Operating System = Mac OS X version 10.7.4 running on x86_64
Java; VM; Vendor = 1.7.0_08-ea; Java HotSpot(TM) 64-Bit Server VM 23.4-b01; Oracle Corporation
Runtime = Java(TM) SE Runtime Environment 1.7.0_08-ea-b05

Reported by a NetBeans user - At some point the IDE displays the beachball of death and needs to be killed from the console or the activity monitor. Thread dump after this happens is attached; it is stuck in sun.lwawt.macosx.CCursorManager.nativeGetCursorPosition(Native Method).

When this happened the console contained the following log messages:

Sep 4 14:58:29 dhcp-prague08-second-floor-10-163-21-238.cz.oracle.com
java[89995] <Error>: kCGErrorRangeCheck: CGSSetSurfaceShape
Sep 4 14:58:29 dhcp-prague08-second-floor-10-163-21-238.cz.oracle.com
java[89995] <Error>: kCGErrorFailure: Set a breakpoint @ CGErrorBreakpoint() to
catch errors as they are logged.
2012-09-04 14:58:29.073 java[89995:707] CGSSetSurfaceShape failed - error 1007
(window:34399, surface:375770944, bounds:{{0, 0}, {2147483647, 2608}}
2012-09-04 14:58:29.084 java[89995:707] *** -[NSMutableArray
initWithCapacity:]: Request for large capacity 25165824
WA

Related NetBeans bug:
http://netbeans.org/bugzilla/show_bug.cgi?id=217817