148593

Oh wow my head hurts.

#0  PyFrame_New (tstate=0x0, code=0x819e680, globals=0x8118fbc, locals=0x0)
    at Objects/frameobject.c:209
#1  0x08079dfe in PyEval_EvalCodeEx (co=0x819e680, globals=0x8118fbc,
    locals=0x0, args=0x81b3a30, argcount=1, kws=0x0, kwcount=0, defs=0x0,
    defcount=0, closure=0x0) at Python/ceval.c:2377
#2  0x080c225c in function_call (func=0x81b3174, arg=0x81b3a24, kw=0x0)
    at Objects/funcobject.c:374
#3  0x080b15f7 in PyObject_Call (func=0x0, arg=0x0, kw=0x0)
    at Objects/abstract.c:1684
#4  0x080b831b in instancemethod_call (func=0x81b3174, arg=0x81b3a24, kw=0x0)
    at Objects/classobject.c:2276
#5  0x080b15f7 in PyObject_Call (func=0x0, arg=0x0, kw=0x0)
    at Objects/abstract.c:1684
#6  0x0807b369 in PyEval_CallObjectWithKeywords (func=0x0, arg=0x810982c,
    kw=0x0) at Python/ceval.c:3049
#7  0x400814e0 in errorcb () at rpmmodule.c:195
#8  0x401da832 in vrpmlog () from /usr/lib/librpmio-4.1.so
#9  0x401da8cf in rpmlog () from /usr/lib/librpmio-4.1.so

...

I think that's quite enough for tonight. I really don't want to know why the internal python thread state is getting hosed. I just kicked off a build of python with internal python debugging support and then I can prod at it more tomorrow. Debugging this sort of stuff always makes me wish that I was a lot better with gdb.