An issue was discovered in Matrix Sydent before 1.0.3 and Synapse before 0.99.3.1. Random number generation is mishandled, which makes it easier for attackers to predict a Sydent authentication token or a Synapse random ID.
The product uses a Pseudo-Random Number Generator (PRNG) in a security context, but the PRNG’s algorithm is not cryptographically strong.
Name | Vendor | Start Version | End Version |
---|---|---|---|
Sydent | Matrix | * | 1.0.3 (excluding) |
Synapse | Matrix | * | 0.99.3.1 (excluding) |
Matrix-synapse | Ubuntu | bionic | * |
Matrix-synapse | Ubuntu | cosmic | * |
Matrix-synapse | Ubuntu | disco | * |
Matrix-synapse | Ubuntu | esm-apps/bionic | * |
Matrix-synapse | Ubuntu | upstream | * |
When a non-cryptographic PRNG is used in a cryptographic context, it can expose the cryptography to certain types of attacks. Often a pseudo-random number generator (PRNG) is not designed for cryptography. Sometimes a mediocre source of randomness is sufficient or preferable for algorithms that use random numbers. Weak generators generally take less processing power and/or do not use the precious, finite, entropy sources on a system. While such PRNGs might have very useful features, these same features could be used to break the cryptography.