A new high-severity local privilege escalation (LPE) vulnerability named Fragnesia (CVE-2026-46300) has been discovered in the Linux Kernel. The flaw, which affects the XFRM ESP-in-TCP subsystem, allows a local, unprivileged user to gain full root privileges. The vulnerability was disclosed on May 13, 2026, with a fully functional proof-of-concept (PoC) exploit released simultaneously on GitHub, creating a zero-day situation for many users. Fragnesia is a bypass for a recently patched, related vulnerability family ('Dirty Frag'), making it particularly dangerous as systems that were just updated may still be vulnerable. The exploit works by corrupting the kernel's page cache, a technique similar to the infamous 'Dirty Pipe' vulnerability, allowing modification of read-only, root-owned files like /usr/bin/su.
CVE-2026-43284, CVE-2026-43500). It exploits a condition where an unprivileged user can manipulate network socket configurations (specifically ESP-in-TCP for IPsec) to cause memory corruption in the kernel. This corruption can be leveraged to write arbitrary data into the page cache for read-only files. The public PoC demonstrates this by overwriting the su binary in the cache, allowing the attacker to create a shell with root permissions.CLONE_NEWUSER), which is the default on many distributions. Distributions that restrict this by default, such as Ubuntu via its AppArmor profile, may be more resilient to the public PoC but are not necessarily immune to the underlying flaw.A working proof-of-concept exploit was released publicly alongside the vulnerability disclosure. This significantly lowers the bar for exploitation and means that attacks are highly likely. System administrators should assume that attackers are actively scanning for and exploiting this vulnerability on any multi-user Linux system.
The business impact of Fragnesia is critical for any organization running multi-user Linux systems, including cloud providers, hosting companies, and enterprises with shared development servers. A successful exploit grants an attacker complete control over the affected system. This could lead to:
The following patterns may help identify vulnerable or compromised systems:
unshare -r or unshare -Uunshare command by an unprivileged user.socket(AF_INET, SOCK_STREAM, IPPROTO_TCP) with TCP_ULP set to espauditd to log the use of the unshare and socket system calls by unprivileged users. Look for anomalous patterns./usr/bin/su.uname -r and compare it against patched versions from your distribution vendor.unshare by unexpected user accounts in your auditd or system logs. Correlate this with network-related system calls or kernel module loading activity.CVE-2026-46300 has been successfully applied.CVE-2026-46300. Monitor your Linux distribution's security advisories and apply the update as soon as it is available.install esp4_tunnel /bin/true and install esp6_tunnel /bin/true to a file in /etc/modprobe.d/. This prevents the vulnerable kernel modules from being loaded.kernel.unprivileged_userns_clone to 0. Be aware that this may break legitimate applications (like container runtimes) that rely on this functionality.New details on 'Fragnesia' (CVE-2026-46300) Linux kernel flaw include attack flow, expanded impact on containers/cloud, and confirmation of emergency patches from major distributions.
Apply the kernel patch provided by the Linux distribution vendor as soon as it becomes available.
Mapped D3FEND Techniques:
As a temporary workaround, disable the vulnerable kernel modules (esp4_tunnel, esp6_tunnel) to prevent the code path from being triggered.
Harden the OS by disabling unprivileged user namespaces (`kernel.unprivileged_userns_clone=0`) to block this and other similar LPE exploits, if application compatibility allows.
Mapped D3FEND Techniques:
The primary and most effective remediation for CVE-2026-46300 is to update the Linux kernel. Administrators must closely monitor security advisories from their specific Linux distribution (e.g., Red Hat, Debian, Canonical) and apply the patched kernel version as soon as it is released. For environments with high uptime requirements where reboots are disruptive, deploying a live patch using services like TuxCare's KernelCare or kpatch is the recommended alternative. This allows the vulnerability to be fixed in the running kernel without a system restart, providing immediate protection against the publicly available exploit. Verifying the new kernel version with uname -r or checking the live patching service status is a critical final step.
As a powerful hardening and mitigation technique, use system call filtering to restrict the actions of unprivileged users. Specifically for Fragnesia and similar LPEs, this involves blocking access to the unshare system call with the CLONE_NEWUSER flag for non-root users. This can be implemented using seccomp-bpf profiles or through broader OS-level configuration by setting the sysctl kernel.unprivileged_userns_clone to 0. While this is a very effective control, it must be tested carefully, as it can break legitimate applications like Docker, Podman, and some sandboxed browser features that rely on user namespaces for isolation. This mitigation is best suited for multi-user servers where user activity can be more strictly controlled.
To detect active exploitation of Fragnesia, use an advanced File Integrity Monitoring (FIM) or Endpoint Detection and Response (EDR) tool capable of performing system file analysis. The Fragnesia PoC works by modifying a read-only, root-owned binary like /usr/bin/su in the kernel's page cache, not on the disk itself. A traditional FIM that only checks on-disk hashes will miss this. A more advanced agent should be configured to monitor for anomalous memory-write operations targeting the page cache of critical setuid files. Furthermore, configure auditd with rules to log all executions of setuid binaries (-a always,exit -F path=/usr/bin/su -F perm=x -F auid>=1000 -F auid!=unset -k su_exec) and look for executions that occur under suspicious circumstances, such as immediately following unshare commands.
The Fragnesia vulnerability (CVE-2026-46300) and a working PoC exploit were publicly disclosed by researcher William Bowling.

Cybersecurity professional with over 10 years of specialized experience in security operations, threat intelligence, incident response, and security automation. Expertise spans SOAR/XSOAR orchestration, threat intelligence platforms, SIEM/UEBA analytics, and building cyber fusion centers. Background includes technical enablement, solution architecture for enterprise and government clients, and implementing security automation workflows across IR, TIP, and SOC use cases.
Help others stay informed about cybersecurity threats
Every tactic, technique, and sub-technique used in this threat has been identified and mapped to the MITRE ATT&CK framework for consistent, actionable threat language.
Observables and indicators of compromise (IOCs) have been extracted and cataloged. Risk has been assessed and correlated with known threat actors and historical campaigns.
Detection rules, incident response steps, and D3FEND-aligned mitigation strategies are included so your team can act on this intelligence immediately.
Structured threat data is packaged as a STIX 2.1 bundle and can be visualized as an interactive graph — relationships between actors, malware, techniques, and indicators.
Sigma detection rules are derived from the threat techniques in this article and can be converted for deployment across any major SIEM or EDR platform.