Access violation in Python garbage collector (visit_decref) - how to debug?

[Geoff Bache]

I've so far only tried within my application, but I'm aware it would be
easier if I could reproduce it outside. Even simplifying the context within
the application has proved difficult though, so I suspect this will be
hard. But I can try a bit more...

The file isn't "large" by production standards, only by test data
standards. It's about 500kb and not at all deeply nested, basically a long
list of dictionaries. But I don't seem to be able to reduce it further
either.

/Geoff

On Fri, Oct 4, 2019 at 9:53 PM Chris Angelico <rosuav at gmail.com> wrote:

> On Sat, Oct 5, 2019 at 5:38 AM Geoff Bache <geoff.bache at gmail.com> wrote:
> >
> > Hi all,
> >
> > We are running Python embedded in our C++ product and are now
> experiencing
> > crashes (access violation reading 0xffffffffff on Windows) in the Python
> > garbage collector.
> >
> > We got this on Python 3.6.4 originally, but I can reproduce it with both
> > Python 3.6.8 and Python 3.7.4.
> >
> > The chances of producing a minimal example that reproduces it reliably
> are
> > currently small I would say. All attempts to simplify the set up seem to
> > cause the problem to go away.
> > Indeed I can only reproduce it by sending a fairly large amount of data 2
> > or 3 times to our server - sending either half of the data does not
> > reproduce it.
>
> Have you tried to reproduce the issue outside of your application?
> Even if it means creating a gigantic Python script with a whopping
> triple-quoted string for the input data, it'd be helpful to try this.
> If you CAN repro the problem, it'd be way easier to diagnose (since we
> don't need your code, just your test case); and if you CAN'T, it may
> mean an issue with the embedding aspects.
>
> What's your definition of "fairly large" here? How many
> kilobytes/megabytes/gigabytes, and how deeply nested is the JSON
> object?
>
> ChrisA
> --
> https://mail.python.org/mailman/listinfo/python-list
>