Skip to content

Openfire has potential identity spoofing issue via unsafe CN parsing

Moderate severity GitHub Reviewed Published Sep 15, 2025 in igniterealtime/Openfire • Updated Sep 16, 2025

Package

maven org.igniterealtime.openfire:xmppserver (Maven)

Affected versions

< 5.0.2

Patched versions

5.0.2

Description

Summary

Identity spoofing in X.509 client certificate authentication in Openfire allows internal attackers to impersonate other users via crafted certificate subject attributes, due to regex-based extraction of CN from an unescaped, provider-dependent DN string.

Analysis

Openfire’s SASL EXTERNAL mechanism for client TLS authentication contains a vulnerability in how it extracts user identities from X.509 certificates. Instead of parsing the structured ASN.1 data, the code calls X509Certificate.getSubjectDN().getName() and applies a regex to look for CN=. This method produces a provider-dependent string that does not escape special characters. In SunJSSE (sun.security.x509.X500Name), for example, commas and equals signs inside attribute values are not escaped.

As a result, a malicious certificate can embed CN= inside another attribute value (e.g. OU="CN=admin,"). The regex will incorrectly interpret this as a legitimate Common Name and extract admin. If SASL EXTERNAL is enabled and configured to map CNs to user accounts, this allows the attacker to impersonate another user.

Impact

When there is no explicit configuration override, Openfire defaults to using certificate attributes as follows:

  • Server-to-server certificates: first the Subject Alternative Name (SAN), and if that is not available, the Common Name (CN).
  • Client-to-server certificates: only the Common Name (CN).

Use of CN for identity mapping was phased out by the CA/Browser Forum. CAB Forum-compliant CAs no longer allow arbitrary Subject RDNs and always include SANs. As a result, certificates issued by public CAs for use on the open Internet are unlikely to be exploitable, though older certificates still within their validity period could theoretically be abused.

The primary risks exist in private CA environments and client certificate authentication, where identity mapping may rely solely on the CN. Both of these are niche deployment scenarios.

Patches

The path has been prepared by replacing the use of getSubjectDN().getName() with standards-compliant LdapName parsing of the RFC2253 representation of X500Principal. This avoids unescaped provider-dependent strings and prevents regex from matching malicious substrings.

The fix is included in Openfire 5.0.2 and 5.1.0. Users should upgrade to this version as soon as it becomes available.

Starting in Openfire 5.0.2, Openfire will by default prefer a Subject Alternative Name-based identity over a Common Name based one for client-to-server authentication (issue OF-3123). For server-to-server authentication, this already is the case.

In Openfire 5.1.0, Openfire will no longer, by default, use Common Name based identities, although this functionality can be restored through configuration changes (issue OF-3122).

Workarounds

The vulnerability is fixed in the upcoming release, but there are operational mitigations you can apply today.

Full workaround (drop-in replacement JAR using the fixed mapper)

This is a complete workaround:

  1. Take the fixed mapper implementation (the code that uses LdapName / RFC2253 parsing) from the patched Openfire source.
  2. Create a new Java class in your proprietary namespace (for example com.acme.openfire.SafeCNCertificateIdentityMapping) that implements the same Openfire certificate-identity mapping interface and contains the fixed parsing logic. Ensure the logic exactly mirrors the patched behavior (use X500Principal.getName() with RFC2253 and javax.naming.ldap.LdapName to parse RDNs, do not rely on getSubjectDN().getName() or provider-dependent strings).
  3. Package the class into a JAR file.
  4. Place the JAR into Openfire’s lib/ directory (e.g. $OPENFIRE_HOME/lib/your-fixed-mapper.jar).
  5. Configure Openfire to use your class for mapping by setting the following properties (in the admin console / system properties). Note that it remains advisable to configure Openfire to keep preferring SAN-based identities:
# For server-to-server identity mapping
provider.serverCertIdentityMap.classList=org.jivesoftware.util.cert.SANCertificateIdentityMapping,com.acme.openfire.SafeCNCertificateIdentityMapping

# For client-to-server identity mapping
provider.clientCertIdentityMap.classList=org.jivesoftware.util.cert.SANCertificateIdentityMapping,com.acme.openfire.SafeCNCertificateIdentityMapping

Both properties accept a comma-separated list of fully-qualified class names; listing only your safe mapper forces Openfire to use your implementation instead of the vulnerable built-in mapper. After placing the JAR and updating properties, restart Openfire.

Notes & caveats for the full workaround

  • This is effectively a complete mitigation for the vulnerable CN-extraction behavior without deploying a new official release.
  • Maintain the JAR: you are responsible for keeping the custom class updated if you upgrade Openfire to newer major versions.
  • Ensure your custom mapper truly implements the patched parsing (do not copy the vulnerable getSubjectDN().getName() + regex approach). Use X500PrincipalgetName(RFC2253)new LdapName(...) → iterate RDNs in a robust way.
  • Because this runs in the Openfire JVM, follow your organization’s binary-signing and review policies before deploying.

Other, lesser workarounds

Use SAN-only mapping

Configure the mapper list to only include Openfire’s SAN mapper: org.jivesoftware.util.cert.SANCertificateIdentityMapping. This prevents CN parsing-based spoofing but breaks authentication for certificates that contain only a CN and no SAN.

Disable certificate-based authentication:

  • For server-to-server, disabling certificate auth leaves only Server Dialback, which is less secure; depending on your environment, this may be a worse trade-off.
  • For client-to-server, disable “mutual authentication” in the admin console to stop client cert authentication altogether.

References

References

@guusdk guusdk published to igniterealtime/Openfire Sep 15, 2025
Published by the National Vulnerability Database Sep 15, 2025
Published to the GitHub Advisory Database Sep 16, 2025
Reviewed Sep 16, 2025
Last updated Sep 16, 2025

Severity

Moderate

CVSS overall score

This score calculates overall vulnerability severity from 0 to 10 and is based on the Common Vulnerability Scoring System (CVSS).
/ 10

CVSS v3 base metrics

Attack vector
Network
Attack complexity
High
Privileges required
High
User interaction
None
Scope
Unchanged
Confidentiality
High
Integrity
High
Availability
None

CVSS v3 base metrics

Attack vector: More severe the more the remote (logically and physically) an attacker can be in order to exploit the vulnerability.
Attack complexity: More severe for the least complex attacks.
Privileges required: More severe if no privileges are required.
User interaction: More severe when no user interaction is required.
Scope: More severe when a scope change occurs, e.g. one vulnerable component impacts resources in components beyond its security scope.
Confidentiality: More severe when loss of data confidentiality is highest, measuring the level of data access available to an unauthorized user.
Integrity: More severe when loss of data integrity is the highest, measuring the consequence of data modification possible by an unauthorized user.
Availability: More severe when the loss of impacted component availability is highest.
CVSS:3.1/AV:N/AC:H/PR:H/UI:N/S:U/C:H/I:H/A:N

EPSS score

Exploit Prediction Scoring System (EPSS)

This score estimates the probability of this vulnerability being exploited within the next 30 days. Data provided by FIRST.
(5th percentile)

Weaknesses

Authentication Bypass by Spoofing

This attack-focused weakness is caused by incorrectly implemented authentication schemes that are subject to spoofing attacks. Learn more on MITRE.

CVE ID

CVE-2025-59154

GHSA ID

GHSA-w252-645g-87mp

Credits

Loading Checking history
See something to contribute? Suggest improvements for this vulnerability.