CVE-2026-99538 in Linuxinfo

Summary

by MITRE • 11/19/2026

In the Linux kernel, the following vulnerability has been resolved:

ksmbd: ignore network devices of other namespaces

With no interface list configured, ksmbd binds every interface that comes up. ksmbd_netdev_event() does not look at the device's namespace, and create_socket() opens the listener in the namespace of the task raising NETDEV_UP. A container with its own address on a bridge brings up its interface and finds port 445 in its namespace already taken by ksmbd, so an SMB server running inside the container fails with "Address already in use".

Interfaces are also matched by name only, so a container's eth0 going down would stop the listener on the host's eth0.

Only handle netdev events for devices in init_net.

If you want to get best quality of vulnerability data, you may have to visit VulDB.

Analysis

by VulDB Data Team • 11/20/2026

The vulnerability identified within the Linux kernel involves the ksmbd SMB server implementation failing to properly isolate network device handling across different network namespaces. This flaw stems from a lack of namespace awareness during interface binding and event processing, which can lead to service disruption and potential security implications in containerized environments. The core issue lies in how ksmbd manages its listener sockets when dealing with dynamic network changes, specifically the arrival or removal of network interfaces.

Under default configuration conditions where no specific interface list is explicitly configured for ksmbd, the daemon operates by binding to every available network interface that comes up on the system. This behavior is driven by the ksmbd_netdev_event function, which reacts to NETDEV_UP events without verifying whether the device belongs to the initial network namespace or a different one created by containers or virtualization layers. Consequently, when create_socket attempts to open a listener socket for port 445, it does so within the network namespace of the task that triggered the event rather than restricting itself to the host's primary namespace.

This architectural oversight creates significant operational conflicts in multi-tenant environments such as those utilizing Docker or Kubernetes. When a container provisions its own bridge interface with an IP address and brings up its eth0 device, ksmbd interprets this NETDEV_UP signal and attempts to bind port 445 within that specific container namespace. Since the host's instance of ksmbd has already bound to port 445 on the underlying physical or virtual interfaces visible in other namespaces, the attempt by the container-local process fails with an Address Already In Use error. This results in the SMB server running inside the container failing to start, effectively causing a denial of service for that specific instance and disrupting any services relying on it.

Furthermore, the vulnerability extends beyond just binding issues due to name-based interface matching rather than unique identifier or namespace-aware matching. If an interface named eth0 goes down within a container's network namespace, ksmbd may incorrectly interpret this as a signal to stop listening on the host's physical eth0 interface. This cross-namespace interference means that actions taken by unprivileged containers can inadvertently shut down critical server services running at the host level, leading to unexpected service outages and potential data accessibility issues for legitimate users of the SMB share.

From a security perspective, this flaw highlights risks associated with improper resource isolation between namespaces. While primarily causing availability issues through binding conflicts and unintended socket closure, it also represents an attack surface where malicious containers could potentially influence host-level network services by manipulating interface states or names. This aligns with CWE-787 Out-of-bounds Write concepts in a broader sense of accessing resources outside intended boundaries, though more accurately categorized under CWE-20 Improper Input Validation regarding the lack of namespace verification and CWE-400 Uncontrolled Resource Consumption if multiple containers trigger rapid interface events. In terms of MITRE ATT&CK, this relates to T1562 Impair Defenses through Disable or Modify Tools, as a compromised container could disrupt host-level security monitoring services that rely on network connectivity provided by ksmbd or similar daemons.

The resolution implemented in the kernel update restricts netdev event handling exclusively to devices located within init_net, which is the initial and primary network namespace of the system. By filtering out events from other namespaces, ksmbd ensures it only binds to interfaces that are part of the host's native networking stack. This prevents containers with their own virtual networks from interfering with host-level socket bindings or triggering unintended shutdowns of listener sockets based on interface name collisions across different contexts.

To mitigate similar risks in environments where containerization is prevalent, administrators should ensure they are running patched versions of the Linux kernel that include this fix. Additionally, configuring ksmbd to explicitly bind only to specific trusted interfaces rather than relying on auto-binding behavior can provide an additional layer of defense against namespace-related conflicts. Regular auditing of network configurations and monitoring for Address Already In Use errors in container logs can help detect early signs of such interference before it leads to service disruption.

Responsible

Linux

Reservation

09/25/2026

Disclosure

11/19/2026

Moderation

accepted

CPE

ready

EPSS

0.00168

KEV

no

Activities

very low

Sources

Interested in the pricing of exploits?

See the underground prices here!