
PEACH
Un cadre d’isolation des locataires
CVE-2026-18355 is a heap buffer overflow vulnerability in the SASL I/O layer of 389 Directory Server (389-ds-base), caused by an integer underflow in sasl_io_read_packet(). The flaw affects Red Hat's 389-ds-base across multiple product streams including redhat-ds:11, redhat-ds:12, and 389-ds:1.4 on Red Hat Enterprise Linux. It was reported on July 30, 2026 and publicly disclosed on September 7, 2026. It carries a CVSS v3.1 base score of 7.5 (High) (Red Hat CVE, GitHub Advisory).
The root cause is an integer underflow (CWE-191) in the SASL I/O layer. In sasl_io_start_packet(), the 4-byte wrapped-record length read from the wire is validated only against an upper bound (a check introduced for CVE-2026-11774), but not against a lower bound. When an attacker sends a SASL-wrapped record with a wire length of 0, 1, or 2, the computed encrypted_buffer_count (e.g., 4, 5, or 6 after adding sizeof(uint32_t)) falls below the already-consumed encrypted_buffer_offset (7 header bytes). The resulting unsigned subtraction in sasl_io_read_packet() underflows to approximately 0xFFFFFFFD, causing PR_Recv to attempt reading ~4 GiB into a 1024-byte heap buffer — a heap overflow with attacker-controlled content. Exploitation requires a prior successful SASL bind with integrity protection (SSF > 0), making this a post-authentication network attack (Red Hat CVE, Red Hat Bugzilla).
Successful exploitation can result in denial of service (server crash) or potentially remote code execution on the affected 389 Directory Server instance. Because 389-ds-base serves as an LDAP directory backend, compromise could expose sensitive directory data including user credentials, group memberships, and organizational information, and could enable lateral movement within environments relying on LDAP for authentication. All three CIA pillars — confidentiality, integrity, and availability — are rated High in the CVSS scoring (Red Hat CVE).
No public proof-of-concept exploit code or in-the-wild exploitation has been reported as of the disclosure date. The EPSS score is 0.0, reflecting low current exploitation probability. The vulnerability is not listed in the CISA Known Exploited Vulnerabilities catalog. Exploitation requires a valid SASL-authenticated session with integrity protection (SSF > 0), which limits the attacker pool to authenticated users or compromised accounts. The vulnerability was discovered and reported by Adam Korczynski, Arthur Chan, David Korczynski (Ada Logics), and a team from Anthropic (Red Hat CVE, GitHub Advisory).
sasl_io_start_packet(), the small length value causes encrypted_buffer_count to be less than encrypted_buffer_offset, producing an unsigned integer underflow (~0xFFFFFFFD) in sasl_io_read_packet().PR_Recv, which attempts to read ~4 GiB of attacker-controlled data into the 1024-byte encrypted_buffer heap allocation, overflowing adjacent heap memory.ns-slapd process; core dump files generated in the 389-ds working directory following LDAP session activity.core.*) in /var/log/dirsrv/ or the server's working directory indicating a crash event.Red Hat rates this issue as Important and recommends updating the affected 389-ds-base package as soon as a fix becomes available. No practical configuration-based mitigation has been identified by Red Hat. As a partial risk reduction, administrators can restrict SASL bind access to trusted, known-good accounts and limit network exposure of the LDAP service. Deployments that do not negotiate a SASL security layer (SSF = 0) after bind are not exposed to this specific code path (Red Hat CVE).
The vulnerability was credited to researchers from Ada Logics (Adam Korczynski, Arthur Chan, David Korczynski) and a team from Anthropic, suggesting it may have been discovered through fuzzing or AI-assisted security research. Red Hat has classified the issue as Important severity. No significant broader media coverage or notable community commentary beyond standard CVE tracking sites has been observed as of the disclosure date (Red Hat CVE).
Source: Ce rapport a été généré à l’aide de l’IA
Évaluation gratuite des vulnérabilités
Évaluez vos pratiques de sécurité cloud dans 9 domaines de sécurité pour évaluer votre niveau de risque et identifier les failles dans vos défenses.
Obtenez une démo personnalisée
"La meilleure expérience utilisateur que j’ai jamais vue, offre une visibilité totale sur les workloads cloud."
"Wiz fournit une interface unique pour voir ce qui se passe dans nos environnements cloud."
"Nous savons que si Wiz identifie quelque chose comme critique, c’est qu’il l’est réellement."