New submission from Richard Eames: I've been porting a project to the latest version of Django, and due to one of the changes in the Django, caused a recursion error in my code. However, the error (under certain conditions) then causes the python interpreter to core dump. I'm not 100% sure what causes this to happen, but it does seem to be similar to https://bugs.python.org/issue6028
I've created a minimal django project: https://github.com/Naddiseo/python-core-dump However, it does rely on some interaction between mysql, pymysql, and django to be reproduced, the latter two being 100% python code. I'm sorry that I could not reduce the test case further. One of the interesting/weird things about this bug is that (on my machine at least) it requires exactly 15 entries in the `MIDDLEWARE` variable in "coredump/settings.py" in my test project, any more, or any less will cause the interpreter to issue a `RecursionError` as expected, but not to core dump. This appears to happen in 3.5, and not in 3.6 so perhaps whatever fix was applied to 3.6 can be backported to 3.5 so that it doesn't core dump? ---------- components: Interpreter Core messages: 282745 nosy: Richard Eames priority: normal severity: normal status: open title: "Fatal Python error: Cannot recover from stack overflow." from RecursionError type: crash versions: Python 3.5 _______________________________________ Python tracker <rep...@bugs.python.org> <http://bugs.python.org/issue28913> _______________________________________ _______________________________________________ Python-bugs-list mailing list Unsubscribe: https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
