解决fly.io部署Django项目时创建两台机器的问题

在fly.io上部署Django项目时，系统自动尝试创建2台机器，超出免费额度导致报错，报错信息如下：

Running nameless-paper-520 release_command: python hw10/manage.py migrate
release_command 5683d609cd778e completed successfully
This deployment will:

• create 2 "app" machines

No machines in group app, launching a new machine
Error: error creating a new machine: failed to launch VM: To create more than 1 machine per app please add a payment method.

已尝试在fly.toml的[experimental]段设置max_machines = 1但无效，当前fly.toml配置：

app = "fly-builder-throbbing-forest-5735"
primary_region = "waw"
console_command = "/code/myapp/manage.py shell"

[deploy]
  release_command = "python myapp/manage.py migrate"

[env]
  ...
[http_service]
  internal_port = 8000
  force_https = true
  auto_stop_machines = true
  auto_start_machines = true
  min_machines_running = 0

[[statics]]
  guest_path = "/code/static"
  url_prefix = "/static/"
  
[[services]]
  internal_port = 8000
  protocol = "tcp"
  [services.concurrency]
    hard_limit = 1
    soft_limit = 1

[experimental]
  max_machines = 1

使用的Dockerfile：

ARG PYTHON_VERSION=3.10

FROM python:${PYTHON_VERSION}

ENV PYTHONDONTWRITEBYTECODE 1
ENV PYTHONUNBUFFERED 1

RUN mkdir -p /code

WORKDIR /code

RUN pip install pipenv
COPY Pipfile Pipfile.lock /code/
RUN pipenv install --deploy --system
COPY . /code

EXPOSE 8000

CMD ["gunicorn", "--bind", ":8000", "--workers", "2", "myapp.wsgi"]

需要修改配置以仅部署1台机器。

• 移除重复服务配置：fly.toml中同时存在[http_service]和[[services]]配置，这会触发fly.io为两种服务部署机器，导致多机器创建。删除整个[[services]]代码段，保留[http_service]即可。
• 设置正确的机器数量限制：[experimental]下的max_machines已被废弃，需在[http_service]段内添加max_machines_running = 1，修改后的[http_service]配置如下：

  [http_service]
    internal_port = 8000
    force_https = true
    auto_stop_machines = true
    auto_start_machines = true
    min_machines_running = 0
    max_machines_running = 1
  

• 可选优化：Dockerfile中Gunicorn设置了--workers 2，针对免费规格的单机器，可调整为--workers 1以匹配资源，修改后的CMD如下：

  CMD ["gunicorn", "--bind", ":8000", "--workers", "1", "myapp.wsgi"]
  

内容的提问来源于stack exchange，提问作者Sergio