hc-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Ortwin Gl├╝ck <...@odi.ch>
Subject Re: Logging in core / core-nio
Date Tue, 01 Apr 2008 18:23:01 GMT
Tatu Saloranta wrote:
> To me, logging statements in production are generally
> to be used for reporting unusual things happening

That totally depends on the application. We run a lot of server code that 
reflects not directly in any user interface. The log is the ONLY way to know 
what has been going on (especially in an error condition, when the transaction 
is rolled back and you thus have no change on the DB). Hooking the debugger into 
a production application server is not feasible most of the time, as setting a 
breakpoint is not selective to a specific thread and would block ANY thread that 
accidentially hits the break point. Even worse in non-blocking NIO code: a 
breakpoint will inadvertly affect hundreds of sessions.
To find bugs in server side logic, for me the log is the most important thing.

Ortwin

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@hc.apache.org
For additional commands, e-mail: dev-help@hc.apache.org


Mime
View raw message