Robert,
Which version of log4j is this? In log4j 1.1b6 there were deadlocking problems with
the AsyncAppender. Can you also describe what appenders and layouts you are using? In
most JVMs there is a keyboard combination that dumps information on threads and the
monitors they are waiting on. Under windows it's Ctrl-Pause. When deadlock occurs
could you see what that gives you? Regards, Ceki
At 16:29 28.06.2001 +1000, Robert Leftwich wrote:
>I am seeing what I think is log4j failing to handle a large number of log messages
>and wanted to check if anyone else has seen a similar problem before I spend time
>tracking it down (and I couldn't find anything similar mentioned in the archives).
>
>The scenario is Windows NT4+SP6, JDK 1.3.1, Enhydra App Server running Barracuda (a
>presentation framework that uses log4j for it's logging). If I enable DEBUG (or INFO)
>logging at the root category level with a console or file appender the app server
>shuts down after a short period of time with no error messages reported. Using WARN
>as the rootCategory level everything works as expected. Note that we have been using
>log4j on the Barracuda project successfully for some time, but this is the first time
>that I have tried to turn on debug level logging for everything (this is in response
>to another Barracuda user reporting the same problem). Barracuda generates a huge
>amount of logging at debug or info levels, but log4j should be able to handle it,
>shouldn't it ? Note that I tried starting the app server with -Xms65m -Xmx256m and it
>made no difference.
>Anyone else seen a similar problem ?
>
>TIA
>
>Robert
>
>---------------------------------------------------------------------
>To unsubscribe, e-mail: [EMAIL PROTECTED]
>For additional commands, e-mail: [EMAIL PROTECTED]
>
--
Ceki Gülcü
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
