Comment 22 for bug 1365646

Barry Warsaw (barry) wrote :

Not that you care, but it's tricky to craft a test case for this because the crash doesn't actually affect any behavior! Since the UpdateDownloaded signal is emitted before the checking lock is released, everything still works. The only effect is that s-i-d tracebacks, and you only see that in the crash report because of the way dbus-python's error handling works.

This probably means that the best thing to do is just explicitly swallow any exceptions from .release().