I have recently worked on enabling 2-step authentication via SSH on the Gentoo developer machine. I have selected google-authenticator-libpam amongst different available implementations as it seemed the best maintained and having all the necessary features, including a friendly tool for users to configure it. However, its design has a weakness: it stores the secret unprotected in user’s home directory.
This means that if an attacker manages to gain at least temporary access to the filesystem with user’s privileges — through a malicious process, vulnerability or simply because someone left the computer unattended for a minute — he can trivially read the secret and therefore clone the token source without leaving a trace. It would completely defeat the purpose of the second step, and the user may not even notice until the attacker makes real use of the stolen secret.
In order to protect against this, I’ve created google-authenticator-wrappers (as upstream decided to ignore the problem). This package provides a rather trivial setuid wrapper that manages a write-only, authentication-protected secret store for the PAM module. Additionally, it comes with a test program (so you can test the OTP setup without jumping through the hoops or risking losing access) and friendly wrappers for the default setup, as used on Gentoo Infra.
The recommended setup (as utilized by sys-auth/google-authenticator-wrappers package) is to use a dedicated user for the password store. In this scenario, the users are unable to read their secrets, and all secret operations (including authentication via the PAM module) are done using an unprivileged user. Furthermore, any operation regarding the configuration (either updating it or removing the second step) require regular PAM authentication (e.g. typing your own password).
This is consistent with e.g. how shadow operates (users can’t read their passwords, nor update them without authenticating first), how most sites using 2-factor authentication operate (again, users can’t read their secrets) and follows the RFC 6238 recommendation (that
keys […] SHOULD be protected against unauthorized access and usage). It solves the aforementioned issue by preventing user-privileged processes from reading the secrets and recovery codes. Furthermore, it prevents the attacker with this particular level of access from disabling 2-step authentication, changing the secret or even weakening the configuration.