CVE-2026-12413: Libreswan Pluto IKEv2 Denial of Service Vulnerability Analysis

Technical summary

The vulnerability exists in the reassemble_v2_incoming_fragments() function within Libreswan's pluto daemon. The function processes IKEv2 protocol fragments and stores payload information in a fixed-size array (msg_digest.digest[PAYLIMIT]). The code fails to properly validate the count of stored payloads; an off-by-one error in the assertion PASSERT(logger, md->digest_roof < elemsof(md->digest)) allows the digest_roof counter to equal or exceed the array bounds. When this occurs, the assertion fails and the daemon aborts. An attacker sending an IKEv2 packet with an invalid fragment structure—specifically one that causes the payload count to exceed expectations—triggers this condition. IKEv1 implementations are unaffected. The vulnerability is only exploitable against configurations where IKEv2 fragmentation is enabled (the default state; fragmentation=no must be explicitly set to disable it).

Business impact

Organizations relying on Libreswan for VPN gateway functionality or remote access infrastructure face service availability risk. Repeated exploitation forces the pluto daemon to restart, interrupting VPN tunnels and blocking legitimate IPsec connections. In production environments handling remote workforce access, incident response, or site-to-site connectivity, a successful denial-of-service attack could isolate critical network segments or prevent authorized users from establishing secure connections. The attack requires no authentication and can be mounted from the network interface that accepts IKEv2 traffic, making it accessible to external threat actors if the VPN endpoint is internet-facing. Recovery is automatic (the daemon restarts), but sustained attacks degrade availability.

Affected systems

Libreswan is affected. The vulnerability applies to any deployment running a vulnerable Libreswan version with IKEv2 enabled and fragmentation not explicitly disabled. Administrators who have set fragmentation=no in their configuration are not vulnerable. Common affected scenarios include: (1) Libreswan-based VPN gateways (on-premises or cloud) accepting remote IKEv2 connections; (2) road-warrior setups using Libreswan clients configured to accept fragmented responses; (3) site-to-site IPsec tunnels where the peer sends fragmented IKEv2 packets. Check your Libreswan version against vendor advisories to confirm vulnerability status and patch availability.

Exploitability

The attack is straightforward to execute: an attacker must craft an IKEv2 packet with an invalid fragment structure and send it to the IKEv2 port (typically UDP 500 or 4500). No prior authentication is required, no user interaction is needed, and the attack succeeds over the network with minimal latency requirements. The CVSS score of 7.5 (HIGH) reflects these low barriers to exploitation. However, the attack causes only denial of service, not data breach or system compromise. Weaponization requires only the ability to construct malformed IKEv2 frames, a skill well within the reach of determined threat actors. No known in-the-wild exploitation has been confirmed to date (KEV status: not listed), but the low complexity and immediate impact make proactive patching critical.

Remediation

Patch Libreswan to a version that contains the off-by-one fix in reassemble_v2_incoming_fragments(). Consult the official Libreswan security advisory for the specific patched version applicable to your release series. As a temporary mitigation, you may disable IKEv2 fragmentation by setting fragmentation=no in your Libreswan configuration, though this may reduce interoperability with peers that require fragmentation support. Restart the pluto daemon after configuration changes. If patching is not immediately feasible, consider restricting network access to IKEv2 ports (UDP 500, 4500) to trusted peer IP ranges using firewall rules.

Patch guidance

Contact Libreswan or check the official security advisories for the patched version applicable to your installation. Patches will be released for currently supported Libreswan branches; verify your version is still within the support lifecycle before assuming a patch is available. After applying the patch, perform a controlled daemon restart and verify IKEv2 tunnel establishment with trusted peers. If you operate a redundant VPN infrastructure, stagger patching across nodes to maintain availability. Test fragmentation=no configuration in a non-production environment first if you choose mitigation over patching.

Detection guidance

Monitor pluto daemon logs (typically in /var/log/pluto or journalctl) for repeated crash and restart cycles coinciding with specific source IPs. IKEv2-aware network sensors may observe malformed fragment payloads in IKEv2 Initial Exchange or subsequent messages. Unusual spikes in IKEv2 messages to your VPN endpoints followed by daemon restarts warrant investigation. Implement alerting on pluto daemon restarts; a high frequency of restarts (multiple per minute) is a strong indicator of active exploitation. Packet capture on port 500/4500 during an incident can reveal the malformed packet structure, though the off-by-one nature means the payload count will marginally exceed the valid range rather than being obviously corrupt.

Why prioritize this

Despite not being on the Known Exploited Vulnerabilities (KEV) list, this vulnerability merits urgent attention because: (1) exploit complexity is very low—any network-positioned attacker can trigger it; (2) impact is service-critical for VPN operations; (3) the vulnerability affects the default configuration (fragmentation enabled), requiring no unusual setup to be vulnerable; (4) no code execution or data exposure is possible, allowing faster patch deployment windows without a separate risk of lateral movement; (5) organizations that patch this issue will eliminate a clear and present DoS vector. Prioritize this as a HIGH-severity issue for VPN infrastructure but recognize that patching is lower-risk than higher-severity code execution flaws.

Risk score, explained

The CVSS 7.5 score reflects: Network-accessible attack vector (AV:N), low attack complexity (AC:L), no privilege or user interaction required (PR:N, UI:N), scope unchanged (S:U), no confidentiality or integrity impact (C:N, I:N), and high availability impact (A:H). The score appropriately captures the ease of exploitation and the real-world harm to service continuity, while the absence of CIA impact prevents a CRITICAL rating. For risk prioritization, also factor in: whether your Libreswan instance is internet-facing, the criticality of the VPN tunnels it supports, and your organization's tolerance for brief connectivity disruptions.

Frequently asked questions