OK. Last post but just in case somebody hits this in the future. I have a
simple unit test (below) that fails about 50% of the time... However it
doesn't fail in 2.5-rc1 so as it is so obscure anyway I won't raise a issue.
Sorry for the spam!
package com.google.gwt.sample.dynatablerf.client;
import com.google.gwt.junit.client.GWTTestCase;
import com.google.gwt.user.client.Timer;
public class GwtTestStall extends GWTTestCase {
@Override
public String getModuleName() {
return "com.google.gwt.sample.dynatablerf.DynaTableRf";
}
public void testStall() {
this.delayTestFinish(2 *60 * 1000);
Timer slowJob = new Timer() {
@Override
public void run() {
System.out.println("Slow job: done");
}
};
System.out.println("Slow job: start");
slowJob.schedule(3000);
Timer repeatingJob = new Timer() {
int done;
@Override
public void run() {
System.out.println("timer " + done);
if (done < 100) {
done++;
schedule(100);
} else {
System.out.println("Repeating job: done");
finishTest();
}
}
};
System.out.println("Repeating job: start");
repeatingJob.schedule(1000);
}
}
On Wednesday, July 11, 2012 1:17:48 PM UTC+1, salk31 wrote:
>
> Hi Thomas,
>
> It is called at the loading of the page but the HtmlUnitThread only runs
> while that call is blocking...
>
> I'm struggling to point the finger at a particular line of code but:
> * Fiddling with the number of seconds GWT specifies to HTMLUnit can make
> the problem appear and disappear (a colleague has duplicated but I had to
> mess with Maven dependencies so not as reliable at test as wanted)
> * I noticed that when a test stalls the HTMLUnit JavaScript thread was
> "gone".
>
> I'll try and write a unit test before raising an issue but is not easy as
> seems non-deterministic.
>
> btw Congratulations on your new role in the steering committee. Bit scary
> as the code as been sooooooooo good so far but fingers crossed.
>
> Cheers
>
> Sam
>
> On Wednesday, July 11, 2012 10:34:40 AM UTC+1, Thomas Broyer wrote:
>>
>>
>>
>> On Wednesday, July 11, 2012 9:21:10 AM UTC+2, salk31 wrote:
>>>
>>> Sorry for talking to myself but...
>>>
>>> waitForBackgroundJavaScriptStartingBefore looks like a possible cause.
>>> It is called here:
>>>
>>> http://code.google.com/p/google-web-toolkit/source/browse/trunk/user/src/com/google/gwt/junit/RunStyleHtmlUnit.java#158
>>>
>>> with a value of 2s the implementation of this method seems a bit odd:
>>>
>>> http://grepcode.com/file/repo1.maven.org/maven2/net.sourceforge.htmlunit/htmlunit/2.9/com/gargoylesoftware/htmlunit/WebClient.java#WebClient.waitForBackgroundJavaScriptStartingBefore%28long%29
>>>
>>> It seems to block while the number of jobs is different to the number of
>>> jobs after the time passed to it...
>>>
>>> This time starts _before_ servlet startup so in our case that could
>>> easily be longer than 2s....
>>>
>>> Just me or both these bits of code a bit dodgy? Maybe gwt should pass
>>> the unit test timeout value to htmlunit or just some huge value?
>>>>
>>>>
>>>>
>> IIUC, that code is called when loading the page, to wait for the GWT code
>> to launch, but before tests are executed. I doubt it's the culprit.
>> BTW, have you set the logs to SPAM level? that could help find where
>> things hang. Also maybe a threaddump (not sure I could help, but pretty
>> sure the people knowing the internals better than me would find them useful)
>>
>
--
You received this message because you are subscribed to the Google Groups
"Google Web Toolkit" group.
To view this discussion on the web visit
https://groups.google.com/d/msg/google-web-toolkit/-/zS1vaa6miXEJ.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to
[email protected].
For more options, visit this group at
http://groups.google.com/group/google-web-toolkit?hl=en.
