| Title | The Trusted Domain Project OpenDMARC 1.4.2 Improper Authentication |
|---|
| Description | OpenDMARC 1.4.2 fails to normalize certain Unicode characters in the domain of the visible `From:` address before performing a DMARC policy lookup. A crafted domain containing a Unicode character that maps to an ASCII domain separator, such as a fullwidth full stop, may therefore be treated as a different and policy-less domain. OpenDMARC returns `dmarc=none` instead of applying the DMARC policy published for the corresponding ASCII domain.
A remote unauthenticated attacker can exploit this issue by sending a forged message whose `From:` domain visually resembles a protected domain but uses such a Unicode character. No signing key or control of the victim’s DNS records is required. As a result, a published `p=reject` or `p=quarantine` policy may be bypassed, increasing the risk of email impersonation and phishing. |
|---|
| Source | ⚠️ https://weitongli.com/share/opendmarc-unicode-dot-bypass.html |
|---|
| User | WeitongLi (UID 100239) |
|---|
| Submission | 08/06/2026 12:53 (2 months ago) |
|---|
| Moderation | 09/27/2026 13:33 (2 months later) |
|---|
| Status | Accepted |
|---|
| VulDB entry | 410885 [Trusted Domain Project OpenDMARC up to 1.4.2 policy.c improper validation of unsafe equivalence in input] |
|---|
| Points | 20 |
|---|