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.
Did a GetLatestVersion. When it was finished, did a compile all. After that was finished, scrolled in the editor. Then NB froze. - JDK 1.4.1_01 - NB Q-Build NetBeansIDE-QBE200301150100-200301192142.zip - VSS 6.0c Attaching ThreadDump.
Created attachment 8709 [details] ThreadDump.
Hmm, it does not look it's really frozen. Isn't it a rather performance problem? Isn't garbage collector running? There is no deadlock in the thread dump you've attached.
Waited for 2 minutes (on a 1.6 GHz machine), then made the ThreadDump and then killed NB.
BTW, using -J-XX:+UseConcMarkSweepGC, so there should be no hangs at all from the GC.
Well, I have bad experience with UseConcMarkSweepGC. It works O.K. for some time, but if you have NetBeans running for the whole day and use them extensively, after some time it becomes very slow. It looks like it's frozen, but it's not. When I did exit, it took 20-30 minutes! I believe, that this is not a problem of NetBeans, because I've never had this problem without UseConcMarkSweepGC. This issue looks like the same problem. NetBeans are not frozen, so from the thread dump it's not apparent where the problem is. I'm resolving this as invalid, because it either is not a problem of NetBeans but JVM, or it is some kind of performance problem, that can not be detected from the attached thread dump.
Actually, NB was running over night, so maybe this is a problem with the JDK, maybe this one? http://developer.java.sun.com/developer/bugParade/bugs/4803 222.html
Yes, it looks like the described problem. It's actually a duplicate of http://developer.java.sun.com/developer/bugParade/bugs/4758307.html which suggests not to use the concurrent collector at all. The bug is in progess so hopefully they will fix it soon.
Problem is, without the Concurrent GC, NB is much less responsive.
Okay, then verifying as invalid bug.
*** Issue 30300 has been marked as a duplicate of this issue. ***
*** Issue 30137 has been marked as a duplicate of this issue. ***
*** Issue 29770 has been marked as a duplicate of this issue. ***
*** Issue 30904 has been marked as a duplicate of this issue. ***