Kubernetes CVE

Directory traversal vulnerability in Kubernetes, as used in Red Hat OpenShift Enterprise 3.0, allows attackers to write to arbitrary files via a crafted object type name, which is not properly handled before passing it to etcd.

The API server in Kubernetes does not properly check admission control, which allows remote authenticated users to access additional resources via a crafted patched object.

Kubernetes before 1.2.0-alpha.5 allows remote attackers to read arbitrary pod logs via a container name.

The API server in Kubernetes, as used in Red Hat OpenShift Enterprise 3.2, in a multi tenant environment allows remote authenticated users with knowledge of other project names to obtain sensitive project and user information via vectors related to the watch-cache list.

Kubernetes version 1.5.0-1.5.4 is vulnerable to a privilege escalation in the PodSecurityPolicy admission plugin resulting in the ability to make use of any existing PodSecurityPolicy object.

Kubernetes in OpenShift3 allows remote authenticated users to use the private images of other users should they know the name of said image.

Default access permissions for Persistent Volumes (PVs) created by the Kubernetes Azure cloud provider in versions 1.6.0 to 1.6.5 are set to "container" which exposes a URI that can be accessed without authentication on the public internet. Access to the URI string requires privileged access to the Kubernetes cluster or authenticated access to the Azure portal.

In Kubernetes versions 1.3.x, 1.4.x, 1.5.x, 1.6.x and prior to versions 1.7.14, 1.8.9 and 1.9.4 containers using subpath volume mounts with any volume type (including non-privileged pods, subject to file permissions) can access files/directories outside of the volume, including the host's filesystem.

In Kubernetes versions 1.3.x, 1.4.x, 1.5.x, 1.6.x and prior to versions 1.7.14, 1.8.9 and 1.9.4 containers using a secret, configMap, projected or downwardAPI volume can trigger deletion of arbitrary files/directories from the nodes where they are running.

A vulnerability in the container management subsystem of Cisco Digital Network Architecture (DNA) Center could allow an unauthenticated, remote attacker to bypass authentication and gain elevated privileges. This vulnerability is due to an insecure default configuration of the Kubernetes container management subsystem within DNA Center. An attacker who has the ability to access the Kubernetes service port could execute commands with elevated privileges within provisioned containers. A successful exploit could result in a complete compromise of affected containers. This vulnerability affects Cisco DNA Center Software Releases 1.1.3 and prior. Cisco Bug IDs: CSCvi47253.

CoreOS Tectonic 1.7.x before 1.7.9-tectonic.4 and 1.8.x before 1.8.4-tectonic.3 mounts a direct proxy to the kubernetes cluster at /api/kubernetes/ which is accessible without authentication to Tectonic and allows an attacker to directly connect to the kubernetes API server. Unauthenticated users are able to list all Namespaces through the Console, resulting in an information disclosure. Tectonic's exposure of an unauthenticated API endpoint containing information regarding the internal state of the cluster can provide an attacker with information that may assist in other attacks against the cluster. For example, an attacker may not have the permissions required to list all namespaces in the cluster but can instead leverage this vulnerability to enumerate the namespaces and then begin to check each namespace for weak authorization policies that may allow further escalation of privileges.

Kubernetes CRI-O version prior to 1.9 contains a Privilege Context Switching Error (CWE-270) vulnerability in the handling of ambient capabilities that can result in containers running with elevated privileges, allowing users abilities they should not have. This attack appears to be exploitable via container execution. This vulnerability appears to have been fixed in 1.9.

In Kubernetes versions 1.5.x, 1.6.x, 1.7.x, 1.8.x, and prior to version 1.9.6, the kubectl cp command insecurely handles tar data returned from the container, and can be caused to overwrite arbitrary local files.

A exposure of sensitive information vulnerability exists in Jenkins Kubernetes Plugin 1.7.0 and older in ContainerExecDecorator.java that results in sensitive variables such as passwords being written to logs.

The F5 BIG-IP Controller for Kubernetes 1.0.0-1.5.0 (k8s-bigip-crtl) passes BIG-IP username and password as command line parameters, which may lead to disclosure of the credentials used by the container.

An exposure of sensitive information vulnerability exists in Jenkins Kubernetes Plugin 1.10.1 and earlier in KubernetesCloud.java that allows attackers to capture credentials with a known credentials ID stored in Jenkins.

It was found that Kubernetes as used by Openshift Enterprise 3 did not correctly validate X.509 client intermediate certificate host name fields. An attacker could use this flaw to bypass authentication requirements by using a specially crafted X.509 certificate.

Following the Gardener architecture, the Kubernetes apiserver of a Gardener managed shoot cluster resides in the corresponding seed cluster. Due to missing network isolation a shoot's apiserver can access services/endpoints in the private network of its corresponding seed cluster. Combined with other minor Kubernetes security issues, the missing network isolation theoretically can lead to compromise other shoot or seed clusters in the "Gardener" context. The issue is rated high due to the high impact of a potential exploitation in "Gardener" context. This was fixed in Gardener release 0.12.4.

The Kubernetes integration in GitLab Enterprise Edition 11.x before 11.2.8, 11.3.x before 11.3.9, and 11.4.x before 11.4.4 has SSRF.

In Kubernetes versions 1.9.0-1.9.9, 1.10.0-1.10.5, and 1.11.0-1.11.1, user input was handled insecurely while setting up volume mounts on Windows nodes, which could lead to command line argument injection.