Hacker News new | past | comments | ask | show | jobs | submit login
Deploying Tornado in production (agiletesting.blogspot.com)
39 points by icey on Dec 18, 2009 | hide | past | favorite | 5 comments



There is some sound information in this. I've been planning on writing about the the curve in using Tornado, as it is deceivingly simple. It does take diving into the code to really pump out a polished site.

I've been working on a stub project which serves as the foundation for the tornado sites I've built: http://github.com/gmr/Tornado-Project-Stub

The nice thing with it is that it makes going from 0 to a working site that covers all the basics very easy.

I've chosen some different methodologies for dealing with daemonizing and logging than Evite, but that's the nice thing about the Python standard libraries; many ways to skin that cat.


How do you do daemonize and log?


Daemonizing is done by process forking using the os module and logging is done via the standard logging module with an optional syslog handler.


are you taking advantage of the commit that implements pre-forking?

http://github.com/facebook/tornado/commit/6fb90ae694190fcedc...


I gave this a try and couldn't get it to work as I'd expect based upon previous experience with Tornado. I'm waiting until they polish up a 0.3 release to adopt using it.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: