
PEACH
Un cadre d’isolation des locataires
CVE-2026-99239 is a Time-of-Check Time-of-Use (TOCTOU) race condition in the Linux kernel's Bluetooth L2CAP implementation that can allow a local attacker to trigger memory corruption or kernel crash, potentially leading to local code execution. The vulnerability arises because an l2cap_chan can be incorrectly linked to an l2cap_conn more than once due to insufficient state checking and locking between l2cap_chan_connect() and l2cap_sock_bind(). It was published on November 2, 2026, and affects multiple stable Linux kernel branches. Affected versions include 5.10.265–5.10.272, 5.15.216–5.15.223, 6.1.183–6.1.190, 6.6.145–6.6.160, 6.12.97–6.12.114, 6.18.39–6.18.59, 7.1.4, and 7.2. It carries a CVSS v3.1 base score of 7.8 (High) (GitHub Advisory, Red Hat).
The root cause is a TOCTOU race condition (CWE-367) in the Linux kernel Bluetooth L2CAP subsystem. The invariant that an l2cap_chan may be linked to an l2cap_conn at most once is violated when two tasks concurrently execute l2cap_chan_connect() (Task 1) and l2cap_sock_bind() (Task 2): Task 1 sets chan->state to BT_CONNECT, while Task 2 sets it to BT_BOUND before Task 1 acquires the socket lock, causing l2cap_sock_connect() — which does not check sk->sk_state — to call __l2cap_chan_add() a second time. The fix adds a WARN_ON_ONCE invariant check in __l2cap_chan_add(), a pre-check in l2cap_chan_connect() to avoid relying on chan->state, and introduces l2cap_lock_chan in l2cap_sock_bind() to properly guard chan->state writes (GitHub Advisory, Red Hat Bugzilla).
Successful exploitation allows a local user with the ability to create Bluetooth sockets to trigger memory corruption or undefined behavior in the kernel's L2CAP cleanup logic, potentially resulting in a kernel crash (denial of service) or arbitrary code execution at the kernel level. The confidentiality, integrity, and availability impacts are all rated High, meaning a successful exploit could grant full control over the affected system. Lateral movement is possible if the attacker leverages kernel-level access to escalate privileges or pivot within the environment (GitHub Advisory).
There is no public proof-of-concept exploit and no evidence of in-the-wild exploitation as of the disclosure date (GitHub Advisory). The EPSS score is 0.0, indicating very low probability of exploitation in the near term. The vulnerability is not listed in the CISA Known Exploited Vulnerabilities (KEV) catalog. Exploitation requires local access and the ability to create Bluetooth sockets, which limits the attack surface compared to remote vulnerabilities.
CAP_NET_RAW or membership in the bluetooth group).uname -r.connect() (triggering l2cap_chan_connect()) in one thread and bind() (triggering l2cap_sock_bind()) in another thread, racing to violate the l2cap_chan/l2cap_conn linkage invariant.chan->state = BT_CONNECT while Task 2 sets chan->state = BT_BOUND, causing l2cap_sock_connect() to call __l2cap_chan_add() a second time on the same channel.dmesg or /var/log/kern.log) showing WARN_ON_ONCE triggered in __l2cap_chan_add, or kernel oops/panic traces referencing l2cap_chan_del, l2cap_chan_connect, or l2cap_sock_bind.AF_BLUETOOTH, BTPROTO_L2CAP), especially from non-privileged users or unexpected binaries.ss -x or Bluetooth monitoring tools./tmp that interact with Bluetooth socket APIs.Upgrade to a patched Linux kernel version: 5.10.273 or later, 5.15.224 or later, 6.1.191 or later, 6.6.161 or later, 6.12.115 or later, 6.18.60 or later, 7.2.14 or later, or 7.3.1 or later (GitHub Advisory, Red Hat). As a workaround where Bluetooth is not required, restrict or disable Bluetooth socket creation by removing the bluetooth kernel module (modprobe -r bluetooth) or using AppArmor/SELinux policies to deny CAP_NET_RAW to untrusted users. Distributions such as Red Hat are tracking this via their security advisories and should be consulted for distribution-specific guidance (Red Hat Bugzilla).
Corrigez la disponibilité entre les grandes distributions Linux et leurs versions.
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 afin d’évaluer votre niveau de risque et d’identifier les lacunes 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."