CVE-2023-54021

What it actually does

Incorrect assignment of goal extent in ext4 block allocator (ac_f_ex instead of ac_g_ex) combined with missing range validation on ar->pright - size allows underflow that triggers BUG_ON in ext4_get_group_no_and_offset during mb_normalize_request.

Why it matters to the business

Kernel panics or filesystem instability on ext4 volumes can cause unplanned downtime and service interruption; low exploitation likelihood reduces urgency but still requires patching of affected kernels and dependent container images.

How an attack unfolds

Local attacker triggers repeated ext4 block allocation requests crafted so that normalize_request receives a size that underflows the right-bound calculation, hitting the BUG_ON path; no named threat actors or ransomware campaigns observed.

How to fix it

Upgrade Linux kernel to commit 2479bb6cbdb4d56b807bbe5229e3e26a6f1f4530 or later; update affected packages to versions containing the fix (e.g., 0.13.0-r1 equivalents).

How to confirm the fix worked

Confirm presence of the range check before calling ext4_get_group_no_and_offset and verify ac_g_ex is set in mb_normalize_request; test allocation paths that previously produced underflow.

Typical attack path

1. Obtain local execution context on target system
2. Perform filesystem operations that invoke ext4_mb_normalize_request with boundary values
3. Trigger underflow in ar->pright calculation
4. Force kernel BUG_ON and resulting panic

How to detect it

Search kernel logs and crash dumps for BUG_ON triggered inside ext4_get_group_no_and_offset following calls from ext4_mb_normalize_request.

• BUG_ON triggered in ext4_get_group_no_and_offset
• ext4_mb_normalize_request underflow

What your security team should do

• Prioritize kernel and container image updates within standard change windows
• Monitor kernel logs for ext4 allocation BUG_ON messages
• Validate ext4 filesystems after patching for continued stability