CVE-2026-72137 PUBLISHED

xfrm: nat_keepalive: avoid double free on send error

Assigner: Linux
Reserved: 09.08.2026 Published: 15.08.2026 Updated: 17.08.2026

In the Linux kernel, the following vulnerability has been resolved:

xfrm: nat_keepalive: avoid double free on send error

nat_keepalive_send() frees the keepalive skb whenever the IPv4 or IPv6 send helper reports an error.

That cleanup is only correct before the skb is handed to the output path. Once ip_build_and_send_pkt() or ip6_xmit() takes ownership, the networking stack may already have consumed the skb before returning an error, so freeing it again is unsafe.

Handle the pre-handoff failure cases inside nat_keepalive_send_ipv4() and nat_keepalive_send_ipv6(), where the caller still owns the skb, and keep nat_keepalive_send() responsible only for family dispatch and the unsupported-family cleanup path.

Metrics

CVSS Vector: CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H
CVSS Score: 9.8

AV:N - On internet-facing IPsec VPN gateways using kernel NAT keepalive, a remote attacker can induce post-handoff outbound send failures (e.g., ENOMEM from network-driven memory pressure) when periodic keepalives are transmitted, reaching the vulnerable double-free path. AC:L - After an outbound ESP-in-UDP SA has NAT keepalive configured, an attacker can reliably force send errors where the stack already consumed the skb (ip6_xmit EMSGSIZE, ip_finish_output2 neighbour/ENOMEM paths) without uncontrollable races or memory layout dependencies. PR:N - Remote exploitation requires no privileges on the victim; it only needs a typical IPsec NAT-T deployment where userspace already set XFRMA_NAT_KEEPALIVE_INTERVAL. Local exploitation via user namespaces needs only CAP_NET_ADMIN (PR:L), not root in the init namespace. UI:N - No victim or administrator interaction is needed during exploitation; NAT keepalives are sent automatically by the kernel delayed work queue on the configured interval once the XFRM SA exists. S:U - Impact is kernel heap corruption and privilege escalation within the same security authority; this is not a VM escape, container breakout, or IOMMU/DMA boundary bypass. C:H - The skb double-free corrupts the slab allocator and yields use-after-free primitives that can be leveraged for arbitrary kernel memory disclosure, not merely a bounded leak or crash-only NULL dereference. I:H - Double-free heap corruption enables attacker-controlled reallocation of freed sk_buff objects, supporting arbitrary kernel memory writes, control-flow hijacking, and local privilege escalation. A:H - The double-free reliably causes slab corruption, kernel oops, or panic during keepalive send-error cleanup, providing immediate and repeatable denial of service on affected systems.

Product Status

Vendor Linux
Product Linux
Versions Default: unaffected
  • affected from f531d13bdfe3f4f084aaa8acae2cb0f02295f5ae to d0a4dc7efa825bce60a8da8f7d43c864a159abde (excl.)
  • affected from f531d13bdfe3f4f084aaa8acae2cb0f02295f5ae to 5b0c4c916f202b8fd13d12afb6af62b385622f81 (excl.)
  • affected from f531d13bdfe3f4f084aaa8acae2cb0f02295f5ae to a8a7e6a9ff8a4c1f067694ddbd44be67fdf36693 (excl.)
  • affected from f531d13bdfe3f4f084aaa8acae2cb0f02295f5ae to 226f4a490d1a938fc838d8f8c46a4eca864c0d78 (excl.)
Vendor Linux
Product Linux
Versions Default: affected
  • Version 6.11 is affected
  • unaffected from 0 to 6.11 (excl.)
  • unaffected from 6.12.101 to 6.12.* (incl.)
  • unaffected from 6.18.40 to 6.18.* (incl.)
  • unaffected from 7.1.5 to 7.1.* (incl.)
  • unaffected from 7.2 to * (incl.)

References