.worker.main worker -l info -c 100 -P threads
+else
+    echo "Invalid start mode. Use 'api' or 'celery'."
+    exit 1
+fi