|
Rank: Newbie Groups: Member
Joined: 4/1/2014 Posts: 4
|
I would like to report a problem regarding the EO.WebBrowser for .NET. We use it from ArchiCAD and the browser crashes randomly inside it. Sometimes it just stops working, the whole webpage inside it stops working. After reopen the browser plugin shows only a white screen. Sometimes after several minutes it also stops working without any reason. It is working properly and suddenly crashes. There are cases when I leave the computer and when I get back I see it has crashed. Nothing was running in the back, I've done nothing, just left the computer. We debugged our web application, and in the cases of these crashes it continues to work, we just can not reach it through the browser plugin. The crashes are totally free from our app's behavior. See the video (1:17): VIDEOPlease, suggest or provide some solution for this problem, in this current situation we can't use the plugin. Thank You in advance!
|
|
Rank: Administration Groups: Administration
Joined: 5/27/2007 Posts: 24,229
|
Hi, Please try to download the latest build from our download page and see if it fixes the problem. We have recently fixed a problem that might be the same as your problem: http://www.essentialobjects.com/forum/postst8452_EOWebBrowser-This-WebView-has-already-been-destroyed-issue.aspxIf the problem still occurs, please try to isolate the problem into a test project and then send us the test project. We will look into it as soon as we have the test project. See here for more information about submitting test project: http://www.essentialobjects.com/forum/test_project.aspxThanks!
|
|
Rank: Member Groups: Member
Joined: 8/5/2016 Posts: 17
|
This problem is still existing on newest Version of EO. (16.2.44)! We are using EO browser in a kiosk environment that runs 24hours a day. Sometimes we leave the computer and when we come back, browser engine has crashed and shows a white screen.
If we restart the engine, nothing gets rendered. After we restart the PC, browser comes back.
Today I installed the browser on a different computer (Intel Compute Stick) and it shows the white screen from the beginning. There's no way to make it render the page (content is loaded and present in the DOM - JS is executed too).
What I found out so far: If I set the option "DisableGPU" to false, everything gets rendered fine until I turn the option back on, then the engine crashes immediately.
|
|
Rank: Administration Groups: Administration
Joined: 5/27/2007 Posts: 24,229
|
Hi,
Do you mean setting DisableGPU to true? The default value for DisableGPU is false.
If that's the case, then it usually indicates that you have a problem with your display driver. In that case it is not something that we can fix on our end. As such you may want to keep that option as true.
Thanks!
|
|
Rank: Advanced Member Groups: Member
Joined: 7/21/2014 Posts: 130
|
We upgrade from 16.1.75.0 to the latest version of EO.WebBrowser and some of our customers are reporting to us the same "white screen" problem...
This problem seems to affect only computers on Windows 7, we don't have any error or something else : all the opened webbrowsers (same engine) just disappears and show a white screen instead.
|
|
Rank: Administration Groups: Administration
Joined: 5/27/2007 Posts: 24,229
|
Fabien wrote:We upgrade from 16.1.75.0 to the latest version of EO.WebBrowser and some of our customers are reporting to us the same "white screen" problem...
This problem seems to affect only computers on Windows 7, we don't have any error or something else : all the opened webbrowsers (same engine) just disappears and show a white screen instead. Please try to set this property to true and see if it resolves the issue for you: https://www.essentialobjects.com/doc/eo.base.runtime.enablelargeaddressspace.aspxMake sure you try this with the latest build. Thanks!
|
|
Rank: Advanced Member Groups: Member
Joined: 7/21/2014 Posts: 130
|
The EnableLargeAddressspace is already set to true.
We actually use the 16.2.37, we will try with the 16.2.44 and see... but the changelog does not leave much hope.
|
|
Rank: Administration Groups: Administration
Joined: 5/27/2007 Posts: 24,229
|
Hi, Please follow these steps and we will check if we see any crash report: https://www.essentialobjects.com/doc/common/crash_report.aspxIf we do not see any crash report, then it's probably because the process crashed very early. The most common cause for such problem is you have some third party AntiVirus/security software on that machine that incorrectly killed our process. If that's not the case, then we will need you to create a small test application to demonstrate the problem and also give us remote access to the system that demonstrates the problem. We may need to have Visual Studio installed on that system in order to get to the root of the problem. Thanks!
|
|
Rank: Advanced Member Groups: Member
Joined: 7/21/2014 Posts: 130
|
We've upgraded to 16.2.44 and problem still the same.
Application Id is 211_A26_AUSS.
We've downgraded to 16.1.75.0 until this issue is resolved
Thanks!
|
|
Rank: Member Groups: Member
Joined: 8/5/2016 Posts: 17
|
We've upgraded too and problem still exists.
|
|
Rank: Administration Groups: Administration
Joined: 5/27/2007 Posts: 24,229
|
Fabien wrote:We've upgraded to 16.2.44 and problem still the same.
Application Id is 211_A26_AUSS.
We've downgraded to 16.1.75.0 until this issue is resolved
Thanks! We do not see any crash report on our end. Can you try to set this property to true and see if it helps: https://www.essentialobjects.com/doc/eo.webbrowser.runtime.disablegpu.aspxThanks!
|
|
Rank: Advanced Member Groups: Member
Joined: 7/21/2014 Posts: 130
|
DisableGpu property as set to true and the app has been deploy on a customer PC.
Now expected his return ...
|
|
Rank: Member Groups: Member
Joined: 8/5/2016 Posts: 17
|
Hi guys, today I found another reason, why or WHEN the white screen appears (randomly).
I was writing a proxy that stands between browser and webserver in order to serve content even when browser's offline. What I noticed is, that if there's a "Content-Length" in the HTTP header, browser randomly closes the reading stream for this ressource too early, what means, that some data can't be send properly and then the white screen appears and engine crashes.
If I remove the "Content-Length" (it's legit not using it, because not mandatory) the problem disappears and I never see the white screen again. If I send the content length again, browser crashes again - it is 100% reproducible and happens mainly on larger files (> 800 kb).
This might help you...
|
|
Rank: Advanced Member Groups: Member
Joined: 7/21/2014 Posts: 130
|
With the property Disable Gpu = True, our users no longer seem to encounter the error
|
|
Rank: Member Groups: Member
Joined: 8/5/2016 Posts: 17
|
In my option this is not a proper solution, because almost every site requires gpu support to be rendered smooth. CSS animations without gpu acceleration look pretty bad....
|
|
Rank: Advanced Member Groups: Member
Joined: 7/21/2014 Posts: 130
|
@Touchway I Agree with you but at this time it seems to be the only option to avoid crash... Now, we waiting for a fix ;)
|
|
Rank: Administration Groups: Administration
Joined: 5/27/2007 Posts: 24,229
|
Touchway wrote:In my option this is not a proper solution, because almost every site requires gpu support to be rendered smooth. CSS animations without gpu acceleration look pretty bad....
Usually this occurs when there is a problem with the display card driver. In that case it is not possible for us to "fix" that problem. The only viable option is to avoid it by disabling the GPU. Thanks!
|
|
Rank: Advanced Member Groups: Member
Joined: 7/21/2014 Posts: 130
|
16.1.75.0 (and before) -> no issue 16.2.37+ -> some users reporting the "blank screen" issue
So may be the display card driver is the source of a problem BUT if we downgrade to 16.1.75 release this crach never happens...
|
|
Rank: Administration Groups: Administration
Joined: 5/27/2007 Posts: 24,229
|
Fabien wrote:16.1.75.0 (and before) -> no issue 16.2.37+ -> some users reporting the "blank screen" issue
So may be the display card driver is the source of a problem BUT if we downgrade to 16.1.75 release this crach never happens... We do not know why 16.1.75 works but 16.2.37 doesn't work. We do have many cases where the GPU has caused the crash. If you can give us remote access to the machine of that has the problem, we will be more than happy to investigate further.
|
|
Rank: Advanced Member Groups: Member
Joined: 7/21/2014 Posts: 130
|
Unfortunately I do not have the ability to give you access to our customers computers (privacy issue)
|
|