GHSA-XF44-6RJP-V8CM
Vulnerability from github – Published: 2026-09-14 15:32 – Updated: 2026-09-14 21:31Description
Three separate mechanisms allowed a web page on an unrelated origin to read responses that Storm's HTTP components served to an authenticated user.
The Logviewer reflected the request's Origin header back in Access-Control-Allow-Origin while also
sending Access-Control-Allow-Credentials: true. The published security model documents a permissive
Access-Control-Allow-Origin: * posture as accepted, which is safe precisely because browsers refuse to
honour * together with credentials; reflecting the concrete origin removes that protection.
The shared CORS filter used by the UI, the Logviewer and DRPC was configured with a response header name where an initialisation parameter name was expected. The container ignored the setting and applied its own defaults, which allow credentials.
Finally, the UI and Logviewer wrapped API responses in a caller-supplied JSONP callback for every GET request. A script element on any origin can load such a response, which bypasses the same-origin policy entirely rather than negotiating it, and there was no way to turn the behaviour off.
In each case the effect is that a page visited by an authenticated operator can read cluster, topology and log data on their behalf.
Mitigation
Upgrade to 3.1.0, where the Logviewer no longer reflects the request origin in a credentialed response, the
CORS filter is configured explicitly, and JSONP wrapping is governed by ui.enable.jsonp, which defaults to
false.
Note that disabling JSONP is a behaviour change for tooling that passes a callback query parameter; such
tooling should be moved to ordinary JSON requests.
Users who cannot upgrade immediately should place the UI, Logviewer and DRPC HTTP endpoints behind a reverse
proxy that strips Access-Control-Allow-Origin and Access-Control-Allow-Credentials from responses and
rejects requests carrying a callback parameter.
Credit
The ASF -- found using Claude agents to study the security of open-source projects, validated and reported by Apache Storm.
{
"affected": [],
"aliases": [
"CVE-2026-82438"
],
"database_specific": {
"cwe_ids": [
"CWE-346"
],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2026-09-14T15:17:10Z",
"severity": "HIGH"
},
"details": "Description\n\nThree separate mechanisms allowed a web page on an unrelated origin to read responses that Storm\u0027s HTTP\ncomponents served to an authenticated user.\n\nThe Logviewer reflected the request\u0027s `Origin` header back in `Access-Control-Allow-Origin` while also\nsending `Access-Control-Allow-Credentials: true`. The published security model documents a permissive\n`Access-Control-Allow-Origin: *` posture as accepted, which is safe precisely because browsers refuse to\nhonour `*` together with credentials; reflecting the concrete origin removes that protection.\n\nThe shared CORS filter used by the UI, the Logviewer and DRPC was configured with a response header name\nwhere an initialisation parameter name was expected. The container ignored the setting and applied its own\ndefaults, which allow credentials.\n\nFinally, the UI and Logviewer wrapped API responses in a caller-supplied JSONP callback for every GET\nrequest. A script element on any origin can load such a response, which bypasses the same-origin policy\nentirely rather than negotiating it, and there was no way to turn the behaviour off.\n\nIn each case the effect is that a page visited by an authenticated operator can read cluster, topology and\nlog data on their behalf.\n\nMitigation\n\nUpgrade to 3.1.0, where the Logviewer no longer reflects the request origin in a credentialed response, the\nCORS filter is configured explicitly, and JSONP wrapping is governed by `ui.enable.jsonp`, which defaults to\nfalse.\n\nNote that disabling JSONP is a behaviour change for tooling that passes a `callback` query parameter; such\ntooling should be moved to ordinary JSON requests.\n\nUsers who cannot upgrade immediately should place the UI, Logviewer and DRPC HTTP endpoints behind a reverse\nproxy that strips `Access-Control-Allow-Origin` and `Access-Control-Allow-Credentials` from responses and\nrejects requests carrying a `callback` parameter.\n\nCredit\n\nThe ASF -- found using Claude agents to study the security of open-source projects, validated and reported by Apache Storm.",
"id": "GHSA-xf44-6rjp-v8cm",
"modified": "2026-09-14T21:31:41Z",
"published": "2026-09-14T15:32:49Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-82438"
},
{
"type": "WEB",
"url": "https://lists.apache.org/thread/2o7tl3hcdd865njxsn4d9cxp1frkctz3"
},
{
"type": "WEB",
"url": "http://www.openwall.com/lists/oss-security/2026/09/13/17"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:N",
"type": "CVSS_V3"
}
]
}
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.