DreamSeeker Crash

ID:2520438
 
Nov 3 2019, 11:59 am (Edited on Nov 3 2019, 12:04 pm)
BYOND Version:513.1492
Operating System:Windows 10 Pro
Web Browser:Chrome 78.0.3904.70
Applies to:Dream Seeker
Status: Open
Issue hasn't been assigned a status value.
Descriptive Problem Summary:
DreamSeeker regularly crashes when moving the client mob to other z levels.

Numbered Steps to Reproduce Problem:
Move the client mob to other z levels.

Does the problem occur:
Every time? Or how often? Around 50% of the time.

In other games? Haven't tested.
In other user accounts? Yes.
On other computers? Yes.

When does the problem NOT occur?
It occurs nearly every play session inevitably.

Here are the crash logs for DreamSeeker if useful.

Crash dumps: Dump 1 | Dump 2

Nov 3 2019, 3:01 pm
Update. We reverted to 512.1402 and now receive:


Faulting application name: dreamseeker.exe, version: 5.0.512.1402, time stamp: 0x5a4ea2eb
Faulting module name: MSVCR120.dll, version: 12.0.21005.1, time stamp: 0x524f7ce6
Exception code: 0xc0000409
Fault offset: 0x000a7666
Faulting process id: 0x4d8c
Faulting application start time: 0x01d5928ee058acbe
Faulting application path: C:\Program Files (x86)\BYOND\bin\dreamseeker.exe
Faulting module path: C:\Program Files (x86)\BYOND\bin\MSVCR120.dll
Report Id: 087d54de-172a-4666-9841-165d0d8a3333
Faulting package full name:
Faulting package-relative application ID:
Nov 3 2019, 3:33 pm
What would you go to 512.1402 for? That's ridiculously out of date. 1488 is the current 512 build.

I need a test case for your issue so I can diagnose the problem. Obviously changing Z levels alone isn't the whole issue; there has to be something else in play.
Nov 4 2019, 10:09 am
I took a look at those files you included. Neither of them is a dump file; they're event viewer logs, which are a different animal.

The first one is not a crash at all, but a hang, and so only a proper dump file would be of any use to diagnose a hang. Hangs are notoriously tricky regardless and often need to be caught in the debugger.

The second file reports a crash in an icon destructor, which means there's been some heap corruption. But the crash occurs after the corruption has already happened, so there's no useful information there for me to work on.

I'm definitely going to need a test case to proceed.