CVE-2026-105929: 
Linux Red Hat Analyse et atténuation de la vulnérabilité

Aperçu

CVE-2026-105929 is a path traversal vulnerability (CWE-22) in ansible-core's ansible-galaxy role installation mechanism. A role's explicit name field from meta/requirements.yml is used to construct the on-disk installation path without validating that the destination remains within the configured roles directory — a name containing ../ escapes the directory, and an absolute name replaces it entirely. Affected software includes ansible-core and numerous Red Hat Ansible Automation Platform container images (versions 2.4 through 2.7 and tech-preview streams). The vulnerability was disclosed on November 17–18, 2026, and carries a CVSS v3.1 base score of 8.1 (High) (Red Hat CVE, GitHub Advisory).

Détails techniques

The root cause (CWE-22) lies in three files within ansible-core: lib/ansible/playbook/role/requirement.py (RoleRequirement.role_yaml_parse()) preserves an attacker-supplied role name without sanitization; lib/ansible/galaxy/role.py (GalaxyRole.__init__()) constructs installation paths via os.path.join(roles_path, name), which Python's os.path.join resolves to an arbitrary location when name is absolute or contains ../; and lib/ansible/cli/galaxy.py processes transitive dependencies from an installed role's meta/requirements.yml, meaning the operator need not explicitly choose the malicious destination. The existing tar-member path check does not mitigate this because it constrains archive members relative to the extraction root, while the extraction root itself is attacker-controlled via the role name. When --force is used, GalaxyRole.remove() calls shutil.rmtree(self.path), enabling deletion and replacement of role-shaped directories outside the intended path (Red Hat Bugzilla, Red Hat CVE).

Impact

Successful exploitation allows an unauthenticated malicious role publisher to cause arbitrary files to be written outside the operator's designated roles directory, bounded only by the invoking user's filesystem permissions. When --force is used, existing role-shaped directories outside the intended path can be deleted and replaced with attacker-controlled content, potentially enabling code execution if critical files (e.g., scripts, libraries, or configuration files) are overwritten. Confidentiality impact is rated None, while both Integrity and Availability are rated High, reflecting the risk of file system manipulation and service disruption rather than data exfiltration (Red Hat CVE, GitHub Advisory).

Exploitabilité

There is no public proof-of-concept exploit and no evidence of in-the-wild exploitation as of the disclosure date (Red Hat CVE). The EPSS score is 0.0, indicating very low current exploitation probability. The vulnerability is not listed in the CISA Known Exploited Vulnerabilities catalog. Exploitation requires user interaction — specifically, an operator must install a role from an attacker-controlled source — but no privileges are required on the attacker's side, and the attack vector is network-based (GitHub Advisory).

Étapes d’exploitation

  1. Publish a malicious role: The attacker creates an Ansible role and publishes it to Ansible Galaxy or a reachable role repository. The role's meta/requirements.yml includes a transitive dependency with a crafted name field such as ../../sensitive_dir or an absolute path like /etc/cron.d/backdoor.
  2. Operator installs the role: The victim operator runs ansible-galaxy role install <malicious-role> (optionally with --force), triggering ansible-galaxy to process the role's meta/requirements.yml and resolve transitive dependencies.
  3. Path traversal occurs: RoleRequirement.role_yaml_parse() preserves the attacker-supplied name without validation. GalaxyRole.__init__() constructs the installation path as os.path.join(roles_path, name), resolving to a location outside the roles directory.
  4. Arbitrary file write: Role files are extracted to the attacker-controlled path, writing attacker-supplied content outside the designated roles directory under the operator's filesystem permissions.
  5. Directory deletion (with --force): If --force is used and the target path contains a valid role structure (meta/main.yml), GalaxyRole.remove() calls shutil.rmtree() on the traversed path, deleting the existing directory before replacing it with attacker content (Red Hat Bugzilla).

Indicateurs de compromis

  • File System: Unexpected files or directories created outside the configured Ansible roles directory (e.g., in /etc/, /usr/, home directories, or other sensitive paths); presence of meta/main.yml in unexpected filesystem locations indicating a role was installed there.
  • Logs: ansible-galaxy output or logs showing role installation paths resolving to locations outside the expected roles directory; warnings or errors from os.path.join resolving to absolute paths.
  • Process: ansible-galaxy role install invocations followed by unexpected file creation events in sensitive directories (detectable via auditd or inotify-based monitoring); shutil.rmtree activity on directories outside the roles path when --force is used.
  • Network: Outbound connections from the operator's system to unfamiliar role repositories or Galaxy-compatible endpoints serving roles with suspicious meta/requirements.yml content.

Atténuation et solutions de contournement

Red Hat has confirmed a patch is available; operators should update ansible-core to a patched version as soon as it is released for their platform (Red Hat CVE, GitHub Advisory). As an immediate workaround, Red Hat recommends installing roles and collections only from trusted, verified sources. Before installing any role, manually review its meta/requirements.yml (including all transitive dependencies) for any name value containing ../ or an absolute path. Avoid using ansible-galaxy role install --force against untrusted content, as this enables the additional directory deletion capability (Red Hat CVE).

Réactions de la communauté

The vulnerability was tracked by multiple security intelligence platforms including VulDB, CVEFeed, and national CERTs (e.g., Hungary's NKI and Spain's CCN-CERT) shortly after disclosure, indicating broad community awareness (Red Hat CVE). Red Hat's Product Security team assigned a High severity rating and provided detailed mitigation guidance on their customer portal. No notable independent researcher commentary or significant social media discussion has been identified beyond standard aggregator coverage.

Ressources supplémentaires

État de correction de la distribution Linux

Corrigez la disponibilité entre les grandes distributions Linux et leurs versions.

Ubuntu

Inconnu

bionic (esm-apps)

ansible

Inconnu

devel

ansible

Inconnu

focal (esm-apps)

ansible

Inconnu

jammy

ansible

Inconnu

jammy (esm-apps)

ansible

Inconnu

noble

ansible

Inconnu

noble (esm-apps)

ansible

Inconnu

resolute

ansible

Inconnu

Source: Ce rapport a été généré à l’aide de l’IA

Connexe Linux Red Hat Vulnérabilités :

CVE ID

Gravité

Partition

Technologies

Nom du composant

Exploit CISA KEV

A corrigé

Date de publication

CVE-2026-130644HIGH8.1
  • Linux Red Hat logoLinux Red Hat
  • ansible-core
NonNonNov 18, 2026
CVE-2026-105929HIGH8.1
  • Linux Red Hat logoLinux Red Hat
  • ansible-core
NonNonNov 18, 2026
CVE-2026-114938HIGH8.1
  • Amazon Corretto JDK logoAmazon Corretto JDK
  • java-21-openjdk-static-libs-fastdebug
NonOuiNov 17, 2026
CVE-2026-119192HIGH7.5
  • Chainguard logoChainguard
  • argo-workflows-3.6
NonOuiNov 17, 2026
CVE-2026-6949HIGH7.5
  • Samba logoSamba
  • ctdb-tests
NonOuiNov 16, 2026

Évaluation gratuite des vulnérabilités

Évaluez votre posture de sécurité cloud

Évaluez vos pratiques de sécurité cloud dans 9 domaines de sécurité afin d’évaluer votre niveau de risque et d’identifier les lacunes dans vos défenses.

Évaluation de la demande

Ressources Wiz supplémentaires

Obtenez une démo personnalisée

Prêt(e) à voir Wiz en action ?

"La meilleure expérience utilisateur que j’ai jamais vue, offre une visibilité totale sur les workloads cloud."
David EstlickRSSI
"Wiz fournit une interface unique pour voir ce qui se passe dans nos environnements cloud."
Adam FletcherChef du service de sécurité
"Nous savons que si Wiz identifie quelque chose comme critique, c’est qu’il l’est réellement."
Greg PoniatowskiResponsable de la gestion des menaces et des vulnérabilités