by Peke » Tue Feb 03, 2026 12:05 pm
Dryst wrote: ↑Tue Feb 03, 2026 12:13 am
I asked Poe AI why this keeps happening over and over. Here are the responses I got:
Can you ask Poe AI why this is not happening to 100k other users and only to you, and list potential cases that are specific to your enviroment.
Dryst wrote: ↑Tue Feb 03, 2026 12:13 am
A native heap corruption bug in MediaMonkey’s process, most likely involving FastMM interacting with third‑party native code under heavy multithreading.
This is generic answer, "ntdll.dll" is essential system kernel low level code.
Taken FROM MS Help
Code: Select all
The error code 0xc0000374 typically indicates a heap corruption issue, and it's a bit generic.
Here are a few steps you can take to troubleshoot and potentially resolve the issue:
Check for Windows and .NET Updates: Ensure that your Windows operating system and .NET Framework are up-to-date with the latest patches and updates. Microsoft frequently releases updates to address known issues and improve stability.
Review Recent Changes: Since the issue started after a specific date, review any changes made to your application or the environment around that time. This includes updates to dependencies, libraries, or any modifications to the code.
Analyze Crash Dumps: If possible, analyze crash dumps generated during these crashes. Tools like WinDbg can help you analyze memory dumps and identify the root cause of crashes. Look for any patterns or specific code paths leading to the crash.
Check Event Viewer: The information you provided is from the Event Viewer. Check for additional information in the Application or System logs. There might be more details or specific error messages that can help diagnose the issue.
Memory Issues: Heap corruption often indicates a problem with memory. Check for any memory-related issues in your application. Ensure proper memory management, avoid buffer overflows, and consider using memory analysis tools.
Update .NET Framework Version: If possible, consider updating your project to a more recent version of .NET Framework. This might include migrating to .NET Core or .NET 5/6, as older versions may have known issues that have been addressed in newer releases.
Please note that MM was not written in .NET but in Embarcadero Delphi and C++ so it is most likely not caused by MM and FastMM that is designed to do lighting fast Memory management and preventing MM memory interruption by outside factors as a safeguard from poking into MM internal functions and changing behavior.
Code: Select all
The error code 0xc0000374 indicates a heap corruption issue, which can be caused by various factors, including:
Incorrect memory access or management: This can occur due to bugs in the application code, libraries, or dependencies.
Outdated or corrupted system files: This can happen if the operating system or .NET Framework is not up-to-date with the latest patches and updates.
Malware or viruses: Malicious software can cause heap corruption and crashes.
Hardware issues: Faulty RAM or other hardware problems can lead to heap corruption.
To troubleshoot and potentially resolve the issue, consider the following steps:
Update your system and drivers: Ensure that your operating system, .NET Framework, and drivers are up-to-date with the latest patches and updates.
Run system file checks: Use tools like sfc /scannow and DISM to repair corrupted system files.
Reinstall problematic apps: If the issue is related to a specific application, try reinstalling it.
Perform a malware scan: Use antivirus software to scan your system for malware and viruses.
Use a memory diagnostic tool: Tools like MemTest86+ can help identify hardware issues related to RAM.
Additionally, consider the following:
Review recent changes: If the issue started after a specific date, review any changes made to your application or environment around that time.
Analyze crash dumps: If possible, analyze crash dumps generated during these crashes to identify patterns or specific code paths leading to the crash.
Check Event Viewer: Look for additional information in the Event Viewer, such as error messages or warnings related to the crash.
Please people no matter how much AI feels safer option, we at MM team work on MM 15+ years and do not disclose all our internal work to any AI, nor we have written code with AI, we are here to help and improve MM and your experience.
Just recently I had user that corrupted MM library by following outdated information and resolutions supplied by AI which ended in several hours of my free time to direct help user and hours of correcting and rebuilding MM5.DB that would all be avoided if user contacted us directly.
I can talk in my own name, but I hope others in team will agree that each time someone report crash/issue with MM I firstly assume it is our fault and bug. Mainly because if there is something we made wrong then it can be fixed (we are paid by your trust in buying MM, to make it work and do the best we can do to fix), as because if it is specific case then it is on our conscience if we want to fix other ones issues and do free work for them. Yeah MM broke, but just like you can't blame car manufacturer if you caused accident by failing to stop in time because you used aftermarket custom disk brakes.
I hate to see how AI is becoming so confident that his answer is correct. Yes in most cases is because it gives 20 answers where one is usually correct, but never consider that 1 of other 19 can catastrophically worsen the issue.
Dryst wrote: ↑Tue Feb 03, 2026 12:13 am
MediaMonkey’s age and design make this class of bug extremely expensive to eliminate.
This is orthodox lie. MM2024 is 3 years old and <10 reports of such/similar cases are reported so ... it is 0.0001% of MM users. Such bugs in MM Code are easily fixable when Error points to MM code not Windows System file.
Dryst wrote: ↑Tue Feb 03, 2026 12:13 am
Option 1 — Enable FastMM full debug mode (best diagnostic)
MediaMonkey can be forced to use:
Full FastMM debug
Guard blocks
Memory leak reporting
This often causes:
The crash to happen earlier
With a clear error message
Instead of ntdll.dll
It also produces logs developers can act on.
This is true, but assume that MM dev teams have MS tools to to debug internal ntdll.dll Microsoft messges.
That is why @lowlander asked for full MM Debug log that reference on exact MM state, what is going on in MM code that like AI said gives Developewr logs we can act on.
[quote=Dryst post_id=531691 time=1770095609 user_id=116831]
I asked Poe AI why this keeps happening over and over. Here are the responses I got:
[/quote]Can you ask Poe AI why this is not happening to 100k other users and only to you, and list potential cases that are specific to your enviroment.
[quote=Dryst post_id=531691 time=1770095609 user_id=116831]
A native heap corruption bug in MediaMonkey’s process, most likely involving FastMM interacting with third‑party native code under heavy multithreading.[/quote]This is generic answer, "ntdll.dll" is essential system kernel low level code.
Taken FROM MS Help
[code]The error code 0xc0000374 typically indicates a heap corruption issue, and it's a bit generic.
Here are a few steps you can take to troubleshoot and potentially resolve the issue:
Check for Windows and .NET Updates: Ensure that your Windows operating system and .NET Framework are up-to-date with the latest patches and updates. Microsoft frequently releases updates to address known issues and improve stability.
Review Recent Changes: Since the issue started after a specific date, review any changes made to your application or the environment around that time. This includes updates to dependencies, libraries, or any modifications to the code.
Analyze Crash Dumps: If possible, analyze crash dumps generated during these crashes. Tools like WinDbg can help you analyze memory dumps and identify the root cause of crashes. Look for any patterns or specific code paths leading to the crash.
Check Event Viewer: The information you provided is from the Event Viewer. Check for additional information in the Application or System logs. There might be more details or specific error messages that can help diagnose the issue.
Memory Issues: Heap corruption often indicates a problem with memory. Check for any memory-related issues in your application. Ensure proper memory management, avoid buffer overflows, and consider using memory analysis tools.
Update .NET Framework Version: If possible, consider updating your project to a more recent version of .NET Framework. This might include migrating to .NET Core or .NET 5/6, as older versions may have known issues that have been addressed in newer releases.
[/code]
Please note that MM was not written in .NET but in Embarcadero Delphi and C++ so it is most likely not caused by MM and FastMM that is designed to do lighting fast Memory management and preventing MM memory interruption by outside factors as a safeguard from poking into MM internal functions and changing behavior.
[code]
The error code 0xc0000374 indicates a heap corruption issue, which can be caused by various factors, including:
Incorrect memory access or management: This can occur due to bugs in the application code, libraries, or dependencies.
Outdated or corrupted system files: This can happen if the operating system or .NET Framework is not up-to-date with the latest patches and updates.
Malware or viruses: Malicious software can cause heap corruption and crashes.
Hardware issues: Faulty RAM or other hardware problems can lead to heap corruption.
To troubleshoot and potentially resolve the issue, consider the following steps:
Update your system and drivers: Ensure that your operating system, .NET Framework, and drivers are up-to-date with the latest patches and updates.
Run system file checks: Use tools like sfc /scannow and DISM to repair corrupted system files.
Reinstall problematic apps: If the issue is related to a specific application, try reinstalling it.
Perform a malware scan: Use antivirus software to scan your system for malware and viruses.
Use a memory diagnostic tool: Tools like MemTest86+ can help identify hardware issues related to RAM.
Additionally, consider the following:
Review recent changes: If the issue started after a specific date, review any changes made to your application or environment around that time.
Analyze crash dumps: If possible, analyze crash dumps generated during these crashes to identify patterns or specific code paths leading to the crash.
Check Event Viewer: Look for additional information in the Event Viewer, such as error messages or warnings related to the crash.[/code]
Please people no matter how much AI feels safer option, we at MM team work on MM 15+ years and do not disclose all our internal work to any AI, nor we have written code with AI, we are here to help and improve MM and your experience.
Just recently I had user that corrupted MM library by following outdated information and resolutions supplied by AI which ended in several hours of my free time to direct help user and hours of correcting and rebuilding MM5.DB that would all be avoided if user contacted us directly.
I can talk in my own name, but I hope others in team will agree that each time someone report crash/issue with MM I firstly assume it is our fault and bug. Mainly because if there is something we made wrong then it can be fixed (we are paid by your trust in buying MM, to make it work and do the best we can do to fix), as because if it is specific case then it is on our conscience if we want to fix other ones issues and do free work for them. Yeah MM broke, but just like you can't blame car manufacturer if you caused accident by failing to stop in time because you used aftermarket custom disk brakes.
I hate to see how AI is becoming so confident that his answer is correct. Yes in most cases is because it gives 20 answers where one is usually correct, but never consider that 1 of other 19 can catastrophically worsen the issue.
[quote=Dryst post_id=531691 time=1770095609 user_id=116831]
MediaMonkey’s age and design make this class of bug extremely expensive to eliminate.
[/quote]This is orthodox lie. MM2024 is 3 years old and <10 reports of such/similar cases are reported so ... it is 0.0001% of MM users. Such bugs in MM Code are easily fixable when Error points to MM code not Windows System file.
[quote=Dryst post_id=531691 time=1770095609 user_id=116831]
Option 1 — Enable FastMM full debug mode (best diagnostic)
MediaMonkey can be forced to use:
Full FastMM debug
Guard blocks
Memory leak reporting
This often causes:
The crash to happen earlier
With a clear error message
Instead of ntdll.dll
It also produces logs developers can act on.
[/quote]This is true, but assume that MM dev teams have MS tools to to debug internal ntdll.dll Microsoft messges.
That is why @lowlander asked for full MM Debug log that reference on exact MM state, what is going on in MM code that like AI said gives Developewr logs we can act on.