+1
On 11/20/19 9:37 pm, Prasanta Sadhukhan wrote:
On 21-Nov-19 4:34 AM, Sergey Bylokhov wrote:
On 11/20/19 1:41 am, Prasanta Sadhukhan wrote:
Hi Sergey,
I have ran for 50 iteration per platform with my change and mach5 is green
(link is in JBS).
And it will be red w/o the fix?
Yes, at least 1 of the iteration will fail. mach5 link in JBS.
Regards
Prasanta
On 19-Nov-19 11:38 PM, Sergey Bylokhov wrote:
Hi, Prasanta.
I have checked the logs for this test on different platforms and it seems we
have 3 issues here:
- RealSync does not work properly and produce next exception:
sun.awt.SunToolkit$OperationTimedOut: 20000
at java.desktop/sun.awt.X11.XToolkit.syncNativeQueue(XToolkit.java:2645)
at java.desktop/sun.awt.SunToolkit.realSync(SunToolkit.java:1470)
at java.desktop/sun.awt.SunToolkit.realSync(SunToolkit.java:1402)
at java.desktop/java.awt.Robot.waitForIdle(Robot.java:675)
at bug6980209.main(bug6980209.java:72)
at java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke0(Native
Method)
at
java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
at
java.base/jdk.internal.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
at java.base/java.lang.reflect.Method.invoke(Method.java:564)
at
com.sun.javatest.regtest.agent.MainWrapper$MainThread.run(MainWrapper.java:127)
at java.base/java.lang.Thread.run(Thread.java:830)
- Exception reported in JDK-8198615, "It seems the secondary loop will never
end"
- Timeout in the realSync(), is it possible that we need to increase the
default timeout, or
we have a bug in realSync()?
BTW to check the intermittent test you need to run it more than 3 times(I guess
50 pre platform is fine).
On 11/19/19 1:42 am, Prasanta Sadhukhan wrote:
Hi All,
Please review a testbug fix for an issue seen intermittently in linux nightly.
Issue is that the test times out after the test finishes without actually
propagating the test results to jtreg harness. It seems like the JFrame which
is created is called with EXIT_ON_CLOSE which will call System.exit(0) causing
the vm to terminate and if the test is not run with othervm option, it might
create some issue when run in a bunch, as it is done in nightly testing.
Proposed fix is to remove the EXIT_ON_CLOSE option as we already have
frame.dispose() call. mach5 run is green.
Bug: https://bugs.openjdk.java.net/browse/JDK-8234385
diff -r 9db62a092725 test/jdk/java/awt/EventQueue/6980209/bug6980209.java
--- a/test/jdk/java/awt/EventQueue/6980209/bug6980209.java Thu Nov 14 10:59:45
2019 +0530
+++ b/test/jdk/java/awt/EventQueue/6980209/bug6980209.java Tue Nov 19 15:06:03
2019 +0530
@@ -63,7 +63,6 @@
public void run() {
frame = new JFrame();
frame.setUndecorated(true);
*- frame.setDefaultCloseOperation(JFrame.EXIT_ON_CLOSE);*
setup(frame);
}
});
Regards
Prasanta
--
Best regards, Sergey.
