7 May
2008
7 May
'08
7:19 p.m.
That's the solution you've picked.
I still haven't figured out how to actually do this in the configuration file.
What is the reasoning which lead you to pick this solution?
I am going to be using syslog to log messages from an app I am writing. I need to be able to have the messages start going into a new file at an arbitrary time.
If nothing else, there are significant security and denial of service issues inherent in this idea, especially when accepting UDP syslog messages, trivially spoofable.
I am only going to be handling messages from my app, not a catch-all source.