通过 Fiddler 和 NTLM 身份验证使用 Python 请求库不一致成功

Question

我已经研究这个问题将近 20 个小时了。

我的所有网络流量都通过我机器上的 Fiddler 路由，然后连接到我们的公司代理。一切正常，除了 Python 应用程序尝试使用 https 访问远程服务器（http 始终正常）。

我导出了公司证书并粘贴到文件中：C:\anaconda2\envs\py36\Lib\site-packages\certifi\cacert.pem。我还使用 verify= 在我的 requests.get 调用中明确设置了它。行为没有区别。

我将本地fiddler代理信息设置为环境变量。 Fiddler 还配置为自动验证。使用 http 没有任何问题。

我似乎只能通过 https 连接到远程服务器，如果我先转到 http://www.google.com，然后快速尝试使用 https 连接。随后的尝试产生以下错误

requests.get('http://www.google.com') # always works for any website
<Response [200]>

requests.get('https://www.anaconda.com') # works after visiting http://www.google.com
<Response [200]>

requests.get('https://www.anaconda.com') # always fails, unless visiting http://www.google.com first

---------------------------------------------------------------------------
OSError                                   Traceback (most recent call last)
C:\anaconda2\envs\py36\lib\site-packages\urllib3\connectionpool.py in urlopen(self, method, url, body, headers, retries, redirect, assert_same_host, timeout, pool_timeout, release_conn, chunked, body_pos, **response_kw)
    593             if is_new_proxy_conn:
--> 594                 self._prepare_proxy(conn)
    595

C:\anaconda2\envs\py36\lib\site-packages\urllib3\connectionpool.py in _prepare_proxy(self, conn)
    804         conn.set_tunnel(self._proxy_host, self.port, self.proxy_headers)
--> 805         conn.connect()
    806

C:\anaconda2\envs\py36\lib\site-packages\urllib3\connection.py in connect(self)
    307             # self._tunnel_host below.
--> 308             self._tunnel()
    309             # Mark this connection as not reusable

C:\anaconda2\envs\py36\lib\http\client.py in _tunnel(self)
    918             raise OSError("Tunnel connection failed: %d %s" % (code,
--> 919                                                                message.strip()))
    920         while True:

OSError: Tunnel connection failed: 407 Proxy Authentication Required

During handling of the above exception, another exception occurred:

MaxRetryError                             Traceback (most recent call last)
C:\anaconda2\envs\py36\lib\site-packages\requests\adapters.py in send(self, request, stream, timeout, verify, cert, proxies)
    448                     retries=self.max_retries,
--> 449                     timeout=timeout
    450                 )

C:\anaconda2\envs\py36\lib\site-packages\urllib3\connectionpool.py in urlopen(self, method, url, body, headers, retries, redirect, assert_same_host, timeout, pool_timeout, release_conn, chunked, body_pos, **response_kw)
    637             retries = retries.increment(method, url, error=e, _pool=self,
--> 638                                         _stacktrace=sys.exc_info()[2])
    639             retries.sleep()

C:\anaconda2\envs\py36\lib\site-packages\urllib3\util\retry.py in increment(self, method, url, response, error, _pool, _stacktrace)
    397         if new_retry.is_exhausted():
--> 398             raise MaxRetryError(_pool, url, error or ResponseError(cause))
    399

MaxRetryError: HTTPSConnectionPool(host='www.google.com', port=443): Max retries exceeded with url: / (Caused by ProxyError('Cannot connect to proxy.', OSError('Tunnel connection failed: 407 Proxy Authentication Required',)))

During handling of the above exception, another exception occurred:

ProxyError                                Traceback (most recent call last)
<ipython-input-49-df48f2544f7e> in <module>
----> 1 requests.get('https://www.google.com')

C:\anaconda2\envs\py36\lib\site-packages\requests\api.py in get(url, params, **kwargs)
     73
     74     kwargs.setdefault('allow_redirects', True)
---> 75     return request('get', url, params=params, **kwargs)
     76
     77

C:\anaconda2\envs\py36\lib\site-packages\requests\api.py in request(method, url, **kwargs)
     58     # cases, and look like a memory leak in others.
     59     with sessions.Session() as session:
---> 60         return session.request(method=method, url=url, **kwargs)
     61
     62

C:\anaconda2\envs\py36\lib\site-packages\requests\sessions.py in request(self, method, url, params, data, headers, cookies, files, auth, timeout, allow_redirects, proxies, hooks, stream, verify, cert, json)
    531         }
    532         send_kwargs.update(settings)
--> 533         resp = self.send(prep, **send_kwargs)
    534
    535         return resp

C:\anaconda2\envs\py36\lib\site-packages\requests\sessions.py in send(self, request, **kwargs)
    644
    645         # Send the request
--> 646         r = adapter.send(request, **kwargs)
    647
    648         # Total elapsed time of the request (approximately)

C:\anaconda2\envs\py36\lib\site-packages\requests\adapters.py in send(self, request, stream, timeout, verify, cert, proxies)
    508
    509             if isinstance(e.reason, _ProxyError):
--> 510                 raise ProxyError(e, request=request)
    511
    512             if isinstance(e.reason, _SSLError):

ProxyError: HTTPSConnectionPool(host='www.google.com', port=443): Max retries exceeded with url: / (Caused by ProxyError('Cannot connect to proxy.', OSError('Tunnel connection failed: 407 Proxy Authentication Required',)))

在我发出请求时，其中一位网络人员正在查看公司代理日志。当发出失败的 https 请求时，他没有在其日志中看到与公司代理的连接。

其他尝试过的东西：

requests-ntlm.
重置 Fiddler 的所有证书。
明确将代理和公司证书信息传递给请求。

谢谢。

Answer 1

对于有同样问题的人。

进一步的研究让我下载了 Python 应用程序 Px (px.exe) Px on GitHub 和 turf Fiddler，它只对 Python 试图退出的应用程序间歇性工作到互联网。

在我的例子中，

PX 本身需要 ZERO 配置。我只需要设置 http_proxy 和 https_proxy 环境变量，以便任何 Python 应用程序都知道将流量集中到哪里。然后我只是运行 Px，一切正常。

希望这对人们有所帮助。

通过 Fiddler 和 NTLM 身份验证使用 Python 请求库不一致成功

Using Python requests library through Fiddler and NTLM authentication is inconsistently successful

python-requests

fiddler

ntlm-authentication