<AWT Dev> [14] RFR JDK-8234385:[TESTBUG] java/awt/EventQueue/6980209/bug6980209.java fails in linux nightly

[Prasanta Sadhukhan]

Hi Sergey,

I have ran for 50 iteration per platform with my change and mach5 is 
green (link is 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
>>
>
>