jakarta-jcs-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Matthew Cooke <mpcoo...@lineone.net>
Subject Re: Additional problems.
Date Thu, 29 Apr 2004 11:15:15 GMT

I've profiled our production Remote JCS cache and one in the office. I 
haven't quite got to the root of the problem but it may be something to 
do with Log4j.
It may well be that JCS jar I built can't find it's log4j file or that 
debugging output is somehow collecting in memory, this doesn't happen 
when I run it locally in our dev enviroment and profile it.

After a few minutes jprofiler shows 1,568,647 Stack trace elements (even 
after a garbage collect).
Compared with only 3,672 hashmap entries.

I can't imagine even in the worst case log4j should try and keep all 
this stuff in memory, but it does look like log4j is doing something weird.


>Any more detail on the memory issue would be appreciated.
>To unsubscribe, e-mail: turbine-jcs-user-unsubscribe@jakarta.apache.org
>For additional commands, e-mail: turbine-jcs-user-help@jakarta.apache.org

To unsubscribe, e-mail: turbine-jcs-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: turbine-jcs-user-help@jakarta.apache.org

View raw message