Executive brief
Vaultwarden, an open-source password management server, contains a flaw that allows attackers to bypass security measures designed to prevent repeated login attempts. By exploiting a specific email-based login feature, an attacker can rapidly test thousands of passwords against a user's account without being blocked. If successful, this could lead to unauthorized access to a user's entire vault of stored passwords and sensitive credentials.
Technical details
A brute-force protection bypass exists in Vaultwarden's 'send_email_login' function (located in email.rs) via the '/api/two-factor/send-email-login' endpoint. While standard login endpoints are protected by the 'check_limit_login' rate-limiting function, this specific 2FA endpoint was left unprotected. It acts as an authentication oracle: if a username/password combination is correct, it returns a 200 OK or a 'Two factor not found' message; if incorrect, it returns 'Username or password is incorrect'. This allows a network-based attacker to perform high-speed password guessing against any user on the server, regardless of whether that specific user has email 2FA enabled, provided the server itself has the email 2FA feature active. The issue is resolved in version 1.35.4 by implementing proper rate limiting on the affected endpoint.
Affected products
- Vaultwarden Vaultwarden < 1.35.4
Timeline
- 2026-02-23: patched: Fix merged in PR 6867 and released in version 1.35.4
- 2026-04-25: advisory: GitHub Security Advisory published
- 2026-05-11: disclosed: CVE published to NVD