-
Bug
-
Resolution: Fixed
-
P3
-
7, 8
-
generic
-
generic
Test runtime/threads/IntSleep still timesout due to the same issue described in CR 6921060.
With big timeout inserted before interrupt calls test passed.
I was able to reproduce issue only on some Windows hosts (stt-39,stt-67.ru.oracle.com).
Test uses synchronization based on thread state. That way is error-prone.
Synchronization algorithm for this test should be adjusted to avoid such hangs.
Log example:
http://vice.russia.sun.com/results/1.6.0_33/b03/vm/win2008-x64-64_vm.runtime.testlist/ResultDir/IntSleep/IntSleep.log
With big timeout inserted before interrupt calls test passed.
I was able to reproduce issue only on some Windows hosts (stt-39,stt-67.ru.oracle.com).
Test uses synchronization based on thread state. That way is error-prone.
Synchronization algorithm for this test should be adjusted to avoid such hangs.
Log example:
http://vice.russia.sun.com/results/1.6.0_33/b03/vm/win2008-x64-64_vm.runtime.testlist/ResultDir/IntSleep/IntSleep.log
- relates to
-
JDK-6546236 Thread interrupt() of Thread.sleep() can be lost on Solaris due to race with signal handler
-
- Closed
-