I'm writing a chapter about debugging JSP applications and I want to over logging with Log4J. I understand how to configure Log4J in a web app, but I'm trying to understand the philosophy of using Log4J in a Servlet/JSP web appliction.
It is my understanding, and my experience with at least one app server, that if you log to the ServletContext.log() methods then you get the following benefits: * The app server will handle log buffering so that log messages are only written to disk every X seconds, where X is a configurable log-flush-interval value. * The app server will handle log rotation so that log files are renamed or deleted every Y seconds or log files are not allowed to grow beyound some size Z. * You can use the app server's Administrative UI to view log files on distributed servers without having to telnet to or otherwise visit each server. I know that Log4J handles log buffering, log rotation, and log viewing (chainsaw?), but in some cases users might want to use Log4J and at the same time take advantage of the app server's logging support described above. So I have a couple of questions: 1) are the above points valid points? 2) does Log4J include a ServletContextAppender that writes to the ServletContext log methods? 3) if no ServletContextAppender exists, then why? Thanks for your assistance. - Dave -- To unsubscribe, e-mail: <mailto:[EMAIL PROTECTED]> For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>
