This Bugzilla instance is a read-only archive of historic NetBeans bug reports. To report a bug in NetBeans please follow the project's instructions for reporting issues.
Created attachment 101260 [details] My log file when netbeans took 100% CPU I have a JFrame that uses standard swing menu components and loads fine in Matisse. However, occasionally (every ~10 mins) when editing my JMenu items (renaming, adding, moving them, etc.) Netbeans will take up 100% of my CPU until I kill it (I've waited ~8 mins). I've attached my netbeans log which ends in a nice stack trace and a jconsole screenshot of the stack trace for the AWT thread (which didn't seem to change). I'm running JDK 1.6_20 and I am using the jide libraries, but not directly in the jmenu that I'm coding against. I can't attach 2 files, so here is the jconsole screenshot: http://yfrog.com/bcnetbeanslockedupp
Can you create a self-profiling snapshot for at least part of the 100% CPU period? http://wiki.netbeans.org/FaqProfileMeNow It may not be possible to start the self-profiling if AWT event queue is blocked by the 100% CPU operations. In such case, could you try jvisualvm from JDK, or at least take a few thread dumps? Thanks
Created attachment 101742 [details] threaddump taken using visualjvm once netbeans is taking up 100% cpu w/o responding
Here is your thread dump using visual jvm. I also have a heap dump if you think it would be helpful, it is 120mb though.
The thread dump shows the problem of wrong synchronization exposed by the additional property sheet thread. However, I don't see how it could be cycled and eat 100% CPU. The synchronization should be already fixed, see e.g. bug 186528. Could you verify with a current (development) daily build? Or use VisualVM to create a profiling snapshot (using the sampler) for a longer period of time when frozen (in 6.9)?