Home › Developer Forums › Echo 3.x Platform (Echo3) › Development (3.x)

Vista+IE7 users, I need a quick favor (testing echo3)

tliebeck's picture
Thu, 01/22/2009 - 20:12 — tliebeck

Could you deploy the test application from the current version of Echo3 and run the "ghost" test, e..g.

http://localhost:8080/TestApp/app?ghost

I'd like to know if the window occasionally goes blank white for periods of time (while the test iteration# at the top of the screen is still incrementing). I'm not seeing this behavior in Windows XP with IE7 or IE6, but am seeing it on my Windows 7 box with IE8 running in "compatibility mode". (As far as I can tell, compatibility mode just uses the msie7 DLL, but I haven't researched it heavily).

This is an IE-specific bug, it's the same thing we saw a while back with the Extras rich text area component. In this case it seems to be some type of threading/rendering issue with IE that suddenly causes the height of the document to be set to 0. I have workaround code in place for it which is effective in IE6 and 7, but can be hit-or-miss in IE8 in compatibility mode.

‹ Exception during Client.processUpdates(): Result of expression 'this._node' [null] is not an object. Red Screen DropDownMenu ›
Mon, 01/26/2009 - 01:26 — tliebeck
tliebeck's picture

I'm a PC?

C'mon isn't anyone here insane enough to be running Windows Vista? :D

Mon, 01/26/2009 - 06:42 — a.schild
a.schild's picture

I am (But I have also

I am (But I have also ubuntu for real work ;) )

I will look at the test you need and come back.

Mon, 01/26/2009 - 06:51 — a.schild
a.schild's picture

It fails, but not where

It fails,

but not where you expected. Seems the war is incomplete, I did take it 2-3 minutes ago from the echo3go builds.

Here the Stacktrace:

HTTP Status 500 - 

--------------------------------------------------------------------------------

type Exception report

message 

description The server encountered an internal error () that prevented it from fulfilling this request.

exception 

javax.servlet.ServletException: Error instantiating servlet class nextapp.echo.testapp.interactive.InteractiveServlet
	org.apache.catalina.valves.ErrorReportValve.invoke(ErrorReportValve.java:102)
	org.apache.catalina.connector.CoyoteAdapter.service(CoyoteAdapter.java:286)
	org.apache.coyote.http11.Http11Processor.process(Http11Processor.java:845)
	org.apache.coyote.http11.Http11Protocol$Http11ConnectionHandler.process(Http11Protocol.java:583)
	org.apache.tomcat.util.net.JIoEndpoint$Worker.run(JIoEndpoint.java:447)
	java.lang.Thread.run(Unknown Source)


root cause 

java.lang.ExceptionInInitializerError
	sun.reflect.NativeConstructorAccessorImpl.newInstance0(Native Method)
	sun.reflect.NativeConstructorAccessorImpl.newInstance(Unknown Source)
	sun.reflect.DelegatingConstructorAccessorImpl.newInstance(Unknown Source)
	java.lang.reflect.Constructor.newInstance(Unknown Source)
	java.lang.Class.newInstance0(Unknown Source)
	java.lang.Class.newInstance(Unknown Source)
	org.apache.catalina.valves.ErrorReportValve.invoke(ErrorReportValve.java:102)
	org.apache.catalina.connector.CoyoteAdapter.service(CoyoteAdapter.java:286)
	org.apache.coyote.http11.Http11Processor.process(Http11Processor.java:845)
	org.apache.coyote.http11.Http11Protocol$Http11ConnectionHandler.process(Http11Protocol.java:583)
	org.apache.tomcat.util.net.JIoEndpoint$Worker.run(JIoEndpoint.java:447)
	java.lang.Thread.run(Unknown Source)


root cause 

nextapp.echo.webcontainer.util.Resource$ResourceException: Resource does not exist: "nextapp/echo/testapp/interactive/resource/css/Custom.css".
	nextapp.echo.webcontainer.util.Resource.getResource(Resource.java:101)
	nextapp.echo.webcontainer.util.Resource.getResourceAsString(Resource.java:71)
	nextapp.echo.webcontainer.service.StaticTextService.forResource(StaticTextService.java:58)
	nextapp.echo.testapp.interactive.InteractiveServlet.<clinit>(InteractiveServlet.java:56)
	sun.reflect.NativeConstructorAccessorImpl.newInstance0(Native Method)
	sun.reflect.NativeConstructorAccessorImpl.newInstance(Unknown Source)
	sun.reflect.DelegatingConstructorAccessorImpl.newInstance(Unknown Source)
	java.lang.reflect.Constructor.newInstance(Unknown Source)
	java.lang.Class.newInstance0(Unknown Source)
	java.lang.Class.newInstance(Unknown Source)
	org.apache.catalina.valves.ErrorReportValve.invoke(ErrorReportValve.java:102)
	org.apache.catalina.connector.CoyoteAdapter.service(CoyoteAdapter.java:286)
	org.apache.coyote.http11.Http11Processor.process(Http11Processor.java:845)
	org.apache.coyote.http11.Http11Protocol$Http11ConnectionHandler.process(Http11Protocol.java:583)
	org.apache.tomcat.util.net.JIoEndpoint$Worker.run(JIoEndpoint.java:447)
	java.lang.Thread.run(Unknown Source)


note The full stack trace of the root cause is available in the Apache Tomcat/6.0.18 logs.


--------------------------------------------------------------------------------

Apache Tomcat/6.0.18

Mon, 01/26/2009 - 07:08 — n.beekman

Same here.

Same here.

Mon, 01/26/2009 - 13:50 — golfman
golfman's picture

I only run it on my laptop

I only run it on my laptop which I don't use for development. I can try to get a dev environment set up on that and try it.

I'll have to prepare myself for lot's of "You are the administrator and a program that you started as the administrator wants to do something that requires administrator rights - are you sure you are still the administrator?" messages ;)

Step Ahead Software
Makers of Javelin Modeler+Coder for Java.
Automatic persistence of your object model using JDO and Hibernate.
http://stepaheadsoftware.com/products/javelin/javelin.htm

Mon, 01/26/2009 - 12:39 — tliebeck
tliebeck's picture

Doh!

Oops, the build was broken...nightlies have been re-run against SVN 1851, which fixes that CSS issue.

Mon, 01/26/2009 - 23:28 — a.schild
a.schild's picture

So far everything is OK

After ~30 minutes of running it is at iteration 29000 and it still looks ok. (Of course I did not watch all 29000 iterations)

The test PC is with Vista Professional german, IE7 fully patched and a intel dualcore cpu with 3.16GHz.

André

Tue, 01/27/2009 - 02:11 — a.schild
a.schild's picture

After about 3.5 hours of

After about 3.5 hours of running I'm at iteration 101000, and everything is displaying correctly.

One this I notice, is that the memory usage of iexplorer is slowly going up.
Directly after start of ghost, the process uses ~30MB, not at iteration 101000 the process has ~440MB allocated.

André

Tue, 01/27/2009 - 12:10 — tliebeck
tliebeck's picture

Thanks appreciate the

Thanks appreciate the report!

What you'd expect to see is it cycle for several iterations with a blank screen.

Try minimizing the browser and memory usage should go down dramatically. Apparently MS only does major garbage collection when the browser is minimized (I'm not sure on specifics of this, it personally makes absolutely no sense to me whatsoever). Additionally try going to about:blank and minimizing...memory usage drops to about 1MB on IE6/7 for me in such cases.

Tue, 06/23/2009 - 07:13 — a.schild
a.schild's picture

Not sure if it's related,

Not sure if it's related, but seems also to be a nasty one:

http://www.webdevbros.net/2006/11/26/ie7-bug-blank-page-white-screen-after-some-refreshs/

and

http://support.microsoft.com/?scid=kb%3Ben-us%3B933006

Looks like version 7.0.6000.20544 of mshtml.dll would fix the problem.
But so far only version 7.0.6000.16850 is available via windows update, getting 7.0.6000.20544 involves asking MS for the hotfix.
(Of course the fix is available since 28-Feb-2007 16:13, but so far MS seems not to wish to release it for Windows XP, it should be included in SP2 for Vista)

I did not see any blank screens in IE with the last beta releases so far.

André