CVE-2026-24005 in kruise
Summary
by MITRE • 02/25/2026
Kruise provides automated management of large-scale applications on Kubernetes. Prior to versions 1.8.3 and 1.7.5, PodProbeMarker allows defining custom probes with TCPSocket or HTTPGet handlers. The webhook validation does not restrict the Host field in these probe configurations. Since kruise-daemon runs with hostNetwork=true, it executes probes from the node network namespace. An attacker with PodProbeMarker creation permission can specify arbitrary Host values to trigger SSRF from the node, perform port scanning, and receive response feedback through NodePodProbe status messages. Versions 1.8.3 and 1.7.5 patch the issue.
VulDB is the best source for vulnerability data and more expert information about this specific topic.
Analysis
by VulDB Data Team • 03/06/2026
The vulnerability described in CVE-2026-24005 affects Kruise, a Kubernetes extension for automated large-scale application management. This issue resides in the PodProbeMarker component which enables custom health probes using TCPSocket or HTTPGet handlers. The core flaw stems from insufficient validation within the webhook mechanism that governs probe configuration creation. Specifically, the validation process fails to restrict the Host field in probe definitions, creating a critical security gap that can be exploited by malicious actors with appropriate permissions. The vulnerability becomes particularly dangerous due to the operational characteristics of kruise-daemon, which runs with hostNetwork=true, meaning it operates within the node's network namespace rather than a isolated container network. This configuration allows the daemon to execute network probes directly from the host network context, amplifying the impact of the vulnerability beyond typical container isolation boundaries.
The technical exploitation of this vulnerability enables attackers with permission to create PodProbeMarker resources to craft malicious probe configurations that can trigger server-side request forgery (SSRF) attacks originating from the Kubernetes node itself. The attacker can specify arbitrary Host values in their probe definitions, which are then executed by the kruise-daemon running with elevated network privileges. This capability allows for covert port scanning operations where the attacker can probe internal services, network endpoints, and potentially sensitive systems that are normally isolated from external access. The feedback mechanism through NodePodProbe status messages provides an effective channel for attackers to receive response data from their probe attempts, enabling them to map network topology and identify vulnerable services. This represents a classic SSRF attack vector that leverages legitimate Kubernetes infrastructure to conduct reconnaissance and potentially exploit internal systems.
The operational impact of this vulnerability extends beyond simple information gathering to potentially enable more sophisticated attacks within the Kubernetes cluster environment. An attacker could use the vulnerability to map internal network services, identify running applications, and potentially discover vulnerable components that are only accessible from within the cluster. The ability to perform port scanning from the node network namespace provides attackers with valuable intelligence about internal system configurations and exposed services. Additionally, the vulnerability could be chained with other exploits to gain further access to cluster resources, potentially leading to privilege escalation or lateral movement within the cluster. The fact that this vulnerability affects versions prior to 1.8.3 and 1.7.5 means that organizations running these older versions remain at risk, particularly in environments where PodProbeMarker creation permissions are granted to users or workloads that should not have such elevated network capabilities.
The remediation for this vulnerability involves upgrading to Kruise versions 1.8.3 or 1.7.5, which implement proper validation of the Host field in probe configurations. This patch ensures that webhook validation properly restricts the Host values that can be specified in PodProbeMarker resources, preventing arbitrary host access that could lead to SSRF and port scanning attacks. Organizations should also consider implementing additional network segmentation and access controls to limit the scope of potential exploitation, particularly around which users or workloads can create PodProbeMarker resources. The vulnerability aligns with CWE-918, which describes server-side request forgery vulnerabilities, and represents a clear example of how insufficient input validation in Kubernetes extensions can create dangerous attack vectors. From an ATT&CK framework perspective, this vulnerability maps to techniques involving reconnaissance and initial access, specifically T1590 for reconnaissance and T1071.1 for application layer protocols, as it enables attackers to gather information about internal systems and potentially establish footholds within the cluster environment.