Action not permitted
Modal body text goes here.
Modal Title
Modal Body
GHSA-MPF4-983Q-P7J4
Vulnerability from github – Published: 2026-09-02 14:38 – Updated: 2026-09-02 14:38Summary
Tornado parses application/x-www-form-urlencoded bodies with urllib.parse.parse_qs and does not pass max_num_fields. A body made almost entirely of separators produces tens of millions of fields, and the parse happens on the event loop before the handler runs, so a single request stalls the whole server.
Where it is
tornado/escape.py, at HEAD e530031405e2154654dedc4c84d5656b557ea310:
result = urllib.parse.parse_qs(
qs, keep_blank_values, strict_parsing, encoding="latin1", errors="strict"
)
max_num_fields is the parameter CPython added for exactly this, and it is absent.
The path to it is entirely server-side and pre-dispatch. RequestHandler._execute parses the body at tornado/web.py:1821, which reaches HTTPServerRequest._parse_body at tornado/httputil.py:636, and the urlencoded branch of parse_body_arguments calls parse_qs_bytes at tornado/httputil.py:1030.
The size that reaches it is bounded only by the body cap, which defaults to the stream's max_buffer_size of 104857600 at tornado/iostream.py:239, applied as the request body default at tornado/http1connection.py:136-140. A 100 MB body of separators is around fifty million fields.
Impact
Denial of service against the whole process, not one request. Tornado is single-threaded and the parse is synchronous on the event loop, so every other connection waits. No authentication is needed if any route accepts a form post, which is the normal case.
Suggested fix
Pass a bound:
result = urllib.parse.parse_qs(
qs, keep_blank_values, strict_parsing, encoding="latin1", errors="strict",
max_num_fields=max_num_fields,
)
with a conservative default and a way for applications to raise it. CPython raises ValueError when the limit is exceeded, which maps cleanly onto a 400.
Lowering the default body cap for urlencoded specifically would help too, since 100 MB of form fields is not a shape any real client sends.
Why I do not think this is a duplicate
The published tornado advisories cover out-of-bounds access in the C extension, unbounded accumulation of decompressed chunks in AsyncHTTPClient, the Authorization header surviving cross-origin redirects, credential leakage on curl handle reuse, and cookie attribute validation. The decompression one is the nearest in spirit and is on the client side; this is the server parsing a request body. The call is unchanged at HEAD.
{
"affected": [
{
"database_specific": {
"last_known_affected_version_range": "\u003c= 6.5.7"
},
"package": {
"ecosystem": "PyPI",
"name": "tornado"
},
"ranges": [
{
"events": [
{
"introduced": "0"
},
{
"fixed": "6.5.8"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"aliases": [
"CVE-2026-82397"
],
"database_specific": {
"cwe_ids": [
"CWE-1284",
"CWE-400"
],
"github_reviewed": true,
"github_reviewed_at": "2026-09-02T14:38:43Z",
"nvd_published_at": "2026-08-31T22:17:22Z",
"severity": "HIGH"
},
"details": "## Summary\n\nTornado parses `application/x-www-form-urlencoded` bodies with `urllib.parse.parse_qs` and does not pass `max_num_fields`. A body made almost entirely of separators produces tens of millions of fields, and the parse happens on the event loop before the handler runs, so a single request stalls the whole server.\n\n## Where it is\n\n`tornado/escape.py`, at HEAD `e530031405e2154654dedc4c84d5656b557ea310`:\n\n```python\nresult = urllib.parse.parse_qs(\n qs, keep_blank_values, strict_parsing, encoding=\"latin1\", errors=\"strict\"\n)\n```\n\n`max_num_fields` is the parameter CPython added for exactly this, and it is absent.\n\nThe path to it is entirely server-side and pre-dispatch. `RequestHandler._execute` parses the body at `tornado/web.py:1821`, which reaches `HTTPServerRequest._parse_body` at `tornado/httputil.py:636`, and the urlencoded branch of `parse_body_arguments` calls `parse_qs_bytes` at `tornado/httputil.py:1030`.\n\nThe size that reaches it is bounded only by the body cap, which defaults to the stream\u0027s `max_buffer_size` of 104857600 at `tornado/iostream.py:239`, applied as the request body default at `tornado/http1connection.py:136-140`. A 100 MB body of separators is around fifty million fields.\n\n## Impact\n\nDenial of service against the whole process, not one request. Tornado is single-threaded and the parse is synchronous on the event loop, so every other connection waits. No authentication is needed if any route accepts a form post, which is the normal case.\n\n## Suggested fix\n\nPass a bound:\n\n```python\nresult = urllib.parse.parse_qs(\n qs, keep_blank_values, strict_parsing, encoding=\"latin1\", errors=\"strict\",\n max_num_fields=max_num_fields,\n)\n```\n\nwith a conservative default and a way for applications to raise it. CPython raises `ValueError` when the limit is exceeded, which maps cleanly onto a 400.\n\nLowering the default body cap for urlencoded specifically would help too, since 100 MB of form fields is not a shape any real client sends.\n\n## Why I do not think this is a duplicate\n\nThe published tornado advisories cover out-of-bounds access in the C extension, unbounded accumulation of decompressed chunks in `AsyncHTTPClient`, the Authorization header surviving cross-origin redirects, credential leakage on curl handle reuse, and cookie attribute validation. The decompression one is the nearest in spirit and is on the client side; this is the server parsing a request body. The call is unchanged at HEAD.",
"id": "GHSA-mpf4-983q-p7j4",
"modified": "2026-09-02T14:38:43Z",
"published": "2026-09-02T14:38:43Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/tornadoweb/tornado/security/advisories/GHSA-mpf4-983q-p7j4"
},
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-82397"
},
{
"type": "WEB",
"url": "https://github.com/tornadoweb/tornado/pull/3704"
},
{
"type": "WEB",
"url": "https://github.com/tornadoweb/tornado/commit/8d6363ed7b69d5f0da806efe34d256627a2191de"
},
{
"type": "PACKAGE",
"url": "https://github.com/tornadoweb/tornado"
},
{
"type": "WEB",
"url": "https://github.com/tornadoweb/tornado/releases/tag/v6.5.8"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H",
"type": "CVSS_V3"
}
],
"summary": " Tornado: Urlencoded body parsing omits max_num_fields, so one request can stall the event loop"
}
BREW-JUPYTERLAB-CVE-2026-82397 (GHSA-MPF4-983Q-P7J4)
Vulnerability from osv_homebrew – Published: 2026-09-03 09:17 – Updated: 2026-09-17 19:02 – Source websiteSummary
Tornado parses application/x-www-form-urlencoded bodies with urllib.parse.parse_qs and does not pass max_num_fields. A body made almost entirely of separators produces tens of millions of fields, and the parse happens on the event loop before the handler runs, so a single request stalls the whole server.
Where it is
tornado/escape.py, at HEAD e530031405e2154654dedc4c84d5656b557ea310:
result = urllib.parse.parse_qs(
qs, keep_blank_values, strict_parsing, encoding="latin1", errors="strict"
)
max_num_fields is the parameter CPython added for exactly this, and it is absent.
The path to it is entirely server-side and pre-dispatch. RequestHandler._execute parses the body at tornado/web.py:1821, which reaches HTTPServerRequest._parse_body at tornado/httputil.py:636, and the urlencoded branch of parse_body_arguments calls parse_qs_bytes at tornado/httputil.py:1030.
The size that reaches it is bounded only by the body cap, which defaults to the stream's max_buffer_size of 104857600 at tornado/iostream.py:239, applied as the request body default at tornado/http1connection.py:136-140. A 100 MB body of separators is around fifty million fields.
Impact
Denial of service against the whole process, not one request. Tornado is single-threaded and the parse is synchronous on the event loop, so every other connection waits. No authentication is needed if any route accepts a form post, which is the normal case.
Suggested fix
Pass a bound:
result = urllib.parse.parse_qs(
qs, keep_blank_values, strict_parsing, encoding="latin1", errors="strict",
max_num_fields=max_num_fields,
)
with a conservative default and a way for applications to raise it. CPython raises ValueError when the limit is exceeded, which maps cleanly onto a 400.
Lowering the default body cap for urlencoded specifically would help too, since 100 MB of form fields is not a shape any real client sends.
Why I do not think this is a duplicate
The published tornado advisories cover out-of-bounds access in the C extension, unbounded accumulation of decompressed chunks in AsyncHTTPClient, the Authorization header surviving cross-origin redirects, credential leakage on curl handle reuse, and cookie attribute validation. The decompression one is the nearest in spirit and is on the client side; this is the server parsing a request body. The call is unchanged at HEAD.
{
"affected": [
{
"ecosystem_specific": {
"fix": "bump",
"range_state": "fixed",
"resource": "tornado",
"resource_purl": "pkg:pypi/tornado@6.5.9",
"upstream_fixed_in": "6.5.8"
},
"package": {
"ecosystem": "Homebrew",
"name": "jupyterlab",
"purl": "pkg:brew/jupyterlab"
},
"ranges": [
{
"events": [
{
"introduced": "1.2.0"
},
{
"fixed": "4.6.3"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"database_specific": {
"confidence": "high",
"source": "matched",
"strategy": "registry",
"upstream_evidence": [
{
"ecosystem": "PyPI",
"key": "pkg:pypi/tornado@6.5.9",
"name": "tornado",
"resource": "tornado",
"strategy": "registry",
"subject_version": "6.5.9"
}
]
},
"details": "## Summary\n\nTornado parses `application/x-www-form-urlencoded` bodies with `urllib.parse.parse_qs` and does not pass `max_num_fields`. A body made almost entirely of separators produces tens of millions of fields, and the parse happens on the event loop before the handler runs, so a single request stalls the whole server.\n\n## Where it is\n\n`tornado/escape.py`, at HEAD `e530031405e2154654dedc4c84d5656b557ea310`:\n\n```python\nresult = urllib.parse.parse_qs(\n qs, keep_blank_values, strict_parsing, encoding=\"latin1\", errors=\"strict\"\n)\n```\n\n`max_num_fields` is the parameter CPython added for exactly this, and it is absent.\n\nThe path to it is entirely server-side and pre-dispatch. `RequestHandler._execute` parses the body at `tornado/web.py:1821`, which reaches `HTTPServerRequest._parse_body` at `tornado/httputil.py:636`, and the urlencoded branch of `parse_body_arguments` calls `parse_qs_bytes` at `tornado/httputil.py:1030`.\n\nThe size that reaches it is bounded only by the body cap, which defaults to the stream\u0027s `max_buffer_size` of 104857600 at `tornado/iostream.py:239`, applied as the request body default at `tornado/http1connection.py:136-140`. A 100 MB body of separators is around fifty million fields.\n\n## Impact\n\nDenial of service against the whole process, not one request. Tornado is single-threaded and the parse is synchronous on the event loop, so every other connection waits. No authentication is needed if any route accepts a form post, which is the normal case.\n\n## Suggested fix\n\nPass a bound:\n\n```python\nresult = urllib.parse.parse_qs(\n qs, keep_blank_values, strict_parsing, encoding=\"latin1\", errors=\"strict\",\n max_num_fields=max_num_fields,\n)\n```\n\nwith a conservative default and a way for applications to raise it. CPython raises `ValueError` when the limit is exceeded, which maps cleanly onto a 400.\n\nLowering the default body cap for urlencoded specifically would help too, since 100 MB of form fields is not a shape any real client sends.\n\n## Why I do not think this is a duplicate\n\nThe published tornado advisories cover out-of-bounds access in the C extension, unbounded accumulation of decompressed chunks in `AsyncHTTPClient`, the Authorization header surviving cross-origin redirects, credential leakage on curl handle reuse, and cookie attribute validation. The decompression one is the nearest in spirit and is on the client side; this is the server parsing a request body. The call is unchanged at HEAD.",
"id": "BREW-jupyterlab-CVE-2026-82397",
"modified": "2026-09-17T19:02:18Z",
"published": "2026-09-03T09:17:32Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/tornadoweb/tornado/security/advisories/GHSA-mpf4-983q-p7j4"
},
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-82397"
},
{
"type": "WEB",
"url": "https://github.com/tornadoweb/tornado/pull/3704"
},
{
"type": "WEB",
"url": "https://github.com/tornadoweb/tornado/commit/8d6363ed7b69d5f0da806efe34d256627a2191de"
},
{
"type": "PACKAGE",
"url": "https://github.com/tornadoweb/tornado"
},
{
"type": "WEB",
"url": "https://github.com/tornadoweb/tornado/releases/tag/v6.5.8"
}
],
"schema_version": "1.7.3",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H",
"type": "CVSS_V3"
}
],
"summary": " Tornado: Urlencoded body parsing omits max_num_fields, so one request can stall the event loop",
"upstream": [
"GHSA-mpf4-983q-p7j4",
"CVE-2026-82397",
"PYSEC-2026-3928"
]
}
BREW-LIVERELOAD-CVE-2026-82397 (GHSA-MPF4-983Q-P7J4)
Vulnerability from osv_homebrew – Published: 2026-09-03 09:21 – Updated: 2026-09-17 19:41 – Source websiteSummary
Tornado parses application/x-www-form-urlencoded bodies with urllib.parse.parse_qs and does not pass max_num_fields. A body made almost entirely of separators produces tens of millions of fields, and the parse happens on the event loop before the handler runs, so a single request stalls the whole server.
Where it is
tornado/escape.py, at HEAD e530031405e2154654dedc4c84d5656b557ea310:
result = urllib.parse.parse_qs(
qs, keep_blank_values, strict_parsing, encoding="latin1", errors="strict"
)
max_num_fields is the parameter CPython added for exactly this, and it is absent.
The path to it is entirely server-side and pre-dispatch. RequestHandler._execute parses the body at tornado/web.py:1821, which reaches HTTPServerRequest._parse_body at tornado/httputil.py:636, and the urlencoded branch of parse_body_arguments calls parse_qs_bytes at tornado/httputil.py:1030.
The size that reaches it is bounded only by the body cap, which defaults to the stream's max_buffer_size of 104857600 at tornado/iostream.py:239, applied as the request body default at tornado/http1connection.py:136-140. A 100 MB body of separators is around fifty million fields.
Impact
Denial of service against the whole process, not one request. Tornado is single-threaded and the parse is synchronous on the event loop, so every other connection waits. No authentication is needed if any route accepts a form post, which is the normal case.
Suggested fix
Pass a bound:
result = urllib.parse.parse_qs(
qs, keep_blank_values, strict_parsing, encoding="latin1", errors="strict",
max_num_fields=max_num_fields,
)
with a conservative default and a way for applications to raise it. CPython raises ValueError when the limit is exceeded, which maps cleanly onto a 400.
Lowering the default body cap for urlencoded specifically would help too, since 100 MB of form fields is not a shape any real client sends.
Why I do not think this is a duplicate
The published tornado advisories cover out-of-bounds access in the C extension, unbounded accumulation of decompressed chunks in AsyncHTTPClient, the Authorization header surviving cross-origin redirects, credential leakage on curl handle reuse, and cookie attribute validation. The decompression one is the nearest in spirit and is on the client side; this is the server parsing a request body. The call is unchanged at HEAD.
| URL | Type | |||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
||||||||||||||||||||
{
"affected": [
{
"ecosystem_specific": {
"fix": "bump",
"range_state": "fixed",
"resource": "tornado",
"resource_purl": "pkg:pypi/tornado@6.5.8",
"upstream_fixed_in": "6.5.8"
},
"package": {
"ecosystem": "Homebrew",
"name": "livereload",
"purl": "pkg:brew/livereload"
},
"ranges": [
{
"events": [
{
"introduced": "2.7.1"
},
{
"fixed": "2.7.1_3"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"database_specific": {
"confidence": "high",
"source": "matched",
"strategy": "registry",
"upstream_evidence": [
{
"ecosystem": "PyPI",
"key": "pkg:pypi/tornado@6.5.8",
"name": "tornado",
"resource": "tornado",
"strategy": "registry",
"subject_version": "6.5.8"
}
]
},
"details": "## Summary\n\nTornado parses `application/x-www-form-urlencoded` bodies with `urllib.parse.parse_qs` and does not pass `max_num_fields`. A body made almost entirely of separators produces tens of millions of fields, and the parse happens on the event loop before the handler runs, so a single request stalls the whole server.\n\n## Where it is\n\n`tornado/escape.py`, at HEAD `e530031405e2154654dedc4c84d5656b557ea310`:\n\n```python\nresult = urllib.parse.parse_qs(\n qs, keep_blank_values, strict_parsing, encoding=\"latin1\", errors=\"strict\"\n)\n```\n\n`max_num_fields` is the parameter CPython added for exactly this, and it is absent.\n\nThe path to it is entirely server-side and pre-dispatch. `RequestHandler._execute` parses the body at `tornado/web.py:1821`, which reaches `HTTPServerRequest._parse_body` at `tornado/httputil.py:636`, and the urlencoded branch of `parse_body_arguments` calls `parse_qs_bytes` at `tornado/httputil.py:1030`.\n\nThe size that reaches it is bounded only by the body cap, which defaults to the stream\u0027s `max_buffer_size` of 104857600 at `tornado/iostream.py:239`, applied as the request body default at `tornado/http1connection.py:136-140`. A 100 MB body of separators is around fifty million fields.\n\n## Impact\n\nDenial of service against the whole process, not one request. Tornado is single-threaded and the parse is synchronous on the event loop, so every other connection waits. No authentication is needed if any route accepts a form post, which is the normal case.\n\n## Suggested fix\n\nPass a bound:\n\n```python\nresult = urllib.parse.parse_qs(\n qs, keep_blank_values, strict_parsing, encoding=\"latin1\", errors=\"strict\",\n max_num_fields=max_num_fields,\n)\n```\n\nwith a conservative default and a way for applications to raise it. CPython raises `ValueError` when the limit is exceeded, which maps cleanly onto a 400.\n\nLowering the default body cap for urlencoded specifically would help too, since 100 MB of form fields is not a shape any real client sends.\n\n## Why I do not think this is a duplicate\n\nThe published tornado advisories cover out-of-bounds access in the C extension, unbounded accumulation of decompressed chunks in `AsyncHTTPClient`, the Authorization header surviving cross-origin redirects, credential leakage on curl handle reuse, and cookie attribute validation. The decompression one is the nearest in spirit and is on the client side; this is the server parsing a request body. The call is unchanged at HEAD.",
"id": "BREW-livereload-CVE-2026-82397",
"modified": "2026-09-17T19:41:55Z",
"published": "2026-09-03T09:21:24Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/tornadoweb/tornado/security/advisories/GHSA-mpf4-983q-p7j4"
},
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-82397"
},
{
"type": "WEB",
"url": "https://github.com/tornadoweb/tornado/pull/3704"
},
{
"type": "WEB",
"url": "https://github.com/tornadoweb/tornado/commit/8d6363ed7b69d5f0da806efe34d256627a2191de"
},
{
"type": "PACKAGE",
"url": "https://github.com/tornadoweb/tornado"
},
{
"type": "WEB",
"url": "https://github.com/tornadoweb/tornado/releases/tag/v6.5.8"
}
],
"schema_version": "1.7.3",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H",
"type": "CVSS_V3"
}
],
"summary": " Tornado: Urlencoded body parsing omits max_num_fields, so one request can stall the event loop",
"upstream": [
"GHSA-mpf4-983q-p7j4",
"CVE-2026-82397",
"PYSEC-2026-3928"
]
}
BREW-NVCHECKER-CVE-2026-82397 (GHSA-MPF4-983Q-P7J4)
Vulnerability from osv_homebrew – Published: 2026-09-03 09:39 – Updated: 2026-09-17 18:30 – Source websiteSummary
Tornado parses application/x-www-form-urlencoded bodies with urllib.parse.parse_qs and does not pass max_num_fields. A body made almost entirely of separators produces tens of millions of fields, and the parse happens on the event loop before the handler runs, so a single request stalls the whole server.
Where it is
tornado/escape.py, at HEAD e530031405e2154654dedc4c84d5656b557ea310:
result = urllib.parse.parse_qs(
qs, keep_blank_values, strict_parsing, encoding="latin1", errors="strict"
)
max_num_fields is the parameter CPython added for exactly this, and it is absent.
The path to it is entirely server-side and pre-dispatch. RequestHandler._execute parses the body at tornado/web.py:1821, which reaches HTTPServerRequest._parse_body at tornado/httputil.py:636, and the urlencoded branch of parse_body_arguments calls parse_qs_bytes at tornado/httputil.py:1030.
The size that reaches it is bounded only by the body cap, which defaults to the stream's max_buffer_size of 104857600 at tornado/iostream.py:239, applied as the request body default at tornado/http1connection.py:136-140. A 100 MB body of separators is around fifty million fields.
Impact
Denial of service against the whole process, not one request. Tornado is single-threaded and the parse is synchronous on the event loop, so every other connection waits. No authentication is needed if any route accepts a form post, which is the normal case.
Suggested fix
Pass a bound:
result = urllib.parse.parse_qs(
qs, keep_blank_values, strict_parsing, encoding="latin1", errors="strict",
max_num_fields=max_num_fields,
)
with a conservative default and a way for applications to raise it. CPython raises ValueError when the limit is exceeded, which maps cleanly onto a 400.
Lowering the default body cap for urlencoded specifically would help too, since 100 MB of form fields is not a shape any real client sends.
Why I do not think this is a duplicate
The published tornado advisories cover out-of-bounds access in the C extension, unbounded accumulation of decompressed chunks in AsyncHTTPClient, the Authorization header surviving cross-origin redirects, credential leakage on curl handle reuse, and cookie attribute validation. The decompression one is the nearest in spirit and is on the client side; this is the server parsing a request body. The call is unchanged at HEAD.
| URL | Type | |||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
||||||||||||||||||||
{
"affected": [
{
"ecosystem_specific": {
"fix": "bump",
"range_state": "fixed",
"resource": "tornado",
"resource_purl": "pkg:pypi/tornado@6.5.8",
"upstream_fixed_in": "6.5.8"
},
"package": {
"ecosystem": "Homebrew",
"name": "nvchecker",
"purl": "pkg:brew/nvchecker"
},
"ranges": [
{
"events": [
{
"introduced": "2.7"
},
{
"fixed": "2.22"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"database_specific": {
"confidence": "high",
"source": "matched",
"strategy": "registry",
"upstream_evidence": [
{
"ecosystem": "PyPI",
"key": "pkg:pypi/tornado@6.5.8",
"name": "tornado",
"resource": "tornado",
"strategy": "registry",
"subject_version": "6.5.8"
}
]
},
"details": "## Summary\n\nTornado parses `application/x-www-form-urlencoded` bodies with `urllib.parse.parse_qs` and does not pass `max_num_fields`. A body made almost entirely of separators produces tens of millions of fields, and the parse happens on the event loop before the handler runs, so a single request stalls the whole server.\n\n## Where it is\n\n`tornado/escape.py`, at HEAD `e530031405e2154654dedc4c84d5656b557ea310`:\n\n```python\nresult = urllib.parse.parse_qs(\n qs, keep_blank_values, strict_parsing, encoding=\"latin1\", errors=\"strict\"\n)\n```\n\n`max_num_fields` is the parameter CPython added for exactly this, and it is absent.\n\nThe path to it is entirely server-side and pre-dispatch. `RequestHandler._execute` parses the body at `tornado/web.py:1821`, which reaches `HTTPServerRequest._parse_body` at `tornado/httputil.py:636`, and the urlencoded branch of `parse_body_arguments` calls `parse_qs_bytes` at `tornado/httputil.py:1030`.\n\nThe size that reaches it is bounded only by the body cap, which defaults to the stream\u0027s `max_buffer_size` of 104857600 at `tornado/iostream.py:239`, applied as the request body default at `tornado/http1connection.py:136-140`. A 100 MB body of separators is around fifty million fields.\n\n## Impact\n\nDenial of service against the whole process, not one request. Tornado is single-threaded and the parse is synchronous on the event loop, so every other connection waits. No authentication is needed if any route accepts a form post, which is the normal case.\n\n## Suggested fix\n\nPass a bound:\n\n```python\nresult = urllib.parse.parse_qs(\n qs, keep_blank_values, strict_parsing, encoding=\"latin1\", errors=\"strict\",\n max_num_fields=max_num_fields,\n)\n```\n\nwith a conservative default and a way for applications to raise it. CPython raises `ValueError` when the limit is exceeded, which maps cleanly onto a 400.\n\nLowering the default body cap for urlencoded specifically would help too, since 100 MB of form fields is not a shape any real client sends.\n\n## Why I do not think this is a duplicate\n\nThe published tornado advisories cover out-of-bounds access in the C extension, unbounded accumulation of decompressed chunks in `AsyncHTTPClient`, the Authorization header surviving cross-origin redirects, credential leakage on curl handle reuse, and cookie attribute validation. The decompression one is the nearest in spirit and is on the client side; this is the server parsing a request body. The call is unchanged at HEAD.",
"id": "BREW-nvchecker-CVE-2026-82397",
"modified": "2026-09-17T18:30:47Z",
"published": "2026-09-03T09:39:33Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/tornadoweb/tornado/security/advisories/GHSA-mpf4-983q-p7j4"
},
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-82397"
},
{
"type": "WEB",
"url": "https://github.com/tornadoweb/tornado/pull/3704"
},
{
"type": "WEB",
"url": "https://github.com/tornadoweb/tornado/commit/8d6363ed7b69d5f0da806efe34d256627a2191de"
},
{
"type": "PACKAGE",
"url": "https://github.com/tornadoweb/tornado"
},
{
"type": "WEB",
"url": "https://github.com/tornadoweb/tornado/releases/tag/v6.5.8"
}
],
"schema_version": "1.7.3",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H",
"type": "CVSS_V3"
}
],
"summary": " Tornado: Urlencoded body parsing omits max_num_fields, so one request can stall the event loop",
"upstream": [
"GHSA-mpf4-983q-p7j4",
"CVE-2026-82397",
"PYSEC-2026-3928"
]
}
BREW-SICKCHILL-CVE-2026-82397 (GHSA-MPF4-983Q-P7J4)
Vulnerability from osv_homebrew – Published: 2026-09-03 10:06 – Updated: 2026-09-17 18:24 – Source websiteSummary
Tornado parses application/x-www-form-urlencoded bodies with urllib.parse.parse_qs and does not pass max_num_fields. A body made almost entirely of separators produces tens of millions of fields, and the parse happens on the event loop before the handler runs, so a single request stalls the whole server.
Where it is
tornado/escape.py, at HEAD e530031405e2154654dedc4c84d5656b557ea310:
result = urllib.parse.parse_qs(
qs, keep_blank_values, strict_parsing, encoding="latin1", errors="strict"
)
max_num_fields is the parameter CPython added for exactly this, and it is absent.
The path to it is entirely server-side and pre-dispatch. RequestHandler._execute parses the body at tornado/web.py:1821, which reaches HTTPServerRequest._parse_body at tornado/httputil.py:636, and the urlencoded branch of parse_body_arguments calls parse_qs_bytes at tornado/httputil.py:1030.
The size that reaches it is bounded only by the body cap, which defaults to the stream's max_buffer_size of 104857600 at tornado/iostream.py:239, applied as the request body default at tornado/http1connection.py:136-140. A 100 MB body of separators is around fifty million fields.
Impact
Denial of service against the whole process, not one request. Tornado is single-threaded and the parse is synchronous on the event loop, so every other connection waits. No authentication is needed if any route accepts a form post, which is the normal case.
Suggested fix
Pass a bound:
result = urllib.parse.parse_qs(
qs, keep_blank_values, strict_parsing, encoding="latin1", errors="strict",
max_num_fields=max_num_fields,
)
with a conservative default and a way for applications to raise it. CPython raises ValueError when the limit is exceeded, which maps cleanly onto a 400.
Lowering the default body cap for urlencoded specifically would help too, since 100 MB of form fields is not a shape any real client sends.
Why I do not think this is a duplicate
The published tornado advisories cover out-of-bounds access in the C extension, unbounded accumulation of decompressed chunks in AsyncHTTPClient, the Authorization header surviving cross-origin redirects, credential leakage on curl handle reuse, and cookie attribute validation. The decompression one is the nearest in spirit and is on the client side; this is the server parsing a request body. The call is unchanged at HEAD.
| URL | Type | |||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
||||||||||||||||||||
{
"affected": [
{
"ecosystem_specific": {
"fix": "bump",
"range_state": "fixed",
"resource": "tornado",
"resource_purl": "pkg:pypi/tornado@6.5.8",
"upstream_fixed_in": "6.5.8"
},
"package": {
"ecosystem": "Homebrew",
"name": "sickchill",
"purl": "pkg:brew/sickchill"
},
"ranges": [
{
"events": [
{
"introduced": "2023.5.30"
},
{
"fixed": "2024.3.1_8"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"database_specific": {
"confidence": "high",
"source": "matched",
"strategy": "registry",
"upstream_evidence": [
{
"ecosystem": "PyPI",
"key": "pkg:pypi/tornado@6.5.8",
"name": "tornado",
"resource": "tornado",
"strategy": "registry",
"subject_version": "6.5.8"
}
]
},
"details": "## Summary\n\nTornado parses `application/x-www-form-urlencoded` bodies with `urllib.parse.parse_qs` and does not pass `max_num_fields`. A body made almost entirely of separators produces tens of millions of fields, and the parse happens on the event loop before the handler runs, so a single request stalls the whole server.\n\n## Where it is\n\n`tornado/escape.py`, at HEAD `e530031405e2154654dedc4c84d5656b557ea310`:\n\n```python\nresult = urllib.parse.parse_qs(\n qs, keep_blank_values, strict_parsing, encoding=\"latin1\", errors=\"strict\"\n)\n```\n\n`max_num_fields` is the parameter CPython added for exactly this, and it is absent.\n\nThe path to it is entirely server-side and pre-dispatch. `RequestHandler._execute` parses the body at `tornado/web.py:1821`, which reaches `HTTPServerRequest._parse_body` at `tornado/httputil.py:636`, and the urlencoded branch of `parse_body_arguments` calls `parse_qs_bytes` at `tornado/httputil.py:1030`.\n\nThe size that reaches it is bounded only by the body cap, which defaults to the stream\u0027s `max_buffer_size` of 104857600 at `tornado/iostream.py:239`, applied as the request body default at `tornado/http1connection.py:136-140`. A 100 MB body of separators is around fifty million fields.\n\n## Impact\n\nDenial of service against the whole process, not one request. Tornado is single-threaded and the parse is synchronous on the event loop, so every other connection waits. No authentication is needed if any route accepts a form post, which is the normal case.\n\n## Suggested fix\n\nPass a bound:\n\n```python\nresult = urllib.parse.parse_qs(\n qs, keep_blank_values, strict_parsing, encoding=\"latin1\", errors=\"strict\",\n max_num_fields=max_num_fields,\n)\n```\n\nwith a conservative default and a way for applications to raise it. CPython raises `ValueError` when the limit is exceeded, which maps cleanly onto a 400.\n\nLowering the default body cap for urlencoded specifically would help too, since 100 MB of form fields is not a shape any real client sends.\n\n## Why I do not think this is a duplicate\n\nThe published tornado advisories cover out-of-bounds access in the C extension, unbounded accumulation of decompressed chunks in `AsyncHTTPClient`, the Authorization header surviving cross-origin redirects, credential leakage on curl handle reuse, and cookie attribute validation. The decompression one is the nearest in spirit and is on the client side; this is the server parsing a request body. The call is unchanged at HEAD.",
"id": "BREW-sickchill-CVE-2026-82397",
"modified": "2026-09-17T18:24:58Z",
"published": "2026-09-03T10:06:45Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/tornadoweb/tornado/security/advisories/GHSA-mpf4-983q-p7j4"
},
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-82397"
},
{
"type": "WEB",
"url": "https://github.com/tornadoweb/tornado/pull/3704"
},
{
"type": "WEB",
"url": "https://github.com/tornadoweb/tornado/commit/8d6363ed7b69d5f0da806efe34d256627a2191de"
},
{
"type": "PACKAGE",
"url": "https://github.com/tornadoweb/tornado"
},
{
"type": "WEB",
"url": "https://github.com/tornadoweb/tornado/releases/tag/v6.5.8"
}
],
"schema_version": "1.7.3",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H",
"type": "CVSS_V3"
}
],
"summary": " Tornado: Urlencoded body parsing omits max_num_fields, so one request can stall the event loop",
"upstream": [
"GHSA-mpf4-983q-p7j4",
"CVE-2026-82397",
"PYSEC-2026-3928"
]
}
BREW-SNAKEVIZ-CVE-2026-82397 (GHSA-MPF4-983Q-P7J4)
Vulnerability from osv_homebrew – Published: 2026-09-03 10:11 – Updated: 2026-09-17 19:58 – Source websiteSummary
Tornado parses application/x-www-form-urlencoded bodies with urllib.parse.parse_qs and does not pass max_num_fields. A body made almost entirely of separators produces tens of millions of fields, and the parse happens on the event loop before the handler runs, so a single request stalls the whole server.
Where it is
tornado/escape.py, at HEAD e530031405e2154654dedc4c84d5656b557ea310:
result = urllib.parse.parse_qs(
qs, keep_blank_values, strict_parsing, encoding="latin1", errors="strict"
)
max_num_fields is the parameter CPython added for exactly this, and it is absent.
The path to it is entirely server-side and pre-dispatch. RequestHandler._execute parses the body at tornado/web.py:1821, which reaches HTTPServerRequest._parse_body at tornado/httputil.py:636, and the urlencoded branch of parse_body_arguments calls parse_qs_bytes at tornado/httputil.py:1030.
The size that reaches it is bounded only by the body cap, which defaults to the stream's max_buffer_size of 104857600 at tornado/iostream.py:239, applied as the request body default at tornado/http1connection.py:136-140. A 100 MB body of separators is around fifty million fields.
Impact
Denial of service against the whole process, not one request. Tornado is single-threaded and the parse is synchronous on the event loop, so every other connection waits. No authentication is needed if any route accepts a form post, which is the normal case.
Suggested fix
Pass a bound:
result = urllib.parse.parse_qs(
qs, keep_blank_values, strict_parsing, encoding="latin1", errors="strict",
max_num_fields=max_num_fields,
)
with a conservative default and a way for applications to raise it. CPython raises ValueError when the limit is exceeded, which maps cleanly onto a 400.
Lowering the default body cap for urlencoded specifically would help too, since 100 MB of form fields is not a shape any real client sends.
Why I do not think this is a duplicate
The published tornado advisories cover out-of-bounds access in the C extension, unbounded accumulation of decompressed chunks in AsyncHTTPClient, the Authorization header surviving cross-origin redirects, credential leakage on curl handle reuse, and cookie attribute validation. The decompression one is the nearest in spirit and is on the client side; this is the server parsing a request body. The call is unchanged at HEAD.
| URL | Type | |||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
||||||||||||||||||||
{
"affected": [
{
"ecosystem_specific": {
"fix": "bump",
"range_state": "fixed",
"resource": "tornado",
"resource_purl": "pkg:pypi/tornado@6.5.8",
"upstream_fixed_in": "6.5.8"
},
"package": {
"ecosystem": "Homebrew",
"name": "snakeviz",
"purl": "pkg:brew/snakeviz"
},
"ranges": [
{
"events": [
{
"introduced": "2.2.0"
},
{
"fixed": "2.2.2_5"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"database_specific": {
"confidence": "high",
"source": "matched",
"strategy": "registry",
"upstream_evidence": [
{
"ecosystem": "PyPI",
"key": "pkg:pypi/tornado@6.5.8",
"name": "tornado",
"resource": "tornado",
"strategy": "registry",
"subject_version": "6.5.8"
}
]
},
"details": "## Summary\n\nTornado parses `application/x-www-form-urlencoded` bodies with `urllib.parse.parse_qs` and does not pass `max_num_fields`. A body made almost entirely of separators produces tens of millions of fields, and the parse happens on the event loop before the handler runs, so a single request stalls the whole server.\n\n## Where it is\n\n`tornado/escape.py`, at HEAD `e530031405e2154654dedc4c84d5656b557ea310`:\n\n```python\nresult = urllib.parse.parse_qs(\n qs, keep_blank_values, strict_parsing, encoding=\"latin1\", errors=\"strict\"\n)\n```\n\n`max_num_fields` is the parameter CPython added for exactly this, and it is absent.\n\nThe path to it is entirely server-side and pre-dispatch. `RequestHandler._execute` parses the body at `tornado/web.py:1821`, which reaches `HTTPServerRequest._parse_body` at `tornado/httputil.py:636`, and the urlencoded branch of `parse_body_arguments` calls `parse_qs_bytes` at `tornado/httputil.py:1030`.\n\nThe size that reaches it is bounded only by the body cap, which defaults to the stream\u0027s `max_buffer_size` of 104857600 at `tornado/iostream.py:239`, applied as the request body default at `tornado/http1connection.py:136-140`. A 100 MB body of separators is around fifty million fields.\n\n## Impact\n\nDenial of service against the whole process, not one request. Tornado is single-threaded and the parse is synchronous on the event loop, so every other connection waits. No authentication is needed if any route accepts a form post, which is the normal case.\n\n## Suggested fix\n\nPass a bound:\n\n```python\nresult = urllib.parse.parse_qs(\n qs, keep_blank_values, strict_parsing, encoding=\"latin1\", errors=\"strict\",\n max_num_fields=max_num_fields,\n)\n```\n\nwith a conservative default and a way for applications to raise it. CPython raises `ValueError` when the limit is exceeded, which maps cleanly onto a 400.\n\nLowering the default body cap for urlencoded specifically would help too, since 100 MB of form fields is not a shape any real client sends.\n\n## Why I do not think this is a duplicate\n\nThe published tornado advisories cover out-of-bounds access in the C extension, unbounded accumulation of decompressed chunks in `AsyncHTTPClient`, the Authorization header surviving cross-origin redirects, credential leakage on curl handle reuse, and cookie attribute validation. The decompression one is the nearest in spirit and is on the client side; this is the server parsing a request body. The call is unchanged at HEAD.",
"id": "BREW-snakeviz-CVE-2026-82397",
"modified": "2026-09-17T19:58:46Z",
"published": "2026-09-03T10:11:36Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/tornadoweb/tornado/security/advisories/GHSA-mpf4-983q-p7j4"
},
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-82397"
},
{
"type": "WEB",
"url": "https://github.com/tornadoweb/tornado/pull/3704"
},
{
"type": "WEB",
"url": "https://github.com/tornadoweb/tornado/commit/8d6363ed7b69d5f0da806efe34d256627a2191de"
},
{
"type": "PACKAGE",
"url": "https://github.com/tornadoweb/tornado"
},
{
"type": "WEB",
"url": "https://github.com/tornadoweb/tornado/releases/tag/v6.5.8"
}
],
"schema_version": "1.7.3",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H",
"type": "CVSS_V3"
}
],
"summary": " Tornado: Urlencoded body parsing omits max_num_fields, so one request can stall the event loop",
"upstream": [
"GHSA-mpf4-983q-p7j4",
"CVE-2026-82397",
"PYSEC-2026-3928"
]
}
CVE-2026-82397 (GCVE-0-2026-82397)
Vulnerability from cvelistv5 – Published: 2026-08-31 21:28 – Updated: 2026-09-02 17:39| URL | Tags |
|---|---|
| https://github.com/tornadoweb/tornado/security/ad… | x_refsource_CONFIRM |
| https://github.com/tornadoweb/tornado/pull/3704 | x_refsource_MISC |
| https://github.com/tornadoweb/tornado/commit/8d63… | x_refsource_MISC |
| https://github.com/tornadoweb/tornado/releases/ta… | x_refsource_MISC |
| Vendor | Product | Version | |
|---|---|---|---|
| tornadoweb | tornado |
Affected:
< 6.5.8
|
{
"containers": {
"adp": [
{
"metrics": [
{
"other": {
"content": {
"id": "CVE-2026-82397",
"options": [
{
"Exploitation": "none"
},
{
"Automatable": "yes"
},
{
"Technical Impact": "partial"
}
],
"role": "CISA Coordinator",
"timestamp": "2026-09-02T17:39:28.982654Z",
"version": "2.0.3"
},
"type": "ssvc"
}
}
],
"providerMetadata": {
"dateUpdated": "2026-09-02T17:39:41.238Z",
"orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
"shortName": "CISA-ADP"
},
"title": "CISA ADP Vulnrichment"
}
],
"cna": {
"affected": [
{
"product": "tornado",
"vendor": "tornadoweb",
"versions": [
{
"status": "affected",
"version": "\u003c 6.5.8"
}
]
}
],
"descriptions": [
{
"lang": "en",
"value": "Tornado is a Python web framework and asynchronous networking library. Prior to 6.5.8, Tornado parses application/x-www-form-urlencoded request bodies with urllib.parse.parse_qs in tornado/escape.py without passing max_num_fields. RequestHandler._execute in tornado/web.py parses the body before handler dispatch through HTTPServerRequest._parse_body and parse_body_arguments in tornado/httputil.py, so an unauthenticated request body containing millions of separator-delimited fields can synchronously stall the single-threaded event loop and delay every connection. The body is bounded only by max_buffer_size, which defaults to 104857600 bytes. This issue is fixed in version 6.5.8."
}
],
"metrics": [
{
"cvssV3_1": {
"attackComplexity": "LOW",
"attackVector": "NETWORK",
"availabilityImpact": "HIGH",
"baseScore": 7.5,
"baseSeverity": "HIGH",
"confidentialityImpact": "NONE",
"integrityImpact": "NONE",
"privilegesRequired": "NONE",
"scope": "UNCHANGED",
"userInteraction": "NONE",
"vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H",
"version": "3.1"
}
}
],
"problemTypes": [
{
"descriptions": [
{
"cweId": "CWE-400",
"description": "CWE-400: Uncontrolled Resource Consumption",
"lang": "en",
"type": "CWE"
}
]
},
{
"descriptions": [
{
"cweId": "CWE-1284",
"description": "CWE-1284: Improper Validation of Specified Quantity in Input",
"lang": "en",
"type": "CWE"
}
]
}
],
"providerMetadata": {
"dateUpdated": "2026-08-31T21:28:13.958Z",
"orgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
"shortName": "GitHub_M"
},
"references": [
{
"name": "https://github.com/tornadoweb/tornado/security/advisories/GHSA-mpf4-983q-p7j4",
"tags": [
"x_refsource_CONFIRM"
],
"url": "https://github.com/tornadoweb/tornado/security/advisories/GHSA-mpf4-983q-p7j4"
},
{
"name": "https://github.com/tornadoweb/tornado/pull/3704",
"tags": [
"x_refsource_MISC"
],
"url": "https://github.com/tornadoweb/tornado/pull/3704"
},
{
"name": "https://github.com/tornadoweb/tornado/commit/8d6363ed7b69d5f0da806efe34d256627a2191de",
"tags": [
"x_refsource_MISC"
],
"url": "https://github.com/tornadoweb/tornado/commit/8d6363ed7b69d5f0da806efe34d256627a2191de"
},
{
"name": "https://github.com/tornadoweb/tornado/releases/tag/v6.5.8",
"tags": [
"x_refsource_MISC"
],
"url": "https://github.com/tornadoweb/tornado/releases/tag/v6.5.8"
}
],
"source": {
"advisory": "GHSA-mpf4-983q-p7j4",
"discovery": "UNKNOWN"
},
"title": "Tornado: Urlencoded body parsing omits max_num_fields, so one request can stall the event loop"
}
},
"cveMetadata": {
"assignerOrgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
"assignerShortName": "GitHub_M",
"cveId": "CVE-2026-82397",
"datePublished": "2026-08-31T21:28:13.958Z",
"dateReserved": "2026-08-28T22:00:43.512Z",
"dateUpdated": "2026-09-02T17:39:41.238Z",
"state": "PUBLISHED"
},
"dataType": "CVE_RECORD",
"dataVersion": "5.2"
}
PYSEC-2026-3928
Vulnerability from pysec - Published: 2026-09-10 09:44 - Updated: 2026-09-10 11:02Summary
Tornado parses application/x-www-form-urlencoded bodies with urllib.parse.parse_qs and does not pass max_num_fields. A body made almost entirely of separators produces tens of millions of fields, and the parse happens on the event loop before the handler runs, so a single request stalls the whole server.
Where it is
tornado/escape.py, at HEAD e530031405e2154654dedc4c84d5656b557ea310:
result = urllib.parse.parse_qs(
qs, keep_blank_values, strict_parsing, encoding="latin1", errors="strict"
)
max_num_fields is the parameter CPython added for exactly this, and it is absent.
The path to it is entirely server-side and pre-dispatch. RequestHandler._execute parses the body at tornado/web.py:1821, which reaches HTTPServerRequest._parse_body at tornado/httputil.py:636, and the urlencoded branch of parse_body_arguments calls parse_qs_bytes at tornado/httputil.py:1030.
The size that reaches it is bounded only by the body cap, which defaults to the stream's max_buffer_size of 104857600 at tornado/iostream.py:239, applied as the request body default at tornado/http1connection.py:136-140. A 100 MB body of separators is around fifty million fields.
Impact
Denial of service against the whole process, not one request. Tornado is single-threaded and the parse is synchronous on the event loop, so every other connection waits. No authentication is needed if any route accepts a form post, which is the normal case.
Suggested fix
Pass a bound:
result = urllib.parse.parse_qs(
qs, keep_blank_values, strict_parsing, encoding="latin1", errors="strict",
max_num_fields=max_num_fields,
)
with a conservative default and a way for applications to raise it. CPython raises ValueError when the limit is exceeded, which maps cleanly onto a 400.
Lowering the default body cap for urlencoded specifically would help too, since 100 MB of form fields is not a shape any real client sends.
Why I do not think this is a duplicate
The published tornado advisories cover out-of-bounds access in the C extension, unbounded accumulation of decompressed chunks in AsyncHTTPClient, the Authorization header surviving cross-origin redirects, credential leakage on curl handle reuse, and cookie attribute validation. The decompression one is the nearest in spirit and is on the client side; this is the server parsing a request body. The call is unchanged at HEAD.
| Name | purl | tornado | pkg:pypi/tornado |
|---|
{
"affected": [
{
"package": {
"ecosystem": "PyPI",
"name": "tornado",
"purl": "pkg:pypi/tornado"
},
"ranges": [
{
"events": [
{
"introduced": "0"
},
{
"fixed": "6.5.8"
}
],
"type": "ECOSYSTEM"
}
],
"versions": [
"0.2",
"1.0",
"1.1",
"1.1.1",
"1.2",
"1.2.1",
"2.0",
"2.1",
"2.1.1",
"2.2",
"2.2.1",
"2.3",
"2.4",
"2.4.1",
"3.0",
"3.0.1",
"3.0.2",
"3.1",
"3.1.1",
"3.2",
"3.2.1",
"3.2.2",
"4.0",
"4.0.1",
"4.0.2",
"4.1",
"4.1b2",
"4.2",
"4.2.1",
"4.2b1",
"4.3",
"4.3b1",
"4.3b2",
"4.4",
"4.4.1",
"4.4.2",
"4.4.3",
"4.4b1",
"4.5",
"4.5.1",
"4.5.2",
"4.5.3",
"4.5b1",
"4.5b2",
"5.0",
"5.0.1",
"5.0.2",
"5.0a1",
"5.0b1",
"5.1",
"5.1.1",
"5.1b1",
"6.0",
"6.0.1",
"6.0.2",
"6.0.3",
"6.0.4",
"6.0a1",
"6.0b1",
"6.1",
"6.1b1",
"6.1b2",
"6.2",
"6.2b1",
"6.2b2",
"6.3",
"6.3.1",
"6.3.2",
"6.3.3",
"6.3b1",
"6.4",
"6.4.1",
"6.4.2",
"6.4b1",
"6.5",
"6.5.1",
"6.5.2",
"6.5.3",
"6.5.4",
"6.5.5",
"6.5.6",
"6.5.7",
"6.5b1"
]
}
],
"aliases": [
"CVE-2026-82397",
"GHSA-mpf4-983q-p7j4"
],
"details": "## Summary\n\nTornado parses `application/x-www-form-urlencoded` bodies with `urllib.parse.parse_qs` and does not pass `max_num_fields`. A body made almost entirely of separators produces tens of millions of fields, and the parse happens on the event loop before the handler runs, so a single request stalls the whole server.\n\n## Where it is\n\n`tornado/escape.py`, at HEAD `e530031405e2154654dedc4c84d5656b557ea310`:\n\n```python\nresult = urllib.parse.parse_qs(\n qs, keep_blank_values, strict_parsing, encoding=\"latin1\", errors=\"strict\"\n)\n```\n\n`max_num_fields` is the parameter CPython added for exactly this, and it is absent.\n\nThe path to it is entirely server-side and pre-dispatch. `RequestHandler._execute` parses the body at `tornado/web.py:1821`, which reaches `HTTPServerRequest._parse_body` at `tornado/httputil.py:636`, and the urlencoded branch of `parse_body_arguments` calls `parse_qs_bytes` at `tornado/httputil.py:1030`.\n\nThe size that reaches it is bounded only by the body cap, which defaults to the stream\u0027s `max_buffer_size` of 104857600 at `tornado/iostream.py:239`, applied as the request body default at `tornado/http1connection.py:136-140`. A 100 MB body of separators is around fifty million fields.\n\n## Impact\n\nDenial of service against the whole process, not one request. Tornado is single-threaded and the parse is synchronous on the event loop, so every other connection waits. No authentication is needed if any route accepts a form post, which is the normal case.\n\n## Suggested fix\n\nPass a bound:\n\n```python\nresult = urllib.parse.parse_qs(\n qs, keep_blank_values, strict_parsing, encoding=\"latin1\", errors=\"strict\",\n max_num_fields=max_num_fields,\n)\n```\n\nwith a conservative default and a way for applications to raise it. CPython raises `ValueError` when the limit is exceeded, which maps cleanly onto a 400.\n\nLowering the default body cap for urlencoded specifically would help too, since 100 MB of form fields is not a shape any real client sends.\n\n## Why I do not think this is a duplicate\n\nThe published tornado advisories cover out-of-bounds access in the C extension, unbounded accumulation of decompressed chunks in `AsyncHTTPClient`, the Authorization header surviving cross-origin redirects, credential leakage on curl handle reuse, and cookie attribute validation. The decompression one is the nearest in spirit and is on the client side; this is the server parsing a request body. The call is unchanged at HEAD.",
"id": "PYSEC-2026-3928",
"modified": "2026-09-10T11:02:34.559608Z",
"published": "2026-09-10T09:44:59.496635Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/tornadoweb/tornado/security/advisories/GHSA-mpf4-983q-p7j4"
},
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-82397"
},
{
"type": "WEB",
"url": "https://github.com/tornadoweb/tornado/pull/3704"
},
{
"type": "WEB",
"url": "https://github.com/tornadoweb/tornado/commit/8d6363ed7b69d5f0da806efe34d256627a2191de"
},
{
"type": "PACKAGE",
"url": "https://github.com/tornadoweb/tornado"
},
{
"type": "WEB",
"url": "https://github.com/tornadoweb/tornado/releases/tag/v6.5.8"
},
{
"type": "PACKAGE",
"url": "https://pypi.org/project/tornado"
},
{
"type": "ADVISORY",
"url": "https://github.com/advisories/GHSA-mpf4-983q-p7j4"
}
],
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H",
"type": "CVSS_V3"
}
],
"summary": " Tornado: Urlencoded body parsing omits max_num_fields, so one request can stall the event loop"
}
Sightings
| Author | Source | Type | Date | Other |
|---|
Nomenclature
- Seen: The vulnerability was mentioned, discussed, or observed by the user.
- Confirmed: The vulnerability has been validated from an analyst's perspective.
- Published Proof of Concept: A public proof of concept is available for this vulnerability.
- Exploited: The vulnerability was observed as exploited by the user who reported the sighting.
- Patched: The vulnerability was observed as successfully patched by the user who reported the sighting.
- Not exploited: The vulnerability was not observed as exploited by the user who reported the sighting.
- Not confirmed: The user expressed doubt about the validity of the vulnerability.
- Not patched: The vulnerability was not observed as successfully patched by the user who reported the sighting.
The approach is described in our paper Mapping CVEs to MITRE ATT&CK Techniques: A Curated Gold-Set Classifier and the Limits of LLM-Assisted Label Expansion.
Browse all ATT&CK techniques and the vulnerabilities related to each.
Related by attack behaviour
Vulnerabilities whose description is nearest to this one in the vector space of the CIRCL/vulnerability-attack-technique-biencoder model. This is a similarity search over the bi-encoder space (plain cosine), not a classification, and it has no measured accuracy.