Package: release.debian.org
Severity: normal
Tags: bookworm
X-Debbugs-Cc: python-torn...@packages.debian.org
Control: affects -1 + src:python-tornado
User: release.debian....@packages.debian.org
Usertags: pu

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA512

[ Reason ]
This upload intends to fix the vulnerability CVE-2025-47287.

CVE-2025-47287 allows a remote attacker to create an extremely high volume of
log entries, constituting a DoS attack.

[ Impact ]
Users of Debian Bookworm will continue to be vulnerable to the mentioned issues
if the update is not approved.

[ Tests ]
The package comes with the testsuite enabled. The tests were adjusted to match
the new behavior to throw errors instead of logging warnings. All tests succeed.

[ Risks ]
The changes are quite simple. However, regressions are always possible. The
fact that the tests are successful reduce the risk of regressions.

[ Checklist ]
  [x] *all* changes are documented in the d/changelog
  [x] I reviewed all changes and I approve them
  [x] attach debdiff against the package in (old)stable
  [x] the issue is verified as fixed in unstable

[ Changes ]
Instead of logging warning messages, errors are created which preserve the
backtrace. Parsing the body has been moved within the code into
RequestHandler._execute() to be in the right exception handler scope. The tests
have been adjusted to this change.

[ Other info ]
All patches contain links to the original reports and commits.

-----BEGIN PGP SIGNATURE-----
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=Cx9E
-----END PGP SIGNATURE-----
diff -Nru python-tornado-6.2.0/debian/changelog 
python-tornado-6.2.0/debian/changelog
--- python-tornado-6.2.0/debian/changelog       2024-12-31 01:53:59.000000000 
+0100
+++ python-tornado-6.2.0/debian/changelog       2025-05-30 05:19:15.000000000 
+0200
@@ -1,3 +1,15 @@
+python-tornado (6.2.0-3+deb12u2) bookworm; urgency=medium
+
+  * Non-maintainer upload by the Debian LTS team.
+  * d/patches/CVE-2025-47287.patch: Add patch to fix CVE-2025-47287.
+    - When Tornado's 'multipart/form-data' parser encounters certain errors,
+      it logs a warning but continues trying to parse the remainder of the
+      data. This allows remote attackers to generate an extremely high volume
+      of logs, constituting a DoS attack. This DoS is compounded by the fact
+      that the logging subsystem is synchronous (closes: #1105886).
+
+ -- Daniel Leidert <dleid...@debian.org>  Fri, 30 May 2025 05:19:15 +0200
+
 python-tornado (6.2.0-3+deb12u1) bookworm; urgency=medium
 
   * Non-maintainer upload by the Debian LTS team.
diff -Nru python-tornado-6.2.0/debian/patches/CVE-2023-28370.patch 
python-tornado-6.2.0/debian/patches/CVE-2023-28370.patch
--- python-tornado-6.2.0/debian/patches/CVE-2023-28370.patch    1970-01-01 
01:00:00.000000000 +0100
+++ python-tornado-6.2.0/debian/patches/CVE-2023-28370.patch    2025-05-30 
05:19:15.000000000 +0200
@@ -0,0 +1,234 @@
+From: Ben Darnell <b...@bendarnell.com>
+Date: Thu, 8 May 2025 13:29:43 -0400
+Subject: httputil: Raise errors instead of logging in multipart/form-data
+ parsing
+
+We used to continue after logging an error, which allowed repeated
+errors to spam the logs. The error raised here will still be logged,
+but only once per request, consistent with other error handling in
+Tornado.
+
+Reviewed-By: Daniel Leidert <dleid...@debian.org>
+Origin: https://github.com/tornadoweb/tornado/pull/3497
+Bug: 
https://github.com/tornadoweb/tornado/security/advisories/GHSA-7cx3-6m66-7c5m
+Bug-Debian: https://bugs.debian.org/1105886
+Bug-Debian-Security: https://security-tracker.debian.org/tracker/CVE-2025-47287
+Bug-Freexian-Security: 
https://deb.freexian.com/extended-lts/tracker/CVE-2025-47287
+---
+ tornado/httputil.py             | 29 +++++++++++------------------
+ tornado/test/httpserver_test.py |  4 ++--
+ tornado/test/httputil_test.py   | 13 ++++++++-----
+ tornado/web.py                  | 17 +++++++++++++----
+ 4 files changed, 34 insertions(+), 29 deletions(-)
+
+diff --git a/tornado/httputil.py b/tornado/httputil.py
+index f74eced..80c159c 100644
+--- a/tornado/httputil.py
++++ b/tornado/httputil.py
+@@ -34,7 +34,6 @@ import unicodedata
+ from urllib.parse import urlencode, urlparse, urlunparse, parse_qsl
+ 
+ from tornado.escape import native_str, parse_qs_bytes, utf8
+-from tornado.log import gen_log
+ from tornado.util import ObjectDict, unicode_type
+ 
+ 
+@@ -759,25 +758,23 @@ def parse_body_arguments(
+     """
+     if content_type.startswith("application/x-www-form-urlencoded"):
+         if headers and "Content-Encoding" in headers:
+-            gen_log.warning(
+-                "Unsupported Content-Encoding: %s", 
headers["Content-Encoding"]
++            raise HTTPInputError(
++                "Unsupported Content-Encoding: %s" % 
headers["Content-Encoding"]
+             )
+-            return
+         try:
+             # real charset decoding will happen in 
RequestHandler.decode_argument()
+             uri_arguments = parse_qs_bytes(body, keep_blank_values=True)
+         except Exception as e:
+-            gen_log.warning("Invalid x-www-form-urlencoded body: %s", e)
++            raise HTTPInputError("Invalid x-www-form-urlencoded body: %s" % 
e) from e
+             uri_arguments = {}
+         for name, values in uri_arguments.items():
+             if values:
+                 arguments.setdefault(name, []).extend(values)
+     elif content_type.startswith("multipart/form-data"):
+         if headers and "Content-Encoding" in headers:
+-            gen_log.warning(
+-                "Unsupported Content-Encoding: %s", 
headers["Content-Encoding"]
++            raise HTTPInputError(
++                "Unsupported Content-Encoding: %s" % 
headers["Content-Encoding"]
+             )
+-            return
+         try:
+             fields = content_type.split(";")
+             for field in fields:
+@@ -786,9 +783,9 @@ def parse_body_arguments(
+                     parse_multipart_form_data(utf8(v), body, arguments, files)
+                     break
+             else:
+-                raise ValueError("multipart boundary not found")
++                raise HTTPInputError("multipart boundary not found")
+         except Exception as e:
+-            gen_log.warning("Invalid multipart/form-data: %s", e)
++            raise HTTPInputError("Invalid multipart/form-data: %s" % e) from e
+ 
+ 
+ def parse_multipart_form_data(
+@@ -817,26 +814,22 @@ def parse_multipart_form_data(
+         boundary = boundary[1:-1]
+     final_boundary_index = data.rfind(b"--" + boundary + b"--")
+     if final_boundary_index == -1:
+-        gen_log.warning("Invalid multipart/form-data: no final boundary")
+-        return
++        raise HTTPInputError("Invalid multipart/form-data: no final boundary 
found")
+     parts = data[:final_boundary_index].split(b"--" + boundary + b"\r\n")
+     for part in parts:
+         if not part:
+             continue
+         eoh = part.find(b"\r\n\r\n")
+         if eoh == -1:
+-            gen_log.warning("multipart/form-data missing headers")
+-            continue
++            raise HTTPInputError("multipart/form-data missing headers")
+         headers = HTTPHeaders.parse(part[:eoh].decode("utf-8"))
+         disp_header = headers.get("Content-Disposition", "")
+         disposition, disp_params = _parse_header(disp_header)
+         if disposition != "form-data" or not part.endswith(b"\r\n"):
+-            gen_log.warning("Invalid multipart/form-data")
+-            continue
++            raise HTTPInputError("Invalid multipart/form-data")
+         value = part[eoh + 4 : -2]
+         if not disp_params.get("name"):
+-            gen_log.warning("multipart/form-data value missing name")
+-            continue
++            raise HTTPInputError("multipart/form-data missing name")
+         name = disp_params["name"]
+         if disp_params.get("filename"):
+             ctype = headers.get("Content-Type", "application/unknown")
+diff --git a/tornado/test/httpserver_test.py b/tornado/test/httpserver_test.py
+index 367a0ef..3fff206 100644
+--- a/tornado/test/httpserver_test.py
++++ b/tornado/test/httpserver_test.py
+@@ -1025,9 +1025,9 @@ class GzipUnsupportedTest(GzipBaseTest, 
AsyncHTTPTestCase):
+         # Gzip support is opt-in; without it the server fails to parse
+         # the body (but parsing form bodies is currently just a log message,
+         # not a fatal error).
+-        with ExpectLog(gen_log, "Unsupported Content-Encoding"):
++        with ExpectLog(gen_log, ".*Unsupported Content-Encoding"):
+             response = self.post_gzip("foo=bar")
+-        self.assertEqual(json_decode(response.body), {})
++        self.assertEqual(response.code, 400)
+ 
+ 
+ class StreamingChunkSizeTest(AsyncHTTPTestCase):
+diff --git a/tornado/test/httputil_test.py b/tornado/test/httputil_test.py
+index 25faf66..4ceda3b 100644
+--- a/tornado/test/httputil_test.py
++++ b/tornado/test/httputil_test.py
+@@ -12,7 +12,6 @@ from tornado.httputil import (
+ )
+ from tornado.escape import utf8, native_str
+ from tornado.log import gen_log
+-from tornado.testing import ExpectLog
+ 
+ import copy
+ import datetime
+@@ -194,7 +193,9 @@ Foo
+             b"\n", b"\r\n"
+         )
+         args, files = form_data_args()
+-        with ExpectLog(gen_log, "multipart/form-data missing headers"):
++        with self.assertRaises(
++            HTTPInputError, msg="multipart/form-data missing headers"
++        ):
+             parse_multipart_form_data(b"1234", data, args, files)
+         self.assertEqual(files, {})
+ 
+@@ -208,7 +209,7 @@ Foo
+             b"\n", b"\r\n"
+         )
+         args, files = form_data_args()
+-        with ExpectLog(gen_log, "Invalid multipart/form-data"):
++        with self.assertRaises(HTTPInputError, msg="Invalid 
multipart/form-data"):
+             parse_multipart_form_data(b"1234", data, args, files)
+         self.assertEqual(files, {})
+ 
+@@ -221,7 +222,7 @@ Foo--1234--""".replace(
+             b"\n", b"\r\n"
+         )
+         args, files = form_data_args()
+-        with ExpectLog(gen_log, "Invalid multipart/form-data"):
++        with self.assertRaises(HTTPInputError, msg="Invalid 
multipart/form-data"):
+             parse_multipart_form_data(b"1234", data, args, files)
+         self.assertEqual(files, {})
+ 
+@@ -235,7 +236,9 @@ Foo
+             b"\n", b"\r\n"
+         )
+         args, files = form_data_args()
+-        with ExpectLog(gen_log, "multipart/form-data value missing name"):
++        with self.assertRaises(
++            HTTPInputError, msg="multipart/form-data value missing name"
++        ):
+             parse_multipart_form_data(b"1234", data, args, files)
+         self.assertEqual(files, {})
+ 
+diff --git a/tornado/web.py b/tornado/web.py
+index 05b571e..98181c0 100644
+--- a/tornado/web.py
++++ b/tornado/web.py
+@@ -1670,6 +1670,14 @@ class RequestHandler(object):
+         try:
+             if self.request.method not in self.SUPPORTED_METHODS:
+                 raise HTTPError(405)
++
++            # If we're not in stream_request_body mode, this is the place 
where we parse the body.
++            if not _has_stream_request_body(self.__class__):
++                try:
++                    self.request._parse_body()
++                except httputil.HTTPInputError as e:
++                    raise HTTPError(400, "Invalid body: %s" % e) from e
++
+             self.path_args = [self.decode_argument(arg) for arg in args]
+             self.path_kwargs = dict(
+                 (k, self.decode_argument(v, name=k)) for (k, v) in 
kwargs.items()
+@@ -1864,7 +1872,7 @@ def _has_stream_request_body(cls: Type[RequestHandler]) 
-> bool:
+ 
+ 
+ def removeslash(
+-    method: Callable[..., Optional[Awaitable[None]]]
++    method: Callable[..., Optional[Awaitable[None]]],
+ ) -> Callable[..., Optional[Awaitable[None]]]:
+     """Use this decorator to remove trailing slashes from the request path.
+ 
+@@ -1893,7 +1901,7 @@ def removeslash(
+ 
+ 
+ def addslash(
+-    method: Callable[..., Optional[Awaitable[None]]]
++    method: Callable[..., Optional[Awaitable[None]]],
+ ) -> Callable[..., Optional[Awaitable[None]]]:
+     """Use this decorator to add a missing trailing slash to the request path.
+ 
+@@ -2317,8 +2325,9 @@ class _HandlerDelegate(httputil.HTTPMessageDelegate):
+         if self.stream_request_body:
+             future_set_result_unless_cancelled(self.request._body_future, 
None)
+         else:
++            # Note that the body gets parsed in RequestHandler._execute so it 
can be in
++            # the right exception handler scope.
+             self.request.body = b"".join(self.chunks)
+-            self.request._parse_body()
+             self.execute()
+ 
+     def on_connection_close(self) -> None:
+@@ -3183,7 +3192,7 @@ class GZipContentEncoding(OutputTransform):
+ 
+ 
+ def authenticated(
+-    method: Callable[..., Optional[Awaitable[None]]]
++    method: Callable[..., Optional[Awaitable[None]]],
+ ) -> Callable[..., Optional[Awaitable[None]]]:
+     """Decorate methods with this to require that the user be logged in.
+ 
diff -Nru python-tornado-6.2.0/debian/patches/series 
python-tornado-6.2.0/debian/patches/series
--- python-tornado-6.2.0/debian/patches/series  2024-12-31 01:53:59.000000000 
+0100
+++ python-tornado-6.2.0/debian/patches/series  2025-05-30 05:19:15.000000000 
+0200
@@ -7,3 +7,4 @@
 CVE-2024-52804.patch
 CVE-2023-28370-1.patch
 CVE-2023-28370-2.patch
+CVE-2023-28370.patch

Reply via email to