Created attachment 92398 [details]
Thread dump (after kill -TERM but was similar before)
I tried to drag an editor tab to another position. The whole desktop (not just NB) froze. I had to kill -9 the IDE from a console login (kill -TERM did not work). JDK 6u17, Ubuntu 9.10 w/ Gnome, NB dev running w/ Ocean L&F.
Created attachment 92399 [details]
the same, the next time I tried to reorder tabs... guess I will just stop trying?
does it help if you uncheck 'drag window image' under tools - options - misc - appearance?
No, still freezes the same way.
Can't consider this random since it freezes every time I try to drag windows, it seems. Recent change in NB? In JDK? In X11? I don't know.
works for me in vboxed ubuntu 9.10, jdk 1.6.0_17, x64
Not surprising. I have no idea how to go about reproducing in a different environment.
See https://netbeans.org/bugzilla/show_bug.cgi?id=60142#c20 for more information on the cause of the bug.
I can reproduce using that test application. I run Ubuntu 9.10 (all updates), Gnome, JDK 6u17, a single 1400x1050 display, NVIDIA driver version 185.
We should get escalated. It's marked P4, which does not sound right for an unrecoverable freeze.
This bug completely prevents me from using D&D on my Ubuntu system unless I pass --nosplash. Probably the use of the early splash screen should just be disabled on X11 systems until the JDK bug is fixed and widely deployed.
OK. Let's disable this on Unixes: core-main#0d600a0c79a3
Integrated into 'main-golden', will be available in build *201001170201* on http://bits.netbeans.org/dev/nightly/ (upload may still be in progress)
User: Jaroslav Tulach <email@example.com>
Log: #178453: Reverting back to b7bd22f6ea9c: Don't use -splash until fixes on X11
The JDK bug is marked fixed in 6u21:
Is someone able to reproduce the problem on Linux with an older JDK?
I was trying the test application attached to issue 60142#c21 but I don't encounter the freeze so I cannot verify the fix with 6u21+.
(In reply to comment #12)
> Is someone able to reproduce the problem on Linux with an older JDK?
Did you try doing so after reverting 0d600a0c79a3?
> Did you try doing so after reverting 0d600a0c79a3?
I was only playing with the test application attached to the other bug (as linked above) which I assume is a self-contained test case.