Error 0xc0000142

[Philip Schuchardt]

Oh! That's a good find! Windows can give some weird call stack when the
code is built in release mode.

On Thu, Dec 12, 2024 at 11:18 AM Olly Betts <olly at survex.com> wrote:

> On Thu, Dec 12, 2024 at 07:09:26PM +0000, Olly Betts wrote:
> > Quoting myself earlier in the thread:
> >
> > | The C++ symbol decodes as OpenSSLInit::OpenSSLInit() and it's throwing
> > | an exception, but libpodofo's source code doesn't seem to contain a
> > | class called OpenSSLInit.  It's not just something inlined from
> > | OpenSSL's headers either.  The address from the backtrace is definitely
> > | in the range where libpodofo is listed as loaded:
> > |
> > | > ModLoad: 00007ffc`56830000 00007ffc`56b9e000   C:\Program
> Files\Survex\libpodofo.dll
> >
> > I think that symbol is from some sort of C++ wrapper around OpenSSL, but
> I
> > failed to work out what it's in.  Such code definitely exists (but I
> don't
> > think this particular instance is the culprit as fastdds doesn't seem to
> be
> > among the DLLs being loaded; also no C++ exception is being thrown by
> it):
> >
> >
> https://sources.debian.org/src/fastdds/3.1.0+ds-2/src/cpp/security/OpenSSLInit.hpp/?hl=11#L15
>
> Aha, I think this is the issue we are hitting:
>
> https://github.com/podofo/podofo/issues/27
>
> It was fixed in podofo git back in October, and should be in podofo 1.0
> (but their latest release still seems to be 0.10.4).
>
> Looks like it might work if we ship OpenSSL's legacy.dll (which seems to
> get loaded at runtime so would have been missed by walking the DLL
> dependency tree).  I can try that.
>
> Cheers,
>     Olly
>