Hi Sebastien, What exactly is the issue you experience ?
The Log4J problem is that it doesn't release all classes/class loaders at undeploy. But if you run the application for N weeks without undeploy+deploy then it should not lead to any problems. Restarting the whole web server is also fine. So, what exactly happens with Wicket 8.x after few weeks ? The application fails with OutOfMemoryError ? If this is the case then the way to debug it is to take heap dumps and analyze what's eating the memory. Start the application and take a heap dump. Then wait for N days and take another one. If you start the server with -XX:+HeapDumpOnOutOfMemoryError then it will dump it automatically when the OOM occurs. Once you have the dumps you can compare them with VisualVM or analyze deeply with Eclipse MAT. if you upload the dumps somewhere I can analyze them for you ;-) Martin On Tue, Jan 23, 2018 at 11:30 AM, Sebastien <seb...@gmail.com> wrote: > Just to let you know, I've deployed the wicket7.x version of the Wicket > jQuery UI demo webapp on my tomcat server about 2 weeks ago... Everything > is working fine, no leak at all so far! > So I think it's safe to say the leak is coming from wicket 8.x. I saw the > comments about the log4j issue and the PageStore issue (but this one was > reported for wildfly IIRC). > Was the log4j issue affecting the wicket8.x branch only? Is there anything > I can test these days on my wicket8.x version? > > Thanks in advance :) > Sebastien. >
