GHSA-r7vq-6425-j94w json
Python-TUF vulnerable to incorrect threshold signature computation for new root metadata
published 4y ago modified last yr.
- cvss
- —
- not scored
- cisa kev
- Not listed
- not in CISA's catalog
- epss
- —
- no score, EPSS covers CVEs only
- public exploits
- 0
- none known
Summary
### Impact
The function `_verify_root_self_signed()`, introduced in [v0.14.0](https://github.com/theupdateframework/tuf/releases/tag/v0.14.0), and which verifies self-signatures in a new root metadata file, counted multiple signatures by any new root key towards the new threshold. That is, any single new root key could theoretically provide enough signatures to meet the threshold for new key self-signatures required during root metadata update.
A scenario where this attack could be relevant is amazingly unlikely in practice to the point where labeling this issue as a security advisory is potentially overstating the impact of the issue. Given that new root keys only become trusted by the client after a successful root metadata update, which also requires the quorum of signatures from old trusted root keys, this issue has been evaluated as low in severity.
In particular, in order to exploit this vulnerability, an attacker must:
1. Control one new root key. 2. Craft a new root metadata file such that there is a number of signatures by this new root key greater than or equal to the new threshold. 3. Cause a valid threshold of the old root keys to sign this new root metadata file. 4. Cause this new root metadata file to be published on the repository. 5. Cause clients to rotate to this new root metadata file.
### Patches A [fix](https://github.com/theupdateframework/tuf/commit/83ac7be525b733f79a7e9bc573ec580ec835f179) is available since [version 0.16.0](https://github.com/theupdateframework/tuf/releases/tag/v0.16.0).
### Workarounds No workarounds are known for this issue.
### References * Pull request resolving the issue [PR 1218](https://github.com/theupdateframework/tuf/pull/1218) * Pull request which introduced the faulty computation of signature threshold on new root metadata [PR 1101](https://github.com/theupdateframework/tuf/pull/1101) * A similar previous issue with incorrectly computed signature thresholds in tuf is described in [GHSA-pwqf-9h7j-7mv8](https://github.com/theupdateframework/tuf/security/advisories/GHSA-pwqf-9h7j-7mv8)
Patch / fix
- pypi tuf 0.16.0
- github.com/theupdateframework/python-tuf#1218 pull request
- github.com/theupdateframework/tuf#1101 pull request
Is GHSA-r7vq-6425-j94w in your dependencies? Fencer shows which of your repositories are affected as soon as an advisory lands, and opens the fix as a pull request.
Check with FencerAffected packages
| ecosystem | package | affected | fixed in |
|---|---|---|---|
| pypi | tuf | >=0.14.0, <0.16.0 |
0.16.0 |
References
-
github.com
/theupdateframework/python-tuf/pull/1218
web
-
github.com
/theupdateframework/tuf/pull/1101
web
-
github.com
/theupdateframework/python-tuf/security/advisories/GHSA-pwqf-9h7j-7mv8
web
-
github.com
/theupdateframework/python-tuf/security/advisories/GHSA-r7vq-6425-j94w
web
-
github.com
/theupdateframework/python-tuf
package
Sources
- GitHub Advisory GHSA-r7vq-6425-j94w
Fencer Intel keeps one entry per vulnerability: the CVE id wins, then GHSA, and records that alias each other are merged, except that two CVEs always keep separate entries. The same data is available as JSON.