Skip to content
NewA calmer, faster ZeroHour

How I Found a $113,337 AF_ALG Linux Local Privilege Escalation Before Copy Fail

Lobsters · securityby idnsec.com via fro ()14 min read4 · 1 read
AI summary · grok-4.7Summaries by Vani AI

Researcher details CVE-2025-39964, an AF_ALG race that gave unprivileged Linux users root and a $113,337 kernelCTF prize.

Muhammad Alifa Ramdhan, with Bing-Jhong Billy Jheng at STAR Labs, describes finding CVE-2025-39964, a race between writers sharing an AF_ALG socket in the Linux kernel. The flaw, present since about 2011 and analyzed on kernel 6.12.44, lets an unprivileged user become root and can escape a Docker container to root on the host. It was responsibly disclosed and earned a $113,337 Google kernelCTF reward. The author distinguishes it from the later Copy Fail AEAD logic bug, which also uses AF_ALG.

Key facts
  • CVE-2025-39964 is a race between writers on one AF_ALG socket.
  • Unprivileged socket API access can become root on the host.
  • The same bug can escape Docker to root on the host.
  • Present since about 2011; analyzed on Linux kernel 6.12.44.
  • Responsibly disclosed; Google kernelCTF reward was $113,337.
Analyze CVE-2025-39964 with Vani AIImpact, affected versions, exploitation, mitigations and detection

Vulnerabilities mentioned

All
CVEVulnerabilityCVSSEPSSSignalsPublished
CVE-2025-39964
Race Condition in Linux Kernel Crypto Socket (af_alg_sendmsg)
5.52%KEV
KEV

From the original article

14 min read at idnsec.com

In 2025, I found an AF_ALG vulnerability in the Linux kernel that allowed an ordinary user to escalate privileges to root. This is a retrospective on how I found CVE-2025-39964 and how we developed the exploit, before Copy Fail drew wider attention to AF_ALG in 2026. Editorial (IDNSEC): Muhammad Alifa Ramdhan discovered CVE-2025-39964 while conducting research at STAR Labs.

Lobsters · security

By idnsec.com via fro · idnsec.com

Read the full article at Lobsters · security

A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.