Cloudanix VulnDB

CVE-2026-98683

None

CVSS

—

EPSS

0.2%

Published

Oct 18, 2026

CISA due date

—

Cloudanix analysis

Schedule a fix SSVC: Track

A bug in how Linux handles clearing sections of files over network shares could let old information reappear after it should have been erased.

Exploitation
Local Privilege
Attacker needs
Local User
Complexity
Moderate
Impact
Data Tampering
Component
OS kernel
Class
Race Condition
Remediation
Upgrade
Effort
Medium
Patch risk
Low

What it actually does

Race condition in SMB3 client zero-range handling where dirty folios overlapping an interior range (not reaching EOF) are not flushed before truncate_pagecache_range and FSCTL_SET_ZERO_DATA, allowing post-zero writeback of stale data.

Why it matters to the business

Applications reading files over SMB shares may encounter unexpected stale data in regions that were supposed to be zeroed, causing data corruption, incorrect computations, or failed integrity checks with low immediate urgency given recent publication and minimal exploitation signals.

How an attack unfolds

An attacker with local or SMB client access crafts file operations that trigger smb3_zero_range on an interior range while a dirty folio overlaps it; the missing unconditional flush allows the folio to be written back after the server-side zero, refilling the range with stale content observable as random stale reads in tests like generic/363.

How to fix it

Update the Linux kernel to a version containing the unconditional dirty-data flush before truncate_pagecache_range and FSCTL_SET_ZERO_DATA in smb3_zero_range, matching the approach already applied to smb3_punch_hole.

How to confirm the fix worked

Confirm the patch is present by inspecting the smb3_zero_range implementation for an unconditional filemap_write_and_wait_range call prior to cache invalidation, then re-run generic/363 or equivalent zero-range stress tests against Windows servers.

Typical attack path

  1. Mount SMB share from vulnerable Linux client
  2. Create or open file and dirty pages in an interior range
  3. Trigger FALLOC_FL_ZERO_RANGE on that range
  4. Allow writeback to occur after FSCTL completes
  5. Read back the range to observe stale data

How to detect it

Search kernel logs and SMB client traces for repeated zero-range operations followed by unexpected non-zero reads in the same ranges; correlate with filemap_write_and_wait_range absence in stack traces on affected kernels.

What your security team should do

  • Prioritize kernel updates on systems performing frequent SMB writes with fallocate zero-range operations
  • Monitor upstream stable kernel releases for the referenced commit within the next 30 days
  • Review SMB client workloads for use of FALLOC_FL_ZERO_RANGE and test post-upgrade

The record

In the Linux kernel, the following vulnerability has been resolved: smb: client: flush dirty data before zeroing a range smb3_zero_range() emulates FALLOC_FL_ZERO_RANGE by invalidating the page cache over the target range with truncate_pagecache_range() and then issuing FSCTL_SET_ZERO_DATA to the server. Dirty data was only flushed conditionally, when the range reached or extended EOF. For a purely interior zero range that does not reach EOF, no flush happened, so a dirty folio overlapping the range could be written back to the server after the FSCTL and refill the range that was just zeroed with stale data. Fix this by unconditionally flushing and waiting for dirty data in the range before invalidating the page cache and issuing FSCTL_SET_ZERO_DATA, exactly as was done for smb3_punch_hole() in commit d7d2adcd022b ("smb/client: flush dirty data before punching a hole"); the two paths are structurally identical here. This is observed as generic/363 randomly reading stale data where a zeroed range is expected against Windows Server.

Affected packages

Ecosystem Package Introduced Fixed