Use jmap -histo <pid> or a memory profiler (yourkit) or visualvm to look at the heap. It also helps to use jstat -gc <pid> 1000 in cases when you have low heap availability before killing your darlings. It might not be a memory leak but possibly a connection pool running out of connections.
Any number of programming errors can lead to memory leaks, including exceptions. Without actual profiling you won't be able to find the leaks. Martijn On Mon, Jan 18, 2010 at 4:14 PM, Frank Silbermann <frank.silberm...@fedex.com> wrote: > I am monitoring a Wicket 1.4.5 application running on Tomcat 6.0 which > accesses a SQL Server 2000 database. Periodically the application > becomes unresponsive due to a lack of heap space, and I have to bounce > Tomcat. I'm trying to figure out what sort of errors could cause this > to happen. > > Any suggestions? > > I've looked in the Tomcat error logs, but the only errors I see prior to > the out-of-heapspace error are a small number of SQL errors that result > from bad input data. The Wicket application handles these errors by > switching the user to a standard error page. I suppose it's a bad > architecture to rely on SQL errors reported by the database rather than > checking the data, but does this result in a memory leak? > > Frank Silbermann > > --------------------------------------------------------------------- > To unsubscribe, e-mail: users-unsubscr...@wicket.apache.org > For additional commands, e-mail: users-h...@wicket.apache.org > > -- Become a Wicket expert, learn from the best: http://wicketinaction.com Apache Wicket 1.4 increases type safety for web applications Get it now: http://www.apache.org/dyn/closer.cgi/wicket/1.4.4 --------------------------------------------------------------------- To unsubscribe, e-mail: users-unsubscr...@wicket.apache.org For additional commands, e-mail: users-h...@wicket.apache.org
