CVE-2024-56632

What it actually does

Failure to release the tagset allocated for the admin queue when NVMe TCP controller creation fails, causing a memory leak in the kernel nvme-tcp driver.

Why it matters to the business

Repeated controller initialization failures in NVMe over TCP setups can gradually exhaust system memory, leading to degraded performance, service disruptions, or forced reboots in storage-dependent environments.

How an attack unfolds

An attacker induces repeated failures during NVMe TCP controller creation through configuration manipulation or resource pressure, preventing tagset cleanup and accumulating leaked memory until the system becomes unstable.

How to fix it

Update the Linux kernel to a version containing the nvme-tcp memory leak fix for failed controller creation; rebuild and deploy the patched kernel module if custom builds are used.

How to confirm the fix worked

Confirm the fix is present by inspecting nvme-tcp source for tagset release on admin_q failure paths; monitor kernel memory statistics during controller creation stress tests.

Typical attack path

1. Identify systems using nvme-tcp for storage connectivity
2. Trigger repeated admin queue or controller creation attempts that fail
3. Prevent tagset release on each failure path
4. Accumulate memory consumption over multiple attempts
5. Cause resource exhaustion or force system restart

How to detect it

Review kernel logs and dmesg for nvme-tcp controller creation errors; track memory usage patterns in slabinfo or kmemleak reports around nvme module activity.

What your security team should do

• Prioritize kernel updates on systems using nvme-tcp storage
• Audit nvme-tcp controller creation logs for frequent failures
• Implement monitoring for unexpected memory growth in kernel slab caches
• Test patched kernels in staging before production rollout