What steps will reproduce the problem?

import pyodbc; pyodbc.connect(...)

What version of the product are you using? On what operating system?

pyodbc 3.0.6, installed from pip; I just cloned the latest version from the Git repository with the same results

Please provide any additional information below.

(gdb) backtrace

0 0x00007ffff7aa084e in PyDict_SetItem () from /usr/lib/libpython2.7.so.1.0

1 0x00007ffff69bec09 in GetConnectionInfo(_object*, Connection*) ()

from /usr/lib/python2.7/site-packages/pyodbc.so

2 0x00007ffff69be669 in Connection_New(_object*, bool, bool, bool, long, bool) ()

from /usr/lib/python2.7/site-packages/pyodbc.so

3 0x00007ffff69b734f in mod_connect(_object*, _object*, _object*) ()

from /usr/lib/python2.7/site-packages/pyodbc.so

4 0x00007ffff7afe05a in PyEval_EvalFrameEx () from /usr/lib/libpython2.7.so.1.0

5 0x00007ffff7afeedd in PyEval_EvalCodeEx () from /usr/lib/libpython2.7.so.1.0

6 0x00007ffff7afefb2 in PyEval_EvalCode () from /usr/lib/libpython2.7.so.1.0

7 0x00007ffff7b17eea in run_mod () from /usr/lib/libpython2.7.so.1.0

8 0x00007ffff7b18ce2 in PyRun_FileExFlags () from /usr/lib/libpython2.7.so.1.0

9 0x00007ffff7b196fb in PyRun_SimpleFileExFlags () from /usr/lib/libpython2.7.so.1.0

10 0x00007ffff7b2a9f2 in Py_Main () from /usr/lib/libpython2.7.so.1.0

11 0x00007ffff747aa15 in __libc_start_main () from /usr/lib/libc.so.6

12 0x0000000000400741 in _start ()

I'm using Python 2.7.3 on Arch Linux and unixodbc.

I'm able to use pyodbc sometimes without it segfaulting, but I haven't been able to figure out what's making the difference.

Running this file:

https://raw.github.com/bendmorris/rdf-treestore/master/treestore.py

works fine, and afterward I can create a pyodbc connection with no problem. Otherwise, running the same connection line from the creation of the Treestore class in that file gives me a segfault.

I'm still doing some detective work. Importing part of BioPython with the line "import Bio.Phylo" prevents the segfault for some reason.

Can you reproduce this in a script with just pyodbc? I'm not suggesting it isn't pyodbc, but it would greatly simplify researching this.

The problem occurs in a script with just two lines:

import pyodbc con = pyodbc.connect("DSN=Virtuoso;UID=dba;PWD=dba")

For some reason adding a BioPython import first prevents the segfault, but otherwise it happens. I wasn't able to figure out why this is.