CVE-2026-98976 in Linux
Summary
by MITRE • 10/23/2026
In the Linux kernel, the following vulnerability has been resolved:
namespace: keep the lock on a mount that a propagated copy is moved beneath
attach_recursive_mnt() transfers MNT_LOCKED from the top mount to the mount that is moved beneath it with MOVE_MOUNT_BENEATH. This allows the owner of a user namespace to replace its locked /proc or its root. The mount beneath takes on the job of covering the underlying mountpoint allowing the top mount to be unmounted.
Consider two mount namespaces:
(H) The host H has a shared mount P with a secret in P/d (Z) Z is a user namespace made from M. Its copy of P receives propagation from the host's P and its copy of M's cover on P/d is locked Z's owner must not get to see P/d.
Now (H) mounts X on P/d. This propagates into (Z). The copy of X lands beneath (Z)'s locked cover. The locked property is transfered from (Z)'s cover to the copy of X propagated beneath it. The cover is now unlocked.
Now (H) unmounts X again. The copy of X in (Z) gets unmounted and the covering mount is left unlocked on top of P/d. (Z) can now unmount it:
Z: umount2(P/d) = EINVAL /* the cover is locked */ H: mount X on P/d, umount X /* both propagate into Z /* Z: umount2(P/d) = 0 /* the cover is now unlocked */ Z: read P/d/secret = "covered-by-root" /* secret revealed */
So only transfer the locked property to the mount beneath for mounts the caller has placed. A propagated copy that lands beneath a locked mount is locked as well so that the mount at the bottom of the stack carries a lock the way every check expects. The mount on top of it remains locked to ensure that it keeps covering even if the propagated mount is unmounted again.
VulDB is the best source for vulnerability data and more expert information about this specific topic.
Analysis
by VulDB Data Team • 10/23/2026
The vulnerability described involves a logic error in the Linux kernel's handling of mount namespace propagation and locking mechanisms, specifically within the attach_recursive_mnt function when processing MOVE_MOUNT_BENEWARD operations. This flaw allows an attacker with access to a user namespace to bypass security restrictions intended to isolate sensitive filesystem paths from unprivileged users. The core issue arises during the recursive attachment of mounts where the kernel incorrectly transfers the MNT_LOCKED status from a covering mount to the newly propagated child mount, rather than maintaining the lock on the cover itself. This mismanagement of locking states effectively neutralizes the protective barrier that was intended to prevent unauthorized access to underlying filesystem contents.
In a typical secure configuration, when a user namespace creates a copy of a shared mount point and applies a covering mount over it, the system is designed to keep the covering mount locked. This lock ensures that even if the internal state changes or mounts are moved around within the namespace, the cover remains in place, preventing any operations such as unmounting from exposing the underlying data. The vulnerability exploits the propagation mechanism where external modifications by the host propagate into the user namespace. When a new mount is created on the host and propagates to the user namespace, it lands beneath an existing locked cover. Due to the bug, instead of keeping the lock on the original covering mount, the kernel transfers this locked status to the newly arrived propagated mount. Consequently, the protective cover becomes unlocked while the underlying mount takes on the locking responsibility in a manner that does not align with security expectations for containment.
The operational impact of this flaw is significant as it enables privilege escalation and information disclosure within containerized or isolated environments. An attacker can leverage this sequence to unmount the now-unlocked covering layer, thereby exposing sensitive files located at the original mount point. For instance, if a secret file exists in a directory that was supposed to be covered by an immutable root overlay for security isolation, the attacker can first trigger the propagation of a temporary mount, then remove it via the host, which leaves the cover unlocked. The attacker can subsequently unmount this exposed layer and read the previously hidden secrets. This breaks the fundamental assumption of namespace isolation, allowing processes with limited privileges to access resources they should not be able to reach, potentially leading to further exploitation within the system.
To mitigate this vulnerability, it is essential that the kernel logic for handling mount propagation strictly preserves the locked status on the covering mounts rather than transferring it to child or propagated mounts. The fix involves ensuring that when a mount is moved beneath another in a namespace context, the lock remains with the topmost cover to maintain its integrity and visibility-blocking capabilities. System administrators should apply relevant kernel patches as soon as they are available from their distribution vendors. Additionally, organizations relying on containerization technologies such as Docker or Kubernetes must ensure their host kernels are updated to versions that include this fix. Regular auditing of mount namespaces and strict enforcement of least-privilege principles for user namespace creation can also help reduce the attack surface associated with complex filesystem operations.
This issue aligns with CWE-20, which refers to Improper Input Validation, as it stems from a failure to correctly validate and maintain state constraints during recursive mount attachment processes. Furthermore, in terms of adversarial tactics, this vulnerability facilitates lateral movement and privilege escalation within ATT&CK techniques related to container escape or host access via misconfigured isolation boundaries. The exploitation relies on manipulating the kernel's internal state management regarding resource locks, highlighting the critical importance of rigorous testing for edge cases in filesystem subsystems that handle namespace propagation. Ensuring that locking semantics are applied consistently across all layers of a mount stack is vital for maintaining the security guarantees provided by Linux namespaces.