TCMD Command Input?

[vefatica]

When I start TCMD (show Command Input configured) it starts looking like this (not docked).

If I click or hover over its tab (?) it drops down, partially covering the TCC tab's client area.

If I then click in the TCC tab client area, the Command Input window rolls up and shortly thereafter TCMD crashes (disappears). I have dump files.

 

[vefatica]

It might even crash twice. I get, for example,

2024-04-20  17:57     321,186,020  tcmd.exe.11636.dmp
2024-04-20  17:57     321,181,676  tcmd.exe(1).11636.dmp

 

[rconn]

If you have crash dumps, then please post them or send them to [email protected].

 

[rconn]

When I start TCMD (show Command Input configured) it starts looking like this (not docked).

View attachment 4422

WAD. And it is docked (but not pinned).

If I click or hover over its tab (?) it drops down, partially covering the TCC tab's client area.

View attachment 4423

WAD. Like the other dockable windows.

If I then click in the TCC tab client area, the Command Input window rolls up and shortly thereafter TCMD crashes (disappears). I have dump files.

The rolling back up is WAD, the crash is not, but also not reproducible here.

 

[vefatica]

If you have crash dumps, then please post them or send them to [email protected].

Too big for my Exchange server. You can get them from my OneDrive public folder.

In WinDbg, one of them doesn't make it to the point where I can analyze the exception.

 

[rconn]

If the crash is in TCMD, you'll get a tcmd.exception.log file, which is the only one that's going to be useful without a reproducible failcase. If the crash is in Windows, the dump file is going to be largely meaningless.

 

[vefatica]

If the crash is in TCMD, you'll get a tcmd.exception.log file, which is the only one that's going to be useful without a reproducible failcase. If the crash is in Windows, the dump file is going to be largely meaningless.

I have none of them. But for one of the dumps I have analyzing the exception identifies TCMD and does not mention any DLL.

BUCKET_ID: APPLICATION_FAULT_INVALID_POINTER_READ_tcmd+372a4
FAILURE_EXCEPTION_CODE: c0000005
FAILURE_IMAGE_NAME: tcmd.exe
BUCKET_ID_IMAGE_STR: tcmd.exe
FAILURE_MODULE_NAME: tcmd
BUCKET_ID_MODULE_STR: tcmd

The last thing on the stack before exception handling is at tcmd+0x372a4.

 

[vefatica]

I do have this.

2024-03-20 12:37:56  21,340,808,618  C:\ProgramData\JP Software\Take Command 32\tcc.exception.log

That is likely to be thousands of crashes! I wish I could open it to see what was going on; it's too big for anything I have

 

[rconn]

A TCC exception log would be meaningless; you need a TCMD log.

The address you provided is in the GUI library in the shortcut manager code, unrelated to the Command Input window. Did you customize TCMD shortcut keys?

 

[vefatica]

I have customized some keystrokes in TCC. I don't know how to do that in TCMD.

 

[vefatica]

2024-03-20  12:37  21,340,808,618  tcc.exception.log

That's an interesting file. It starts like this.

[2024/03/20 12:20:46.183] D:\TakeCommand32\TCConsole\ntinit.cpp:412 wmain()  SEH Exception: 0x00007FFCFFAFBCB7 - ACCESS_VIOLATION
[2024/03/20 12:36:04.302] D:\TakeCommand32\TCConsole\ntinit.cpp:412 wmain()  SEH Exception: 0x00007FFCFFAFBCB7 - ACCESS_VIOLATION
[2024/03/20 12:36:16.942] D:\TakeCommand32\TCConsole\ntinit.cpp:412 wmain()  SEH Exception: 0x00007FFCB93D96CD - ACCESS_VIOLATION
[2024/03/20 12:36:16.942] D:\TakeCommand32\TCConsole\ntinit.cpp:412 wmain()  SEH Exception: 0x00007FFCB93D96CD - ACCESS_VIOLATION
[2024/03/20 12:36:16.957] D:\TakeCommand32\TCConsole\ntinit.cpp:412 wmain()  SEH Exception: 0x00007FFCB93D96CD - ACCESS_VIOLATION

and ends like this

[2024/03/20 12:37:56.089] D:\TakeCommand32\TCConsole\ntinit.cpp:412 wmain()  SEH Exception: 0x00007FFCB93D96CD - ACCESS_VIOLATION
[2024/03/20 12:37:56.111] D:\TakeCommand32\TCConsole\ntinit.cpp:412 wmain()  SEH Exception: 0x00007FFCB93D96CD - ACCESS_VIOLATION
[2024/03/20 12:37:56.127] D:\TakeCommand32\TCConsole\ntinit.cpp:412 wmain()  SEH Exception: 0x00007FFCB93D96CD - ACCESS_VIOLATION
[2024/03/20 12:37:56.560] D:\TakeCommand32\TCConsole\ntinit.cpp:412 wmain()  SEH Exception: 0x00007FFCB93D96CD - ACCESS_VIOLATION
[2024/03/20 12:37:56.575] D:\TakeCommand32\TCConsole\ntinit.cpp:412 wmain()  SEH Exception: 0x00007FFCB93D96CD - ACCESS_VIOLATION

with ~17,000,000 lines the same except for the timestamp. And in there are entries that are out of order timewise.

[2024/03/20 12:36:19.710] D:\TakeCommand32\TCConsole\ntinit.cpp:412 wmain()  SEH Exception: 0x00007FFCB93D96CD - ACCESS_VIOLATION
[2024/03/20 12:36:16.942] D:\TakeCommand32\TCConsole\ntinit.cpp:412 wmain()  SEH Exception: 0x00007FFCB93D96CD - ACCESS_VIOLATION

Could there be more than one exception log mechanism going at once?

Another funny thing ... if I look at the XHISTORY log (or my own timestamped log) there was nothing going on when that dump_log was being created.

 

[rconn]

That address is not in TCC (or any of TCC's dlls). Perhaps a third-party dll being injected into TCC.

If you have multiple TCC instances running you can have multiple overlapping log entries. And logging an exception doesn't necessarily mean that TCC crashed - TCC will recover (more or less gracefully) from internal exceptions, though crashes in injected dlls (or some Windows dlls) are apt to be fatal.