Odyssey browser vs. Python [SOLVED]

I would like to know if this phenomenon / bug is reproducible.

So the standard Python installation has a script PYTHON:Scripts/PyVersion.py (it displays some system version information).

It produces this output:
Machine: AmigaOne
CPU: IBM PowerPC 750 FX
System:: ppc-amiga
Python Version: 2.5.6
Workbench Version: 53.53
Workbench ARexx Port: WORKBENCH

If Odyssey 1.23 (md5sum b6735a66396eb51add0498b65e2aebdb) is running, the script displays the first two lines, then crashes. If Odyssey is shutdown first, the script does work. Grim Reaper stacktrace shows something about socket.

I should add that if I run Odyssey, then quit it, then restart it, the script survives.

Stack trace from log:
0x81320A38 symbol not available
Sita:System/Python/Lib/lib-dynload/_socket.so:init_socket()+0xa0 (section 8 @ 0x1DC8)
Sita:SObjs/libpython25.so:import_submodule()+0x140 (section 10 @ 0xDD250)
Sita:SObjs/libpython25.so:import_module_level.isra.8()+0x234 (section 10 @ 0xDD940)
Sita:SObjs/libpython25.so:PyImport_ImportModuleLevel()+0xc4 (section 10 @ 0xDE8C0)
Sita:SObjs/libpython25.so:builtin___import__()+0x8c (section 10 @ 0xBC6A8)
Sita:SObjs/libpython25.so:PyObject_Call()+0x3c (section 10 @ 0x146A0)
Sita:SObjs/libpython25.so:PyEval_CallObjectWithKeywords()+0x98 (section 10 @ 0xBEA90)
Sita:SObjs/libpython25.so:PyEval_EvalFrameEx()+0x2d94 (section 10 @ 0xC2130)
Sita:SObjs/libpython25.so:PyEval_EvalCodeEx()+0x964 (section 10 @ 0xC5384)
Sita:SObjs/libpython25.so:PyEval_EvalCode()+0x2c (section 10 @ 0xC53E8)
Sita:SObjs/libpython25.so:PyImport_ExecCodeModuleEx()+0xcc (section 10 @ 0xDBF40)
Sita:SObjs/libpython25.so:load_source_module()+0x300 (section 10 @ 0xDC3A0)
Sita:SObjs/libpython25.so:import_submodule()+0x140 (section 10 @ 0xDD250)
Sita:SObjs/libpython25.so:import_module_level.isra.8()+0x234 (section 10 @ 0xDD940)
Sita:SObjs/libpython25.so:PyImport_ImportModuleLevel()+0xc4 (section 10 @ 0xDE8C0)
Sita:SObjs/libpython25.so:builtin___import__()+0x8c (section 10 @ 0xBC6A8)
Sita:SObjs/libpython25.so:PyObject_Call()+0x3c (section 10 @ 0x146A0)
Sita:SObjs/libpython25.so:PyEval_CallObjectWithKeywords()+0x98 (section 10 @ 0xBEA90)
Sita:SObjs/libpython25.so:PyEval_EvalFrameEx()+0x2d94 (section 10 @ 0xC2130)
Sita:SObjs/libpython25.so:PyEval_EvalCodeEx()+0x964 (section 10 @ 0xC5384)
Sita:SObjs/libpython25.so:PyEval_EvalFrameEx()+0x4dc0 (section 10 @ 0xC415C)
Sita:SObjs/libpython25.so:PyEval_EvalFrameEx()+0x4eb4 (section 10 @ 0xC4250)
Sita:SObjs/libpython25.so:PyEval_EvalFrameEx()+0x4eb4 (section 10 @ 0xC4250)
Sita:SObjs/libpython25.so:PyEval_EvalCodeEx()+0x964 (section 10 @ 0xC5384)
Sita:SObjs/libpython25.so:PyEval_EvalCode()+0x2c (section 10 @ 0xC53E8)
Sita:SObjs/libpython25.so:PyRun_FileExFlags()+0xb8 (section 10 @ 0xE9654)
Sita:SObjs/libpython25.so:PyRun_SimpleFileExFlags()+0xe4 (section 10 @ 0xEAE78)
Sita:SObjs/libpython25.so:Py_Main()+0x9ac (section 10 @ 0xFA47C)
native kernel module newlib.library.kmod+0x000020a4
native kernel module newlib.library.kmod+0x00002d0c
native kernel module newlib.library.kmod+0x00002ee8
Python:_start()+0x180 (section 7 @ 0x17C)
native kernel module dos.library.kmod+0x000255c8
native kernel module kernel+0x00040058
native kernel module kernel+0x000400a0

@Thematic
I doubt it will be reproducible by anyone else, since it sounds like memory corruption (which is probably caused by Odyssey but we can't be sure) . In any case, I had Odyssey running & got this from the Shell:
Quote:
(I used ## to comment-out the version number of Workbench, since I'm not sure if this is a private Beta version.)

I should add that I am running OS4's Debug kernel, which is much more likely to catch memory corruption. You could try enabling that in KickLayout, and see if it catches the memory corruption(?) at an earlier/different place.

@Thematic

Did you install my relatively recent pythonssl package from os4depot?

If so make sure you install the very recent version, there was an error in the shared object build, that will produce exactly that error.

It occurs at random, basically depends on what other programs are in memory when the shared object is loaded. I guess odyssey being fairly big tends to trigger more often than others, but it not caused by odyssey per seh.

@broadblues

I did have different shared objects from the archive. That was it. Thanks!