A ChatGPT web client that supports multiple users, multiple languages, and multiple database connections for persistent data storage. Provides Docker images and quick deployment scripts.
DEBUG=False # Whether to enable debug mode, default False
APP_DOMAIN=gpt.mytestdomain.com:9000
SERVER_WORKERS=3 # The number of worker processes for handling requests.
WORKER_TIMEOUT=180 # Workers silent for more than this many seconds are killed and restarted. default 180s
DB_URL=postgres://postgres:postgrespw@localhost:49153/chatgpt # If this parameter is not set, the built-in Sqlite will be used by default. It should be noted that if you do not connect to an external database, the data will be lost after the container is destroyed.
DJANGO_SUPERUSER_USERNAME=admin # default superuser name
ACCOUNT_EMAIL_VERIFICATION=optional # Determines the e-mail verification method during signup – choose one of "none", "optional", or "mandatory". Default is "optional". If you don't need to verify the email, you can set it to "none".
If you want to use the email verification function, you need to configure the following parameters
感谢作者分享了这个项目,请问这个项目可以安装在 ARM64 架构的 aarch64 主机上吗?目前80,443,8000等端口都已经在 nginx, caddy 的占用中了。我刚接触 docker,还不熟悉。可以请你帮忙检查一下下面的 docker-compose.yml 文件有没有错误?如果修改正确后,是否运行 docker-compose up --pull always -d 就可以自动下载镜像并配置运行了?另外,有可能使用127.0.0.1替换下面文件中的domain,然后用原有的nginx增加不同的分流路径来指向client, wsgi-server, backend-web-server吗?比如用 www.mytestdomain.com/chatgpt 来指向主机上的 127.0.0.1:8800,再指向docker上的client。谢谢!
version: '3' services: client: platform: linux/aarch64 image: wongsaang/chatgpt-ui-client:latest environment:
If you want to use the email verification function, you need to configure the following parameters
volumes:
- db_sqlite3:/app/db.sqlite3
ports:
networks: chatgpt_ui_network: driver: bridge