+- Discovery Gaming Community (https://discoverygc.com/forums)
+-- Forum: Welcome (https://discoverygc.com/forums/forumdisplay.php?fid=399)
+--- Forum: Help & Support (https://discoverygc.com/forums/forumdisplay.php?fid=26)
+--- Thread: Fatal Crash, Unknown Error Code (/showthread.php?tid=206189)
Fatal Crash, Unknown Error Code - Zyliath - 12-22-2024
![[Image: h0YO924.png]](https://imgur.com/h0YO924.png)
Was playing Disco on the server, chilling in Pennsylvania while mining heavy water.
Suddently, while in the middle of mining an asteroid, the game froze and crashed with the error above.
Can someone lend me a hand here?
RE: Fatal Crash, Unknown Error Code - Laz - 12-22-2024
@Aingar from what I can tell, this seems to be part of some recursive logic of a nested element.
Code:
while ( *(_WORD *)(this + 26) )
{
v6 = *(_DWORD **)(this + 28);
v7 = (void *)*v6;
if ( *v6 )
{
sub_635C310(*v6);
sub_6391D5A(v7);
}
}`this+26` looks particularly suspect. sub_6391D5A is a free on the first item of v6.
@Zyliath Can you send me (or perhaps @Aingar is the better choice) the dump file?
RE: Fatal Crash, Unknown Error Code - Zyliath - 12-22-2024
(12-22-2024, 07:35 PM)Laz Wrote: @Aingar from what I can tell, this seems to be part of some recursive logic of a nested element.
Code:while ( *(_WORD *)(this + 26) )
{
v6 = *(_DWORD **)(this + 28);
v7 = (void *)*v6;
if ( *v6 )
{
sub_635C310(*v6);
sub_6391D5A(v7);
}
}
`this+26` looks particularly suspect. sub_6391D5A is a free on the first item of v6.
@Zyliath Can you send me (or perhaps @Aingar is the better choice) the dump file?
Can do one better: I can send the dump file to both of you, see if 2 heads work better than one.
RE: Fatal Crash, Unknown Error Code - Aingar - 12-23-2024
The code crashes in the for loop above the segment you pasted Laz. Call stack for some reason shows nothing, but the stack itself The only thing I can tell is that this code is related to physics and collision states. Something must have went wrong when killing a collision group, but I can't fathom what went wrong as the system isn't reverse engineered at all.
Unless this crash is reliably reproducible I don't think we can fix this, pretty sure any attempts at ignoring the issue without crashing will simply end in a crash somewhere else (like it did for some other crash catch attempts we've made)