资讯详情

前提

Mem0 Restfull API 

github地址：https://github.com/mem0ai/mem0/tree/main/server

介绍文档：https://docs.mem0.ai/open-source/features/rest-api

开始配置

按正常流程，我把Mem0的包上传到虚拟机之后，修改.env基础配置如下（OPENAI_API_KEY修改成自己的）：

OPENAI_API_KEY=abcPOSTGRES_HOST=postgres
POSTGRES_PORT=5432
POSTGRES_DB=postgres
POSTGRES_USER=postgres
POSTGRES_PASSWORD=postgres
POSTGRES_COLLECTION_NAME=memoriesNEO4J_URI=bolt://neo4j:7687
NEO4J_USERNAME=neo4j
NEO4J_PASSWORD=mem0graph

然后修改docker-compose.yaml内容如下：

name: mem0-devservices:mem0:build:context: ..  # Set context to parent directorydockerfile: mem0/dev.Dockerfileports:- "8888:8000"env_file:- .envnetworks:- mem0_networkvolumes:- ./history:/app/history      # History db location. By default, it creates a history.db file on the server folder- .:/app                      # Server code. This allows to reload the app when the server code is updated- ../mem0:/app/packages/mem0  # Mem0 library. This allows to reload the app when the library code is updateddepends_on:postgres:condition: service_healthyneo4j:condition: service_healthycommand: uvicorn main:app --host 0.0.0.0 --port 8000 --reload  # Enable auto-reloadenvironment:- PYTHONDONTWRITEBYTECODE=1  # Prevents Python from writing .pyc files- PYTHONUNBUFFERED=1  # Ensures Python output is sent straight to terminalpostgres:image: ankane/pgvector:v0.5.1restart: on-failureshm_size: "128mb" # Increase this if vacuuming fails with a "no space left on device" errornetworks:- mem0_networkenvironment:- POSTGRES_USER=postgres- POSTGRES_PASSWORD=postgreshealthcheck:test: ["CMD", "pg_isready", "-q", "-d", "postgres", "-U", "postgres"]interval: 5stimeout: 5sretries: 5volumes:- postgres_db:/var/lib/postgresql/dataports:- "8432:5432"neo4j:image: neo4j:5.26.4networks:- mem0_networkhealthcheck:test: wget bolt://neo4j:8687 || exit 1
      interval: 1stimeout: 10sretries: 20start_period: 90sports:- "8474:7474" # HTTP- "8687:7687" # Boltvolumes:- ./neo4j/data:/data  # 将 Neo4j 数据存储到 mem0 目录- ./neo4j/logs:/logs  # 将 Neo4j 日志存储到 mem0 目录- ./neo4j/import:/var/lib/neo4j/import  # 将 Neo4j 导入目录存储到 mem0 目录environment:- NEO4J_AUTH=neo4j/mem0graph- NEO4J_PLUGINS=["apoc"]  # Add this line to install APOC- NEO4J_apoc_export_file_enabled=true- NEO4J_apoc_import_file_enabled=true- NEO4J_apoc_import_file_use__neo4j__config=truevolumes:neo4j_data:postgres_db:networks:mem0_network:driver: bridge

执行报错

然后执行 docker compose up 命令启动，结果报错：

dependency failed to start: container mem0-dev-neo4j-1 is unhealthy

查看neo4j日志：docker-compose logs neo4j

报错内容如下：

/usr/local/lib/python3.6/site-packages/paramiko/transport.py:32: CryptographyDeprecationWarning: Python 3.6 is no longer supported by the Python core team. Therefore, support for it is deprecated in cryptography. The next release of cryptography will remove support for Python 3.6. from cryptography.hazmat.backends import default_backend ERROR: The Compose file './docker-compose.yaml' is invalid because: 'name' does not match any of the regexes: '^x-'
You might be seeing this error because you're using the wrong Compose file version. Either specify a supported version (e.g "2.2" or "3.3") and place your service definitions under the services key, or omit the version key and place your service definitions at the root of the file to use version 1. For more on the Compose file format versions, see https://docs.docker.com/compose/compose-file/

错误信息显示有两个问题：

1. Docker Compose 版本问题：你的 Docker Compose 版本不支持 name 字段（mem0-dev 不被支持 - 这是 Docker Compose V2 的新特性）

2. 健康检查命令还没修改：日志显示还在使用旧的 wget bolt://neo4j:8687 命令

问题修正

后面我就重新修改了docker-compose.yaml配置如下：

version: '3.8'services:mem0:build:context: ..  # Set context to parent directorydockerfile: mem0/dev.Dockerfilecontainer_name: mem0-dev-mem0ports:- "8888:8000"env_file:- .envnetworks:- mem0_networkvolumes:- ./history:/app/history      # History db location. By default, it creates a history.db file on the server folder- .:/app                      # Server code. This allows to reload the app when the server code is updated- ../mem0:/app/packages/mem0  # Mem0 library. This allows to reload the app when the library code is updateddepends_on:postgres:condition: service_healthyneo4j:condition: service_healthycommand: uvicorn main:app --host 0.0.0.0 --port 8000 --reload  # Enable auto-reloadenvironment:- PYTHONDONTWRITEBYTECODE=1  # Prevents Python from writing .pyc files- PYTHONUNBUFFERED=1  # Ensures Python output is sent straight to terminalpostgres:image: ankane/pgvector:v0.5.1container_name: mem0-dev-postgresrestart: on-failureshm_size: "128mb" # Increase this if vacuuming fails with a "no space left on device" errornetworks:- mem0_networkenvironment:- POSTGRES_USER=postgres- POSTGRES_PASSWORD=postgreshealthcheck:test: ["CMD", "pg_isready", "-q", "-d", "postgres", "-U", "postgres"]interval: 5stimeout: 5sretries: 5volumes:- postgres_db:/var/lib/postgresql/dataports:- "8432:5432"neo4j:image: neo4j:5.26.4container_name: mem0-dev-neo4jnetworks:- mem0_networkhealthcheck:# test: ["CMD-SHELL", "curl -f http://localhost:7474 || exit 1"]test: ["CMD-SHELL", "wget -q --spider http://localhost:7474 || exit 1"]interval: 10stimeout: 5sretries: 10start_period: 90sports:- "8474:7474" # HTTP- "8687:7687" # Boltvolumes:- ./neo4j/data:/data  # 将 Neo4j 数据存储到 mem0 目录- ./neo4j/logs:/logs  # 将 Neo4j 日志存储到 mem0 目录- ./neo4j/import:/var/lib/neo4j/import  # 将 Neo4j 导入目录存储到 mem0 目录environment:- NEO4J_AUTH=neo4j/mem0graph- NEO4J_PLUGINS=["apoc"]- NEO4J_apoc_export_file_enabled=true- NEO4J_apoc_import_file_enabled=true- NEO4J_apoc_import_file_use__neo4j__config=truevolumes:neo4j_data:postgres_db:networks:mem0_network:driver: bridge

然后执行命令：

docker-compose down     # 将之前启动的容器关闭并删除
docker-compose up -d     # 重启容器

然后看到运行正常了：

 访问Swagger页面也正常：