Tag Archives: python



In my latest project I try to use the same application server in development and production environment (for simplicity, easier configuration, faster bugfixing, etc.) To achieve that I switched from embedded Flask / Werkzeug server in development and Apache mod_pyhon in production to embeddable, production-ready Waitress server.

One of the features I missed was automatically restarting the server whenever the code changes. I asked about this feature on Github and very helpful community members suggested various solutions, but none of the answers was satisfactory, so I decided to implement my own.


First I checked how other Python servers tackle this problem and all I came across work by spawning a monitor process (not thread!) with the same parameters as the main process, but with a special environment variable to distinguish between them. Examples:
* Werkzeug Reloadercode
* Pyramid pservecode

This can be confusing for a programmer when the server is embedded inside the application, because any code before entering main event loop (updating database schema, opening a config file, displaying “Starting…” message, etc.) will be executed twice. It seems much cleaner to me to use a dedicated program that will monitor the server process and restart it when necessary.

I found some generic utilities that can do that (e.g. inotifywait, watchmedo trick), but none of them behaves exactly how I want, so I created reload.



It monitors current directory and subdirectories for any changes, ignoring paths specified in .reloadignore file as regular expressions. Perhaps I will add support for reading .gitignore and other files later.


In the end the implementation turned out pretty simple. reload is programming language and server independent and it can be used when developing with anything that restarts reasonably quickly.


burstLogging is always a compromise between storing maximum amount of context information and maintaining good performance + saving disk space. In other words some errors occur only in production environment or are very difficult to reproduce locally, but you can’t store all debug messages on production systems.

I’d like to propose an alternative solution to this dilemma: BurstLogging.

The idea is simple – log only some debug messages that were created shortly before an error message was logged.

This is achieved by storing all logs in buffer, logging only informational messages during normal operation and dumping all messages when an error occurs. Chronological order of messages is preserved (debug messages that are older than currently logged info message are dropped) and there is no huge performance penalty (messages are formatted only when they are emitted).

Current implementation is written in Python, but it shouldn’t be difficult to port it to other programming languages or implement a language agnostic solution communicating via a port or a pipe.

The project is really new and it hasn’t been used in production yet. Please tell me what do you think about this idea.