CVE-2025-11395 Overview
A flaw was found in Podman, the daemonless container engine widely used on Linux systems. The vulnerability allows an attacker who can supply a crafted tar archive to the podman load command to create files on the host filesystem. Any files written inherit the privileges of the user running Podman, which can include root when Podman runs in rootful mode. The issue is tracked under [CWE-277: Insecure Inherited Permissions] and stems from insufficient validation of archive contents during image loading.
Critical Impact
An attacker able to deliver a malicious tar archive to podman load can write arbitrary files on the host with the privileges of the invoking user.
Affected Products
- Podman (Red Hat advisory)
- Red Hat distributions shipping affected Podman versions
- Linux systems using podman load to import container images from untrusted archives
Discovery Timeline
- 2026-09-15 - CVE-2025-11395 published to the National Vulnerability Database
- 2026-09-16 - Last updated in NVD database
Technical Details for CVE-2025-11395
Vulnerability Analysis
The vulnerability resides in Podman's image import path. The podman load command accepts a tar archive representing a container image and unpacks it into local storage. When processing a crafted archive, Podman writes files to locations that carry the permissions of the invoking user rather than restricting output to the intended image storage directory. This falls under [CWE-277], where inherited permissions are broader than the operation requires.
The attack vector is Adjacent Network, meaning the attacker needs a logically adjacent path to deliver the archive to the target system. Low privileges and no user interaction are required beyond convincing the user or workflow to invoke podman load against the malicious tar file. Confidentiality, integrity, and availability are all affected at a low level.
Exploitation is not currently known to be occurring in the wild, and the flaw is not listed in the CISA Known Exploited Vulnerabilities catalog. Public exploit code has not been observed at time of publication.
Root Cause
The root cause is improper permission enforcement during tar extraction inside podman load. Podman does not sufficiently constrain the file paths or ownership metadata contained in the archive, allowing entries in the tar to be materialized on the host filesystem outside the expected image storage layout.
Attack Vector
An attacker crafts a malicious tar archive containing entries whose names or metadata cause Podman to write files at attacker-chosen locations on the host. The victim runs podman load -i malicious.tar or an equivalent command. Podman processes the archive and creates the attacker-specified files with the privileges of the running user. If Podman is executed as root, this results in host-level file creation that can be leveraged for privilege escalation or persistence.
Refer to the Red Hat CVE-2025-11395 Advisory and Red Hat Bug Report #2402034 for vendor technical detail.
Detection Methods for CVE-2025-11395
Indicators of Compromise
- Invocations of podman load referencing tar archives sourced from untrusted origins, email attachments, or public download URLs.
- Unexpected files appearing outside the container storage directories (for example, /var/lib/containers/storage/) shortly after a podman load operation.
- Modifications to sensitive host paths such as /etc/, /root/, or systemd unit directories immediately following container image import.
Detection Strategies
- Audit shell history and process telemetry for podman load and podman image load executions, correlating with the source of the archive.
- Monitor filesystem events generated by the Podman process for writes outside the expected storage root.
- Alert when Podman is executed as root against archives originating from user download directories or shared filesystems.
Monitoring Recommendations
- Enable Linux audit rules on execve for the podman binary to capture command-line arguments and the invoking user.
- Ingest Podman event logs (podman events) into a central log store for correlation with file integrity monitoring.
- Baseline expected paths written during legitimate podman load operations and alert on deviations.
How to Mitigate CVE-2025-11395
Immediate Actions Required
- Refrain from running podman load against tar archives obtained from untrusted or unverified sources.
- Where feasible, run Podman in rootless mode so that any exploitation is confined to the unprivileged user account.
- Apply vendor patches as soon as they are available from your Linux distribution.
Patch Information
Patch availability and fixed package versions should be tracked through the Red Hat CVE-2025-11395 Advisory and the corresponding Red Hat Bug Report #2402034. Downstream distributions will publish updated Podman packages that validate tar archive contents and constrain file writes to the image storage directory. Administrators should update Podman on all affected hosts and restart any long-running services that embed Podman.
Workarounds
- Import container images using podman pull from trusted registries instead of podman load where possible.
- Verify the provenance and integrity of any tar archive before loading it, using cryptographic signatures or checksums provided by the image publisher.
- Restrict which users may invoke Podman in rootful mode via sudo policy or role-based access controls.
- Isolate image-import workflows to dedicated, non-production hosts until patched packages are deployed.
Disclaimer: This content was generated using AI. While we strive for accuracy, please verify critical information with official sources.
