When failing to connect to a MongoDB slave, the pymongo driver reports "could not find master/primary". This error is misleading, since the master/primary wasn't even requested. For example,

PS C:\Users\jaraco\> python
Python 2.7.1 (r271:86832, Nov 27 2010, 17:19:03) [MSC v.1500 64 bit (AMD64)] on
win32
Type "help", "copyright", "credits" or "license" for more information.
>>> import pymongo
>>> pymongo.Connection('doesnotexist.example.net', slave_okay=True)
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "c:\python\lib\site-packages\pymongo-1.10.1-py2.7-win-amd64.egg\pymongo\connection.py", line 352, in _init_
self.__find_master()
File "c:\python\lib\site-packages\pymongo-1.10.1-py2.7-win-amd64.egg\pymongo\connection.py", line 576, in __find_master
raise AutoReconnect("could not find master/primary")
pymongo.errors.AutoReconnect: could not find master/primary

From the traceback, it appears as if the driver is attempting to "find_master" and fails to do so, but in reality, I know it's simply failing to make a connection to the slave I requested.

The error message could be improved to not provide misleading information and possibly to provide more detail about which host or hosts failed.