use outermost proxied entry when checking for browser protocol

wee care about what the browser sees, so trust the outermost entry instead of the innermost

This is not secure _in general_, in that these values can be spoofed by malicious proxies,
but for CORS and cookie purposes, we only care about what the browser sees,
however many hops there may be.

A malicious proxy in the chain here isn't a concern because what matters is the immediate
hop from the _browser_, not the immediate hop from the _server_.

jupyterhub/apihandlers/auth.py

@@ -16,6 +16,7 @@ from tornado import web
 from .. import orm
 from .. import roles
 from .. import scopes
+from ..utils import get_browser_protocol
 from ..utils import token_authenticated
 from .base import APIHandler
 from .base import BaseHandler
@@ -115,7 +116,10 @@ class OAuthHandler:
         # make absolute local redirects full URLs
         # to satisfy oauthlib's absolute URI requirement
         redirect_uri = (
-            self.request.protocol + "://" + self.request.headers['Host'] + redirect_uri
+            get_browser_protocol(self.request)
+            + "://"
+            + self.request.host
+            + redirect_uri
         )
         parsed_url = urlparse(uri)
         query_list = parse_qsl(parsed_url.query, keep_blank_values=True)

jupyterhub/apihandlers/base.py

@@ -14,6 +14,7 @@ from tornado import web
 
 from .. import orm
 from ..handlers import BaseHandler
+from ..utils import get_browser_protocol
 from ..utils import isoformat
 from ..utils import url_path_join
 
@@ -60,6 +61,8 @@ class APIHandler(BaseHandler):
         """
         host_header = self.app.forwarded_host_header or "Host"
         host = self.request.headers.get(host_header)
+        if host and "," in host:
+            host = host.split(",", 1)[0].strip()
         referer = self.request.headers.get("Referer")
 
         # If no header is provided, assume it comes from a script/curl.
@@ -71,7 +74,8 @@ class APIHandler(BaseHandler):
             self.log.warning("Blocking API request with no referer")
             return False
 
-        proto = self.request.protocol
+        proto = get_browser_protocol(self.request)
+
         full_host = f"{proto}://{host}{self.hub.base_url}"
         host_url = urlparse(full_host)
         referer_url = urlparse(referer)

jupyterhub/handlers/base.py

@@ -49,6 +49,7 @@ from ..spawner import LocalProcessSpawner
 from ..user import User
 from ..utils import AnyTimeoutError
 from ..utils import get_accepted_mimetype
+from ..utils import get_browser_protocol
 from ..utils import maybe_future
 from ..utils import url_path_join
 
@@ -632,12 +633,10 @@ class BaseHandler(RequestHandler):
         next_url = self.get_argument('next', default='')
         # protect against some browsers' buggy handling of backslash as slash
         next_url = next_url.replace('\\', '%5C')
-        if (next_url + '/').startswith(
-            (
-                f'{self.request.protocol}://{self.request.host}/',
-                f'//{self.request.host}/',
-            )
-        ) or (
+        proto = get_browser_protocol(self.request)
+        host = self.request.host
+
+        if (next_url + '/').startswith((f'{proto}://{host}/', f'//{host}/',)) or (
             self.subdomain_host
             and urlparse(next_url).netloc
             and ("." + urlparse(next_url).netloc).endswith(

jupyterhub/services/auth.py

@@ -53,6 +53,7 @@ from traitlets import validate
 from traitlets.config import SingletonConfigurable
 
 from ..scopes import _intersect_expanded_scopes
+from ..utils import get_browser_protocol
 from ..utils import url_path_join
 
 
@@ -772,7 +773,7 @@ class HubOAuth(HubAuth):
             # OAuth that doesn't complete shouldn't linger too long.
             'max_age': 600,
         }
-        if handler.request.protocol == 'https':
+        if get_browser_protocol(handler.request) == 'https':
             kwargs['secure'] = True
         # load user cookie overrides
         kwargs.update(self.cookie_options)
@@ -812,7 +813,7 @@ class HubOAuth(HubAuth):
     def set_cookie(self, handler, access_token):
         """Set a cookie recording OAuth result"""
         kwargs = {'path': self.base_url, 'httponly': True}
-        if handler.request.protocol == 'https':
+        if get_browser_protocol(handler.request) == 'https':
             kwargs['secure'] = True
         # load user cookie overrides
         kwargs.update(self.cookie_options)

jupyterhub/tests/test_api.py

@@ -118,16 +118,39 @@ async def test_post_content_type(app, content_type, status):
         ("fake.example", {"netloc": "fake.example", "scheme": "https"}, {}, 403),
         # explicit ports, match
         ("fake.example:81", {"netloc": "fake.example:81"}, {}, 200),
-        # Test proxy defined headers taken into account by xheaders=True in
-        # https://github.com/jupyterhub/jupyterhub/blob/2.0.1/jupyterhub/app.py#L3065
+        # Test proxy protocol defined headers taken into account by utils.get_browser_protocol
         (
             "fake.example",
             {"netloc": "fake.example", "scheme": "https"},
-            # note {"X-Forwarded-Proto": "https"} does not work
             {'X-Scheme': 'https'},
             200,
         ),
+        (
+            "fake.example",
+            {"netloc": "fake.example", "scheme": "https"},
+            {'X-Forwarded-Proto': 'https'},
+            200,
+        ),
+        (
+            "fake.example",
+            {"netloc": "fake.example", "scheme": "https"},
+            {
+                'Forwarded': 'host=fake.example;proto=https,for=1.2.34;proto=http',
+                'X-Scheme': 'http',
+            },
+            200,
+        ),
+        (
+            "fake.example",
+            {"netloc": "fake.example", "scheme": "https"},
+            {
+                'Forwarded': 'host=fake.example;proto=http,for=1.2.34;proto=http',
+                'X-Scheme': 'https',
+            },
+            403,
+        ),
         ("fake.example", {"netloc": "fake.example"}, {'X-Scheme': 'https'}, 403),
+        ("fake.example", {"netloc": "fake.example"}, {'X-Scheme': 'https, http'}, 403),
     ],
 )
 async def test_cors_check(request, app, host, referer, extraheaders, status):

jupyterhub/tests/test_utils.py

@@ -2,12 +2,16 @@
 import asyncio
 import time
 from concurrent.futures import ThreadPoolExecutor
+from unittest.mock import Mock
 
 import pytest
 from async_generator import aclosing
 from tornado import gen
 from tornado.concurrent import run_on_executor
+from tornado.httpserver import HTTPRequest
+from tornado.httputil import HTTPHeaders
 
+from .. import utils
 from ..utils import iterate_until
 
 
@@ -88,3 +92,33 @@ async def test_tornado_coroutines():
     # verify that tornado gen and executor methods return awaitables
     assert (await t.on_executor()) == "executor"
     assert (await t.tornado_coroutine()) == "gen.coroutine"
+
+
+@pytest.mark.parametrize(
+    "forwarded, x_scheme, x_forwarded_proto, expected",
+    [
+        ("", "", "", "_attr_"),
+        ("for=1.2.3.4", "", "", "_attr_"),
+        ("for=1.2.3.4,proto=https", "", "", "_attr_"),
+        ("", "https", "http", "https"),
+        ("", "https, http", "", "https"),
+        ("", "https, http", "http", "https"),
+        ("proto=http ; for=1.2.3.4, proto=https", "https, http", "", "http"),
+        ("proto=invalid;for=1.2.3.4,proto=http", "https, http", "", "https"),
+        ("for=1.2.3.4,proto=http", "https, http", "", "https"),
+        ("", "invalid, http", "", "_attr_"),
+    ],
+)
+def test_browser_protocol(x_scheme, x_forwarded_proto, forwarded, expected):
+    request = Mock(spec=HTTPRequest)
+    request.protocol = "_attr_"
+    request.headers = HTTPHeaders()
+    if x_scheme:
+        request.headers["X-Scheme"] = x_scheme
+    if x_forwarded_proto:
+        request.headers["X-Forwarded-Proto"] = x_forwarded_proto
+    if forwarded:
+        request.headers["Forwarded"] = forwarded
+
+    proto = utils.get_browser_protocol(request)
+    assert proto == expected

jupyterhub/utils.py

@@ -683,3 +683,44 @@ def catch_db_error(f):
             return r
 
     return catching
+
+
+def get_browser_protocol(request):
+    """Get the _protocol_ seen by the browser
+
+    Like tornado's _apply_xheaders,
+    but in the case of multiple proxy hops,
+    use the outermost value (what the browser likely sees)
+    instead of the innermost value,
+    which is the most trustworthy.
+
+    We care about what the browser sees,
+    not where the request actually came from,
+    so trusting possible spoofs is the right thing to do.
+    """
+    headers = request.headers
+    # first choice: Forwarded header
+    forwarded_header = headers.get("Forwarded")
+    if forwarded_header:
+        first_forwarded = forwarded_header.split(",", 1)[0].strip()
+        fields = {}
+        forwarded_dict = {}
+        for field in first_forwarded.split(";"):
+            key, _, value = field.partition("=")
+            fields[key.strip().lower()] = value.strip()
+        if "proto" in fields and fields["proto"].lower() in {"http", "https"}:
+            return fields["proto"].lower()
+        else:
+            app_log.warning(
+                f"Forwarded header present without protocol: {forwarded_header}"
+            )
+
+    # second choice: X-Scheme or X-Forwarded-Proto
+    proto_header = headers.get("X-Scheme", headers.get("X-Forwarded-Proto", None))
+    if proto_header:
+        proto_header = proto_header.split(",")[0].strip().lower()
+        if proto_header in {"http", "https"}:
+            return proto_header
+
+    # no forwarded headers
+    return request.protocol