CVE-2026-64386 is a double-free in the Linux kernel's SMB client, specifically in the SMB2_query_info_init() code path when the client attempts to retry a request after a replayable error. The root cause is straightforward: when a retry occurs, the response buffer from the previous attempt is freed during normal error handling, but the bookkeeping state tracking that buffer isn't reset. If the subsequent retry fails before re-sending — specifically at the query_info_init stage — the cleanup code attempts to free the same buffer again, producing the double-free. The fix is described as resetting response bookkeeping before each retry attempt, which tells you everything about the vulnerability's nature: the state from a failed request persists into the next attempt's error path, and every caller of the replay mechanism must manually reset that state. This is not merely a bug in one function — it's a structural flaw in how the SMB client's retry infrastructure manages state across attempts. The replay mechanism is shared across multiple SMB2 operations, meaning the same pattern could exist elsewhere in the client if it hasn't been audited. The CVSS 9.8 reflects kernel-context exploitation severity, but the practical exploitability requires either a man-in-the-middle position or a malicious SMB server triggering the specific failure sequence. What should concern you: if this pattern exists in query_info(), it likely exists in other operations using the same replay path. Treat this CVE as a signal to audit other SMB2 operation handlers that use the retry infrastructure — particularly those added in SMB3 or later protocol extensions where the replay code was extended without formal state-machine redesign. The thinness of the public fix description ('reset response bookkeeping') suggests this may be a per-caller band-aid rather than a holistic refactoring, leaving the underlying architectural gap intact for other code paths. Monitor for follow-on CVEs in the next 6-12 months hitting the same infrastructure in different SMB operations — that would confirm this was a class-level vulnerability, not an isolated fix.
CVE-2026-64386
Official description Straight from the sourceThe vendor's or NVD's own wording, published unedited. Authoritative, but often terse — it says what broke, rarely what to do.
NVD · uneditedIn the Linux kernel, the following vulnerability has been resolved: smb: client: fix query_info() replay double-free A response-bearing attempt can return a replayable error and free its response buffer. If SMB2_query_info_init() fails before the next send, cleanup retains the previous buffer type and frees that response again. Reset response bookkeeping before each attempt to prevent the stale free.
Technical summary Written by usOur analysis, written from the advisory, the CVSS vector and the affected-version data. It adds context the advisory leaves out, and never invents facts that are not in the source.
dbcve analysis · moderate confidenceA double-free vulnerability exists in the Linux kernel's SMB client query_info() replay handling. When a response-bearing attempt returns a replayable error and frees its response buffer, a subsequent failure in SMB2_query_info_init() before the next send causes the cleanup code to retain the previous buffer type and free the same response buffer again.
Verify against the referenced sources before acting — the references below are authoritative for this CVE, this summary is not.
CVSS breakdown How the score is builtThe industry scoring standard. It rates how the flaw is reached, what it takes to exploit, and what an attacker gains — the score is derived from those, not the other way round.
From the vector- Attack vector
- Network
- Complexity
- Low
- Privileges
- None
- User interaction
- None
- Scope
- Unchanged
- Confidentiality
- High
- Integrity
- High
- Availability
- High
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H
Am I affected? How to checkSteps we derive from the advisory and the affected-version data, so you can decide whether this CVE reaches your setup. They are a guide, not a scan — your own configuration is the authority.
dbcve checksWork through these to decide whether this CVE applies to you.
-
Identify kernel versionRun `uname -r` to get the running kernel version. Compare this to the version that includes the fix for CVE-2026-64386.Affected if The kernel version is older than the patched version containing the fix.
-
Confirm CIFS/SMB client is availableRun `lsmod | grep cifs` to check if the CIFS kernel module is loaded, or check `/proc/filesystems` for 'cifs' to see if SMB client support is built into the kernel.Affected if The CIFS module is loaded or the client is built-in, enabling the vulnerable code path.
-
Check for active SMB mountsRun `mount -t cifs` or examine `/etc/mtab` or `/proc/mounts` for any cifs mounts that would trigger the query_info() code path.Affected if There are active CIFS mounts using the vulnerable SMB client code.
-
Inspect SMB client configurationCheck `/proc/fs/cifs/DebugData` (if available) or examine loaded module parameters with `cat /sys/module/cifs/parameters/*` to see client configuration.Affected if The SMB client is configured and active, making the double-free condition reachable.
A system is affected if it runs a kernel version prior to the fix, has the CIFS/SMB client enabled (loaded or built-in), and has active CIFS mounts that would trigger the query_info() code path where the double-free can occur.
Generated from the published advisory. Verify against your own configuration.
Remediation Closing itWhat it takes to close this. Where a vendor fix exists we point at it; where none exists we say so plainly, and can build one. Effort estimates are scoped from the advisory, not from your codebase.
From vendor dataApply the kernel patch that resets response bookkeeping before each attempt to prevent the stale free. This requires updating the Linux kernel with the fix.
- Consultation2.0 h
- Implementation3.0 h
- Testing2.0 h
- Review / QA1.0 h
An estimate, not a bill — we confirm scope with you before any work starts. Need it this week? Rush from $2,272.
Scan for this in your stack
Free · runs locallyCheck whether your project pulls in CVE-2026-64386 — or any other known-vulnerable package — straight from your lock files. Free and open source; it runs locally and uploads nothing.
References Go to the primary sourcePrimary sources — vendor advisories, patches and trackers. Where our summary and a reference disagree, the reference wins.
Primary sourcesCVE-2026-64386 is a double-free in the Linux kernel's SMB client, specifically in the SMB2_query_info_init() code path when the client attempts to retry a request after a replayable error. The root cause is straightforward: when a retry occurs, the response buffer from the previous attempt is freed during normal error handling, but the bookkeeping state tracking that buffer isn't reset. If the subsequent retry fails before re-sending — specifically at the query_info_init stage — the cleanup code attempts to free the same buffer again, producing the double-free. The fix is described as resetting response bookkeeping before each retry attempt, which tells you everything about the vulnerability's nature: the state from a failed request persists into the next attempt's error path, and every caller of the replay mechanism must manually reset that state. This is not merely a bug in one function — it's a structural flaw in how the SMB client's retry infrastructure manages state across attempts. The replay mechanism is shared across multiple SMB2 operations, meaning the same pattern could exist elsewhere in the client if it hasn't been audited. The CVSS 9.8 reflects kernel-context exploitation severity, but the practical exploitability requires either a man-in-the-middle position or a malicious SMB server triggering the specific failure sequence. What should concern you: if this pattern exists in query_info(), it likely exists in other operations using the same replay path. Treat this CVE as a signal to audit other SMB2 operation handlers that use the retry infrastructure — particularly those added in SMB3 or later protocol extensions where the replay code was extended without formal state-machine redesign. The thinness of the public fix description ('reset response bookkeeping') suggests this may be a per-caller band-aid rather than a holistic refactoring, leaving the underlying architectural gap intact for other code paths. Monitor for follow-on CVEs in the next 6-12 months hitting the same infrastructure in different SMB operations — that would confirm this was a class-level vulnerability, not an isolated fix.
Practitioner notes
ContributedPeer-ranked notes from engineers who’ve handled CVE-2026-64386 in production — separate from our analysis above.
The advisory tells you what broke. It rarely tells you what actually worked. If you’ve dealt with this one, that detail is what the next engineer is searching for.
- The version that genuinely resolved it — not the one the vendor claimed
- A config change or rule that shut the vector down
- A gotcha in the upgrade path that cost you an afternoon
A place for practitioners to share what actually worked: a mitigation you’ve tested, a configuration change, a version- or environment-specific caveat, or a link to a verified patch. The most useful notes rise to the top as peers upvote them, so the signal stays high.
- Verified mitigations, workarounds, and config changes
- Version or environment caveats, and links to real fixes
- No weaponised exploit code, or anything meant to cause harm
- No spam, self-promotion, credentials, or personal data