100.2.1 Application crash hiding and showing map

[ChrisSmith12]

I'm getting an application crash using 100.2.1 .Net in my WPF application. I have the map in a Telerik RadPane. 

The crash occurs when I cover/hide the map within my application and then uncover/show it. The crash happens every time on the first two test systems below but I cannot reproduce the crash on the third system. While I am unable to reproduce the crash on the third system, I am able to get the map into a state where it is continually updating but does not show anything.

Test Systems

1. Windows 10 x64 on Dell laptop

2. Windows 10 x64 virtual machine running in VMWare Fusion on Mac OS 10.13.3 host system.

3. Windows 7 x64 on white box laptop

Exception:

Unhandled exception at 0x00007FFBB72D4008 in OperatorConsole.exe: Microsoft C++ exception: Esri_runtimecore::Runtimecore_net::Dx_exception at memory location 0x000000AF882BC160. occurred

[AaronMurphy3]

Hi Morten, 

I tried setting it to Static and the error no longer appears. This is a temporary fix but we would much prefer to use Dynamic for the release.

Do you have an email address I could send my sample project too?

[AaronMurphy3]

Here is the output when I turned on Native code debugging and got it to crash a few different times:

First time:

Critical error detected c0000374
The thread 0x3a9c has exited with code 0 (0x0).
The thread 0x3aa4 has exited with code 0 (0x0).
First-chance exception at 0x773F9D71 (ntdll.dll) in ArcGisSample.exe: 0xC0000374: A heap has been corrupted (parameters: 0x7742D8D0).
The thread 0x36b8 has exited with code 0 (0x0).
The thread 0x3628 has exited with code 0 (0x0).
Unhandled exception at 0x773F9D71 (ntdll.dll) in ArcGisSample.exe: 0xC0000374: A heap has been corrupted (parameters: 0x7742D8D0).

Second time:

First-chance exception at 0x154381A7 (igc32.dll) in ArcGisSample.exe: 0xC0000005: Access violation reading location 0xBAADF00D.
Unhandled exception at 0x154381A7 (igc32.dll) in ArcGisSample.exe: 0xC0000005: Access violation reading location 0xBAADF00D.

Third time:

First-chance exception at 0x0F4B01E3 (wpfgfx_v0400.dll) in ArcGisSample.exe: 0xC0000005: Access violation writing location 0x3FF00000.
Unhandled exception at 0x0F4B01E3 (wpfgfx_v0400.dll) in ArcGisSample.exe: 0xC0000005: Access violation writing location 0x3FF00000.

Fourth time:

Critical error detected c0000374
First-chance exception at 0x773F9D71 (ntdll.dll) in ArcGisSample.exe: 0xC0000374: A heap has been corrupted (parameters: 0x7742D8D0).
Unhandled exception at 0x773F9D71 (ntdll.dll) in ArcGisSample.exe: 0xC0000374: A heap has been corrupted (parameters: 0x7742D8D0).

First-chance exception at 0x1A0F16E0 (igd10iumd32.dll) in ArcGisSample.exe: 0xC0000005: Access violation reading location 0xFEEEFEEE.
An unhandled exception of type 'System.AccessViolationException' occurred in Esri.ArcGISRuntime.dll
Additional information: Attempted to read or write protected memory. This is often an indication that other memory is corrupt.

Fifth time:

First-chance exception at 0x5F843AAF (d3d11.dll) in ArcGisSample.exe: 0xC0000005: Access violation reading location 0xFEEEFEFA.
An unhandled exception of type 'System.AccessViolationException' occurred in Esri.ArcGISRuntime.dll
Additional information: Attempted to read or write protected memory. This is often an indication that other memory is corrupt.

As you will likely notice they all appear to be similar (but not the same) exceptions. There were all retrieved by performing the exact same action every time.

I got some dump files but they are 477MB unfortunately.

[ChrisSmith12]

Morten,

  I sent a dump file yesterday via email. Did you receive it?

[MichaelBranscomb]

Hi Chris,

Yes, received the link to download the dump and downloaded. We've submitted an issue in the backlog for further investigation.

Thanks for your help 

Cheers

Mike

[AaronMurphy3 · Accepted solution]

After further testing I can confirm this bug was introduced in v100.2. I have tried all the recent versions and v100.1 appears to works fine.

[ChrisSmith12]

Aaron,

  Thanks for the confirmation. Any idea when this will get fixed?

[ChrisSmith12]

I can also confirm that setting all Graphics Overlays to Static rendering removes the application crash, however, I have been able to get the map into a bad state where the map is all white and continuously in a rendering state. It stayed that way for 5 minutes before I closed the application. I triggered the bad state by switching back and forth between tabs hiding and showing the map.

[MichaelBranscomb]

Hi Chris,

Thanks for the additional information. This issue is under investigation by one of the development team in the current sprint.

Cheers

Mike

[MichaelBranscomb]

Hi,

We have identified the problem and fixed this issue in the current internal build. Please can you retest in the next release (anticipated to ship late-Q2 / early-Q3)?

Cheers

Mike