Test_org_eclipse_swt_browser_Browser.test_setUrl_remote_with_post fails on GHA Windows build

[iloveeclipse]

Looks like change of Gihub Windows 2025 runner also broke Test_org_eclipse_swt_browser_Browser.test_setUrl_remote_with_post test on Windows (see similar issue with keys #2516).

Seen on

• Add diagnostics to failing Clipboard test and skip on non-GTK #2555,
• Fix warning about using static directly in tests #2560

Stack from https://github.com/eclipse-platform/eclipse.platform.swt/runs/51571267543

Test timed out. ProgressListener not fired 
Test log:
checkInternet() passedSetting Browser url to:https://bugs.eclipse.org/bugs/buglist.cgi
java.lang.AssertionError: 
Test timed out. ProgressListener not fired 
Test log:
checkInternet() passedSetting Browser url to:https://bugs.eclipse.org/bugs/buglist.cgi
	at org.junit.Assert.fail(Assert.java:89)
	at org.junit.Assert.assertTrue(Assert.java:42)
	at org.eclipse.swt.tests.junit.Test_org_eclipse_swt_browser_Browser.test_setUrl_remote_with_post(Test_org_eclipse_swt_browser_Browser.java:1251)
	at java.base/java.lang.reflect.Method.invoke(Method.java:580)
	Suppressed: java.lang.Throwable: Screenshot written to C:\Users\RUNNER~1\AppData\Local\Temp\org.eclipse.swt.tests.junit.Test_org_eclipse_swt_browser_Browser.test_setUrl_remote_with_post[browser flags: 524,288].png
		at org.eclipse.test.Screenshots$ScreenshotOnFailure.failed(Screenshots.java:41)
Running Test_org_eclipse_swt_browser_Browser#test_setUrl_remote_with_post[browser flags: 524,288]
AWT screenshot saved to: C:\Users\RUNNER~1\AppData\Local\Temp\org.eclipse.swt.tests.junit.Test_org_eclipse_swt_browser_Browser.test_setUrl_remote_with_post[browser flags: 524,288].png

[iloveeclipse]

@rgrunber , @akurtakov : the test introduced via e381d20 uses bugzilla for "post" requests. I guess bugzilla will not support that soon? Do you have an idea which other server at eclipse.org can be "misused" for testing post requests?

Should we disable the test method now for Windows on Github only (similar to #2559) or disable itt completely?

[akurtakov]

The test is valuable so ideally shouldn't be just disabled. Ideally, the test should start it's own httpserver (https://docs.oracle.com/en/java/javase/21/docs/api/jdk.httpserver/com/sun/net/httpserver/HttpServer.html ? ) and use it for testing purposes.
I have no clue about Windows and will defer that part to @HeikoKlare .

[rgrunber]

Maybe www.eclipse.org, based on https://bugs.eclipse.org/bugs/show_bug.cgi?id=506001#c4 ? Although I wouldn't be surprised if that stopped working as well.

[HeikoKlare]

My best guess is that this failure is not related to the change to Windows Server 2025 in GHA. I expect this to be a long-standing flaky test like several other tests executed on Edge/WebView2 which all seem to be related to progress listeners, see e.g.:

• Browser tests failing on windows #2363
• Test_org_eclipse_swt_browser_Browser.test_LocationListener_then_ProgressListener fails randomly #2098
• Test_org_eclipse_swt_browser_Browser.test_ProgressListener_completed_Called failed randomly #2113
• ...

We had already invested quite some time in finding the cause of this issue, without any success so far. Fortunately, the failures do not appear too often (though they are still very annoying).