When using the non-default "fallback" crypto back-end, ECC operations in node-jose can trigger a Denial-of-Service (DoS) condition, due to a possible infinite loop in an internal calculation. For some ECC operations, this condition is triggered randomly; for others, it can be triggered by malicious input.
Technical summary
The JOSE logic implemented by node-jose usually relies on an external cryptographic library for the underlying cryptographic primitives that JOSE operations require. When WebCrypto or the Node crypto module are available, they are used. When neither of these libraries is available, node-jose includes its own "fallback" implementations of some algorithms based on node-forge, in particular implementations of ECDH and ECDSA.
A various points, these algorithm implementations need to compute to the X coordinate of an elliptic curve point. This is done by calling the getX() method of the object representing the point, which is an alias of the function pointFpGetX() in lib/deps/ecc/math.js.
Computing the X coordinate from the form in which the point is stored requires computing the modular inverse of the Z coordinate, using the modInverse function from the jsbn library (e.g., this.z.modInverse(this.curve.p)). The output of this function call is multiplied by another value before being reduced with the barrettReduce() function.
The root cause of this issue is that the jsbnmodInverse function sometimes returns negative results. These results are correct in that they are equivalent mod the relevant modulus, but can be problematic for functions that expect modular operations to always return positive results (in the range [0, p), where p is the modulus).
In particular, while the Barrett reduction algorithm in general can handle negative inputs, the implementation in node-jose explicitly does not. Therefore, while the negative value that is returned by modInverse() is mathematically correct, it leads to an error in barrettReduce() causing an infinite loop which may result in a Denial of Service condition.
For a given prime modulus, we estimate that roughly one in every 2^20 inputs produce a negative modInverse(). This estimate was validated with exhaustive testing on small primes (<30 bits) and randomized testing with regard to the P-256 prime.
Impact
This issue is only present in situations where the "fallback" cryptographic implementation is being used, i.e., situations where neither WebCrypto nor the Node crypto module is available.
The following elliptic curve algorithms are impacted by this issue (all in lib/deps/ecc/index.js):
In the first three cases, the points being evaluated are generated randomly, so an attack could only arise due to a bad value being randomly selected (as noted above, with probability roughly 2^{-20}). In the latter two cases, the points being evaluated are provided from outside the library, and thus potentially by attackers.
Patches
Has the problem been patched? What versions should users upgrade to?
Workarounds
Since this issue is only present in the "fallback" crypto implementation, it can be avoided by ensuring that either WebCrypto or the Node crypto module is available in the JS environment where node-jose is being run.
cisco/node-jose (node-jose)
### [`v2.2.0`](https://togithub.com/cisco/node-jose/compare/v2.1.1...v2.2.0)
[Compare Source](https://togithub.com/cisco/node-jose/compare/v2.1.1...v2.2.0)
### [`v2.1.1`](https://togithub.com/cisco/node-jose/compare/v2.1.0...v2.1.1)
[Compare Source](https://togithub.com/cisco/node-jose/compare/v2.1.0...v2.1.1)
### [`v2.1.0`](https://togithub.com/cisco/node-jose/compare/2.0.0...v2.1.0)
[Compare Source](https://togithub.com/cisco/node-jose/compare/2.0.0...v2.1.0)
### [`v2.0.0`](https://togithub.com/cisco/node-jose/blob/HEAD/CHANGELOG.md#200-2020-09-20)
[Compare Source](https://togithub.com/cisco/node-jose/compare/1.1.4...2.0.0)
While all of the changes in this release are only dependency updates and otherwise patch-level changes, it does break compatibility with versions of node.js older than version 10.
##### Update
- switch to `karma-safari-applescript-launcher` for improved local Safari testing ([a0db723](https://togithub.com/cisco/node-jose/commit/a0db7230205d8bb6bfd8d06b7a1ffb4b036700fd))
- update dependencies for `karma`, `karma-mocha`, `mocha`, `webpack-stream`, and `yargs` ([96ee8ff](https://togithub.com/cisco/node-jose/commit/96ee8ff70dc9fbf08309cf4f5240f3affbffb16d))
- update `node-forge` to latest (shahar-h)([277aab5](https://togithub.com/cisco/node-jose/commit/277aab5c4c1c5b6b6c3923663d40aa9412f9f49d))
- cherry-pick `lodash` modules to effect smaller footprint (tychenjianjun)([003ef1c](https://togithub.com/cisco/node-jose/commit/003ef1c1488c5006976e4ead72bf40f63c4b098d))
- replace `browserify-zlib` with `pako` (taymoork2)([85610d6](https://togithub.com/cisco/node-jose/commit/85610d6fdf411c72f58d15197e6f935e0b20a695))
Configuration
📅 Schedule: Branch creation - "" (UTC), Automerge - At any time (no schedule defined).
🚦 Automerge: Disabled by config. Please merge this manually once you are satisfied.
♻ Rebasing: Whenever PR becomes conflicted, or you tick the rebase/retry checkbox.
🔕 Ignore: Close this PR and you won't be reminded about this update again.
[ ] If you want to rebase/retry this PR, check this box
This PR has been generated by Mend Renovate. View repository job log here.
This PR contains the following updates:
^1.1.4->^2.0.0GitHub Vulnerability Alerts
CVE-2023-25653
Description
When using the non-default "fallback" crypto back-end, ECC operations in
node-josecan trigger a Denial-of-Service (DoS) condition, due to a possible infinite loop in an internal calculation. For some ECC operations, this condition is triggered randomly; for others, it can be triggered by malicious input.Technical summary
The JOSE logic implemented by
node-joseusually relies on an external cryptographic library for the underlying cryptographic primitives that JOSE operations require. When WebCrypto or the Nodecryptomodule are available, they are used. When neither of these libraries is available,node-joseincludes its own "fallback" implementations of some algorithms based onnode-forge, in particular implementations of ECDH and ECDSA.A various points, these algorithm implementations need to compute to the X coordinate of an elliptic curve point. This is done by calling the
getX()method of the object representing the point, which is an alias of the functionpointFpGetX()inlib/deps/ecc/math.js.Computing the X coordinate from the form in which the point is stored requires computing the modular inverse of the Z coordinate, using the
modInversefunction from thejsbnlibrary (e.g.,this.z.modInverse(this.curve.p)). The output of this function call is multiplied by another value before being reduced with thebarrettReduce()function.The root cause of this issue is that the
jsbnmodInversefunction sometimes returns negative results. These results are correct in that they are equivalent mod the relevant modulus, but can be problematic for functions that expect modular operations to always return positive results (in the range[0, p), wherepis the modulus).In particular, while the Barrett reduction algorithm in general can handle negative inputs, the implementation in
node-joseexplicitly does not. Therefore, while the negative value that is returned bymodInverse()is mathematically correct, it leads to an error inbarrettReduce()causing an infinite loop which may result in a Denial of Service condition.For a given prime modulus, we estimate that roughly one in every
2^20inputs produce a negativemodInverse(). This estimate was validated with exhaustive testing on small primes (<30 bits) and randomized testing with regard to the P-256 prime.Impact
This issue is only present in situations where the "fallback" cryptographic implementation is being used, i.e., situations where neither WebCrypto nor the Node
cryptomodule is available.The following elliptic curve algorithms are impacted by this issue (all in
lib/deps/ecc/index.js):exports.generateKeyPair)ECPrivateKey.prototype.toPublicKey)ECPrivateKey.prototype.sign)ECPublicKey.prototype.verify)ECPrivateKey.prototype.computeSecret)In the first three cases, the points being evaluated are generated randomly, so an attack could only arise due to a bad value being randomly selected (as noted above, with probability roughly
2^{-20}). In the latter two cases, the points being evaluated are provided from outside the library, and thus potentially by attackers.Patches
Has the problem been patched? What versions should users upgrade to?
Workarounds
Since this issue is only present in the "fallback" crypto implementation, it can be avoided by ensuring that either WebCrypto or the Node
cryptomodule is available in the JS environment wherenode-joseis being run.References
For more information
If you have any questions or comments about this advisory:
Credits
Release Notes
cisco/node-jose (node-jose)
### [`v2.2.0`](https://togithub.com/cisco/node-jose/compare/v2.1.1...v2.2.0) [Compare Source](https://togithub.com/cisco/node-jose/compare/v2.1.1...v2.2.0) ### [`v2.1.1`](https://togithub.com/cisco/node-jose/compare/v2.1.0...v2.1.1) [Compare Source](https://togithub.com/cisco/node-jose/compare/v2.1.0...v2.1.1) ### [`v2.1.0`](https://togithub.com/cisco/node-jose/compare/2.0.0...v2.1.0) [Compare Source](https://togithub.com/cisco/node-jose/compare/2.0.0...v2.1.0) ### [`v2.0.0`](https://togithub.com/cisco/node-jose/blob/HEAD/CHANGELOG.md#200-2020-09-20) [Compare Source](https://togithub.com/cisco/node-jose/compare/1.1.4...2.0.0) While all of the changes in this release are only dependency updates and otherwise patch-level changes, it does break compatibility with versions of node.js older than version 10. ##### Update - switch to `karma-safari-applescript-launcher` for improved local Safari testing ([a0db723](https://togithub.com/cisco/node-jose/commit/a0db7230205d8bb6bfd8d06b7a1ffb4b036700fd)) - update dependencies for `karma`, `karma-mocha`, `mocha`, `webpack-stream`, and `yargs` ([96ee8ff](https://togithub.com/cisco/node-jose/commit/96ee8ff70dc9fbf08309cf4f5240f3affbffb16d)) - update `node-forge` to latest (shahar-h)([277aab5](https://togithub.com/cisco/node-jose/commit/277aab5c4c1c5b6b6c3923663d40aa9412f9f49d)) - cherry-pick `lodash` modules to effect smaller footprint (tychenjianjun)([003ef1c](https://togithub.com/cisco/node-jose/commit/003ef1c1488c5006976e4ead72bf40f63c4b098d)) - replace `browserify-zlib` with `pako` (taymoork2)([85610d6](https://togithub.com/cisco/node-jose/commit/85610d6fdf411c72f58d15197e6f935e0b20a695))Configuration
📅 Schedule: Branch creation - "" (UTC), Automerge - At any time (no schedule defined).
🚦 Automerge: Disabled by config. Please merge this manually once you are satisfied.
♻ Rebasing: Whenever PR becomes conflicted, or you tick the rebase/retry checkbox.
🔕 Ignore: Close this PR and you won't be reminded about this update again.
This PR has been generated by Mend Renovate. View repository job log here.