什么是创建Python守护程序的事实库 [英] What is the de facto library for creating Python Daemons

问题描述

我正在尝试使用在我看来是 python-daemon 库是创建守护程序而不忘记任何内容的最安全方法。该文档非常糟糕，只是 PEP 3143 。

I am attempting to use the python-daemon library which seemed to me to be the safest way to create a Daemon without forgetting anything. The documentation is quite poor, being just PEP 3143.

另一方面，我发现了很多与Sander Marechal的使用Python的简单Unix / Linux守护程序。

On the other hand, I have found a lot of links to Sander Marechal's A simple unix/linux daemon in Python. This looks to be a nicer solution or though I have not yet attempted to use it.

编辑：我已经使用了Sander Marechal的解决方案，并且看起来效果很好。

I have used Sander Marechal's solution and it seems to work nicely.

那么在Python社区中创建Daemon的实际方式是什么，是这些库之一，还是干脆做

So what is the de facto way in the Python community to create a Daemon, is it one of these libraries, or simply doing it all yourself (forking twice etc.)?

此外，您会认为任何带有PEP的库都是更好的选择，因为它更接近于一种全面的方法创建守护程序（或至少以一种更标准的方式）的方法比任何其他解决方案都要好。那么，此 python-daemon 包有何处理，它将包含在标准库中吗？

Also, you would think that any library with a PEP would be a far better choice since it is closer to a comprehensive way of creating a Daemon (or at least a more standard way) than any other solution. So what is the deal with this python-daemon package, would it ever be included in the standard library?

推荐答案

我选择了Sander Marechal的一个简单的Python Unix / Linux守护程序，它很简单，您可以通过创建一个子类并覆盖 run（）方法来使用它，这很自然做事的方式（而不是 with context： python-daemon模块的方法。

I went with Sander Marechal's A simple unix/linux daemon in Python, it is simple, and you work with it by creating a subclass and overriding the run() method, which feels a very natural way to do things (rather than the with context: approach of the python-daemon module.

这篇关于什么是创建Python守护程序的事实库的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！