GHSA-hx5q-v6pj-533r: SAML authentication bypass due to missing validation on unsigned SAML messages
Impact When SAML is used as the authentication mechanism, Central Dogma accepts unsigned SAML messages (assertions, logout requests, etc.) as they are, rather than rejecting them by default. As a result, an attacker can forge a SAML message to authenticate themselves, despite the fact that such an unsigned SAML message should be rejected.
Published Feb 26, 2024 · Updated Mar 15, 2026
Affected packages
com.linecorp.centraldogma:centraldogma-server-auth-saml
- Ecosystem
- Maven
- Affected
- <= 0.64.2
- Fixed in
- 0.64.3
- Check
mvn dependency:tree -Dincludes=com.linecorp.centraldogma:centraldogma-server-auth-saml
Advisory details
From the GitHub Advisory Database (CC BY 4.0).
Impact
When SAML is used as the authentication mechanism, Central Dogma accepts unsigned SAML messages (assertions, logout requests, etc.) as they are, rather than rejecting them by default. As a result, an attacker can forge a SAML message to authenticate themselves, despite the fact that such an unsigned SAML message should be rejected.
Patches
The vulnerability has been patched in Central Dogma 0.64.3 by updating its Armeria dependency to 1.27.2. All users who use SAML as the authentication mechanism must upgrade from 0.64.3 or later.
Workarounds
A user can manually upgrade the armeria-saml module with the one from Armeria 1.27.2 or later, either by replacing the JAR in the Central Dogma distribution or by updating the dependency tree of the build.
References
Credited to lishiki.
References
- https://github.com/line/armeria/security/advisories/GHSA-4m6j-23p2-8c54
- https://github.com/line/centraldogma/security/advisories/GHSA-hx5q-v6pj-533r
- https://github.com/line/centraldogma/commit/16903426be2e954c050b3ee47b8c38ee3218f0eb
- https://github.com/line/centraldogma/commit/16903426be2e954c050b3ee47b8c38ee3218f0ebxz
- https://github.com/line/centraldogma/releases/tag/centraldogma-0.64.3
- https://github.com/advisories/GHSA-hx5q-v6pj-533r
How this page is made: the data panel comes from the GitHub Advisory Database, CISA’s KEV catalog, the NVD and FIRST EPSS, and updates twice a day. No analysis has been written for this entry yet. Methodology