''' We've discovered a privilege escalation issue in the OpenShift platform.

Conditions for the privilege escalation: - a user must be granted the ability to "update, patch" the "pods/ephemeralcontainers" subresource - by default, NEITHER common users NOR Service Accounts are granted this permission by the platform - with the above permission, a user is able to patch a running pod they've got access to and bypass SCC admission - this means, a user can create a "privileged" container, which allows them obtaining access to the pod's node resources

The step-by-step reproducer is described in https://issues.redhat.com/browse/OCPBUGS-7181.

Affected OpenShift Container Platform versions: 4.10 and newer. Affected component: kube-apiserver (and the platforms that use it)

The bug is located within the apiserver-library-go repository in the following module: https://github.com/openshift/apiserver-library-go/tree/master/pkg/securitycontextconstraints. We are yet to determine how to fix it.

The workaround is to remove the permissions to "update, patch" the "pods/ephemeralcontainers" subresource from any low-privileged users, if there are any that currently hold it. '''

From Stanislav Láznička

A flaw in the containers/common Go library, used by container runtimes such as Podman, Buildah, and CRI-O, allows a malicious container image to exploit the FIPS mode to bind mount arbitrary host directories into the container. The flaw arises from the MountsWithUIDGID() function, which attempts to mount a FIPS-related directory inside the container but does not properly verify the path. If the FIPS path is a symlink, it could point to arbitrary directories on the host, allowing an attacker to access sensitive host files.

Impact A malicious container can affect the host by taking advantage of code cri-o added to show the container mounts on the host.

A workload built from this Dockerfile: FROM docker.io/library/busybox as source RUN mkdir /extra && cd /extra && ln -s ../../../../../../../../root etc

FROM scratch

COPY --from=source /bin /bin COPY --from=source /lib /lib COPY --from=source /extra .

and this container config:

{ "metadata": { "name": "busybox" }, "image":{ "image": "localhost/test" }, "command": [ "/bin/true" ], "linux": { } }

and this sandbox config { "metadata": { "name": "test-sandbox", "namespace": "default", "attempt": 1, "uid": "edishd83djaideaduwk28bcsb" }, "linux": { "securitycontext": { "namespaceoptions": { "network": 2 } } } }

will create a file on host /host/mtab

Patches 1.30.1, 1.29.5, 1.28.7

Workarounds Unfortunately not

References Are there any links users can visit to find out more?

A vulnerability was found in Buildah. Cache mounts do not properly validate that user-specified paths for the cache are within our cache directory, allowing a RUN instruction in a Container file to mount an arbitrary directory from the host (read/write) into the container as long as those files can be accessed by the user running Buildah.

crypto: algifaead - Revert to operating out-of-place

Impact What kind of vulnerability is it? Who is impacted? All versions of CRI-O running on cgroupv2 nodes. Unchecked access to an experimental annotation allows a container to be unconfined. Back in 2021, support was added to support an experimental annotation that allows a user to request special resources in cgroupv2. It was supposed to be gated by an experimental annotation: io.kubernetes.cri-o.UnifiedCgroup, which was supposed to be filtered from the list of allowed annotations . However, there is a bug in this code which allows any user to specify this annotation, regardless of whether it's enabled on the node. The consequences of this are a pod can specify any amount of memory/cpu and get it, circumventing the kubernetes scheduler, and potentially be able to DOS a node. Patches Has the problem been patched? What versions should users upgrade to? 1.29.1, 1.28.3, 1.27.3

Workarounds Is there a way for users to fix or remediate the vulnerability without upgrading? use cgroupv1

References Are there any links users can visit to find out more?

A flaw was found in rsync which could be triggered when rsync compares file checksums. This flaw allows an attacker to manipulate the checksum length (s2length) to cause a comparison between a checksum and uninitialized memory and leak one byte of uninitialized stack data at a time.

A flaw was found in libxml2's xmlBuildQName function, where integer overflows in buffer size calculations can lead to a stack-based buffer overflow. This issue can result in memory corruption or a denial of service when processing crafted input.

A symlink traversal vulnerability in the containers/storage library can cause Podman, Buildah, and CRI-O to hang and potentially be DoSed via OOM kill when running a malicious image using an automatically assigned user namespace (--userns=auto in Podman and Buildah). The containers/storage library will read /etc/passwd inside the container, but does not properly validate if that file is a symlink, which can be used to cause the library to read an arbitrary file on the host. This file is only read, and if it does not properly parse as a copy of /etc/passwd it will cause an error (there is a small risk of information disclosure via the error message here as elements of the file that failed to parse can be included, but this is only as the user running Podman/Buildah/CRI-O so it wouldn't be a file they did not already have access to). The report here discovered that you can symlink /etc/passwd in the container to a FIFO on the host, causing a hang as the file cannot be completely read (or an OOM condition if the FIFO is continuously written to, which was then ready by Podman). This hang could occur in a critical section in the c/storage library, blocking other processes from creating containers, but could be easily solved via a SIGKILL of the affected process. The ability to potentially crash the CRI-O service via OOM kill could be more relevant, though the attacker would have to know the path of a FIFO that is regularly being written to on the host in order to do this.