Fatal Memory Error DAbort 0xF50EFD7B

[FKhueson]

Hi, I'm encountering a fatal memory error with the following debug output:

9/26/2022 10:41:01.059 AM
DAbort 0xF50EFD7B: 
c:\penguin\labview\components\mgcore\trunk\20.0\source\MemoryManager.cpp(1340) : DAbort 0xF50EFD7B: 
minidump id: 8af6fc53-c0a8-4ad8-a86e-1c8b8e613956
$Id: //labview/components/mgcore/trunk/20.0/source/MemoryManager.cpp#12 $

Attached is the error zip file.  I'm having trouble tracing this back to something in my program.  The penguin directory above doesn't exist, so this seems like something internal to LabVIEW.  I make use of DSC module for logging data from shared variables, but have been doing this for a long time and hasn't been a problem before, so it's not just the use of DSC that could be causing this.  If anyone could help me parse through this and figure out what component might be a problem, that'd be appreciated.  Thanks!

[wiebe@CARYA]

Note that your email is in 0830192b-1642-47d6-b95d-bf8ae146d414.txt.

The file doesn't really add anything (so (next time) remove it).

[wiebe@CARYA]

It's probably some DLL call that corrupts memory.

It is memory related for sure...

If you can't reproduce it to catch the crash in action, it will be very hard to find the problem.

DAbort 0x37C03D (the error in the file)

DAbort 0x37C03D

Solved: Re: TestStand 2016 Crash DAbort 0x0037C03D in MemoryManager.cpp - NI Community

Crash "DAbort 0x37C03D in MemoryManager.cpp" when calling DLL with "Call Library Function Node" - NI...

Error Message DAbort 0xF50EFD7B in MemoryManager.cpp - NI

DAbort 0xF50EFD7B in MemoryManager.cpp Using NI-DCPower in TestStand - NI

Fatal Internal Error 0xF50EFD7B: MemoryManager.cpp, line 1187 - NI Community

[FKhueson]

Thanks, been perusing those links but unfortunately I'm not using the particular modules that they show to be problematic, and I don't make any calls to DLLs at that point.  Trying to parse through pieces of my code to try to isolate things, this has been a finicky thing to nail down.  The error doesn't always happen with the same actions, so I'm starting to believe it might be some shared variable that's being read and written to at the same time when operating in the background.  Will continue investigating.

[wiebe@CARYA · Accepted solution]

I think DSC functions are mostly dll calls.

A race condition or simultaneous access to a shared variable seems like a potential cause.

I'd try to make a stress test. Make a VI that executes scenarios as fast as possible, try to reproduce the problem faster.

[FKhueson]

To update, I put a semaphore in the loops that accessed shared variables to limit access to one process, and the fatal memory error hasn't showed up since.  Thank you!