Comment 3 for bug 1545577

This still seems to be happening and now it's trying to do a rollback and fails with a similar error. It seems to me like something to do with backrefs and cascade-update. Some stack objects seem to be flushed from the session between get and delete.

We may have to investigate more.

  File "/usr/local/lib/python2.7/dist-packages/sqlalchemy/orm/session.py", line 490, in __exit__
    self.rollback()
  File "/usr/local/lib/python2.7/dist-packages/sqlalchemy/util/langhelpers.py", line 60, in __exit__
    compat.reraise(exc_type, exc_value, exc_tb)
  File "/usr/local/lib/python2.7/dist-packages/sqlalchemy/orm/session.py", line 487, in __exit__
    self.commit()
  File "/usr/local/lib/python2.7/dist-packages/sqlalchemy/orm/session.py", line 392, in commit
    self._prepare_impl()
  File "/usr/local/lib/python2.7/dist-packages/sqlalchemy/orm/session.py", line 372, in _prepare_impl
    self.session.flush()
  File "/usr/local/lib/python2.7/dist-packages/sqlalchemy/orm/session.py", line 2019, in flush
    self._flush(objects)
  File "/usr/local/lib/python2.7/dist-packages/sqlalchemy/orm/session.py", line 2137, in _flush
    transaction.rollback(_capture_exception=True)