[CVE-2026-89990][MODERATE 7.0] ceph: lock mutex in ceph_mds_check_access()

[AL-KERNEL <alkernel@kernelcve.org>]

CVE: CVE-2026-89990
Priority: MODERATE 7.0
AL-KERNEL base severity: MODERATE
KPANIC flag: YES
Patch: ceph: lock mutex in ceph_mds_check_access()
Commit: 729c1ff670ece6f3d4b1738365a0b2f90d487523
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=729c1ff670ece6f3d4b1738365a0b2f90d487523
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-89990
Analysis date: Wed, 16 Sep 2026 14:07:40 -0400
ActionableScore: 5
ActionableScore lower bound: unknown
Actionable bucket: Actionable Moderate / Strong Important candidate requiring manual review
Manual review required: YES

Summary:
No one-sentence report phrase was found.

======================================================================
ABOUT THIS REPORT
======================================================================

The original Linux kernel CVE announcement for CVE-2026-89990 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-89990

The original announcement does not normally provide a security severity
estimate, CVSS assessment, or enough information to determine whether the
reported kernel bug represents a practically relevant security issue.

This report was generated by AL-KERNEL, an AI-assisted Linux kernel
vulnerability analysis system developed by Alexander Larkin. It combines
an autonomous classifier with LLM-assisted technical analysis and a
separate ActionableScore mechanism.

The purpose of this report is to prioritize Linux kernel CVEs before
manual review, identify cases that require prompt investigation, and
support automatic closure of issues that are unlikely to have meaningful
security impact.

Published priority for this report: MODERATE 7.0
Manual review required: YES

A detailed explanation of the methodology and priority rules is included
at the end of this message.

======================================================================
AL-KERNEL CLASSIFICATION RESULT
======================================================================

CVE-2026-89990	MODERATE	https://www.kernelcve.org/list/?q=CVE-2026-89990 CHECK WITH IMPACT FROM ORIG NN IMPORTANT	Maybe valid. Check manually. Hints by AL-KERNEL: The best (paranoid) CVSS is 'AV:L/AC:H/PR:L/UI:N/S:U/C:H/I:H/A:H';*CWE-416;CWE-362;CWE-667;Other CVSS 'AV:L/AC:H/PR:L/UI:N/S:U/C:N/I:N/A:H';BEST CVSS score: '7';DESCR 'A race in CephFS authorization handling allows MDS session OPEN processing to replace and free mdsc s_cap_auths while ceph_mds_check_access is traversing the same array and its strings. This produces a confirmed use after free read during a filesystem open operation and can cause a kernel Oops and denial of service. For the CVSS the PR:L is used because an unprivileged local process with access to the mounted CephFS namespace can invoke the affected access check. AC:H is used because exploitation requires overlap with an MDS session reopen. The issue is not directly reachable through arbitrary unauthenticated network traffic, although reconnect or MDS session activity supplies the concurrent event needed for the race. Impact is at least a local denial of service via kernel crash. In the worst case reused authorization memory may affect an access decision or support broader confidentiality and integrity impact, so the UAF requires manual security review.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Actionable Moderate / Strong Important candidate requiring manual review (with actual score 5)  SKIP CVE-2026-89990 UNKNOWN SKIP The Fixes patch not applied yet, so unlikely that actual: url YES  NO NO unknown 	MAYBE	OOB LOCK DISK UAF HARDWARE LINUS KPANIC  DECREASED_TO_MODERATE7_BASED_ON_FALSEPOSCHECKOFDH	-	-	checked

======================================================================
ACTIONABLESCORE ANALYSIS
======================================================================

ActionableScore=5
PrimitiveClass=CLASS_B
RealUAF=YES
HostAdminRequired=NO
NamespaceScopedPrivilege=UNKNOWN
ImportantCandidate=YES
ManualReview=REQUIRED
AutoDowngradeAllowed=NO

1. ActionableScore

* Raw conservative score: 5
* Conservative score: 5
* Raw paranoid score: 5
* Paranoid score: 5
* No floor or cap changed the score.
* Final recommended bucket: **Actionable Moderate / Strong Important candidate requiring manual review**::

2. Signal breakdown

Positive signals:

* +2 Real UAF / active lifetime violation. MDS session OPEN processing can free and replace `mdsc->s_cap_auths` and its strings while `ceph_mds_check_access()` still traverses them.
* +1 Reliable kernel crash / strong DoS. The supplied trace demonstrates a kernel paging fault and Oops in `ceph_mds_check_access()`.
* +1 Local unprivileged trigger. A non-root process reaches the stale traversal through `openat2()` on an accessible CephFS path. No administrative operation is required from that process during the observed access.
* +1 Important filesystem path. The stale data is consumed by the CephFS authorization path during file opening.

Negative signals:

* No timing penalty. The issue is a race, but the supplied real-world crash does not establish a narrow, one-shot, or materially uncontrollable timing window.
* No privilege penalty. The crashing access path is demonstrated in a non-root process. The CephFS mount requires prior setup, but host administration is not required for each vulnerable file access.

Important non-awarded signals:

* Controlled reclaim or attacker-shaped replacement contents are not demonstrated.
* Structured stale-object leverage is not separately awarded because the local caller is not shown to control replacement or reuse of the authorization array.
* No concrete authorization bypass is awarded. The stale data participates in an authorization decision, but an incorrect DENY-to-ALLOW result is not demonstrated.
* Arbitrary or directed kernel write is not demonstrated.
* Confidentiality and integrity points are not awarded merely because freed authorization data could theoretically be reused.
* No repeated release or refcount mutation exists. The stale operation reads and traverses freed objects rather than releasing them again.
* No extra Availability point is awarded because the observed Oops is already counted as the reliable crash.
* Remote reachability is not awarded. MDS session activity supplies the concurrent writer, but arbitrary unauthenticated network traffic is not shown to reach the complete trigger.

3. Reachability analysis

An ordinary local process with permission to access the mounted CephFS namespace can invoke `ceph_mds_check_access()` through a normal file-open operation. The supplied trace shows this path from a non-root `php-cgi8.4` process, positively establishing below-host-admin reachability for the reader side.

The concurrent freeing operation occurs when MDS session OPEN handling replaces `s_cap_auths`. The local process is not shown to control that session transition directly. It can arise from Ceph client session reopen or reconnection activity, so the complete race combines an ordinary filesystem operation with an asynchronous session event.

A CephFS mount and working MDS connection are prior administrative deployment prerequisites. No special hardware, boot mode, or attacker-controlled namespace capability is required. The evidence does not establish that the complete setup or session-reopen operation is namespace-scoped, so `NamespaceScopedPrivilege=UNKNOWN`.

The race occurred under a real workload, but attacker control over both sides and reliable repetition are unresolved. This uncertainty suppresses additional exploitation-control bonuses but does not justify a race penalty.

4. Primitive and severity interpretation

The exact primitive is a read-side use-after-free of an array of `struct ceph_mds_cap_auth` objects and associated strings. The writer replaces and frees the old authorization data under `mdsc->mutex`, while the vulnerable reader traverses it without that mutex.

This is active structured storage rather than a fixed stale buffer. The stale fields are consumed by `ceph_mds_auth_match()` during an access-control decision. The stale use occurs in an ordinary file-open path, not only in cleanup or destruction.

There is no concrete repeated release or refcount operation. The first free ends the array and string lifetimes, but the later operation traverses and reads the dangling objects rather than mutating their lifetime metadata.

The bug is therefore `CLASS_B`. The demonstrated consequence is a kernel Oops and denial of service. An incorrect access decision or more useful post-reuse behavior is technically possible because authorization structures are consumed after their lifetime ends, but controlled reclaim, DENY-to-ALLOW bypass, disclosure, arbitrary write, and privilege escalation are not established.

`ImportantCandidate=YES` because this is an active-object UAF below host-admin privilege in a security-relevant authorization path, and the post-reuse behavior requires exploitation-focused review. `AutoDowngradeAllowed=NO` because automatically treating it as an ordinary crash would discard unresolved leverage associated with stale authorization objects. The supported operational classification remains Actionable Moderate, with Important-grade impact not yet proven.

5. One-sentence report phrase

A race between CephFS authorization checks and MDS session reopen can cause a non-root file-open operation to traverse freed capability authorization data, producing a kernel Oops while leaving the security effect of post-free reuse unresolved.

6. Manual review recommendation

MANUAL CHECK REQUIRED

The confirmed active-object UAF is reachable from a non-root filesystem operation and consumes stale authorization data, but controlled reuse and an actual access-control bypass remain unresolved.

7. Call-site confidence

Call-site confidence: high

======================================================================
UPSTREAM PATCH SUMMARY
======================================================================

Patch: ceph: lock mutex in ceph_mds_check_access()
Commit: 729c1ff670ece6f3d4b1738365a0b2f90d487523
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=729c1ff670ece6f3d4b1738365a0b2f90d487523

Commit description:

MDS session OPEN handling replaces mdsc->s_cap_auths under
mdsc->mutex, freeing the previous array and its strings.

ceph_mds_check_access() traverses this array without holding the
mutex.  A concurrent session reopen can therefore free the array while
it is being inspected, resulting in a use-after-free like this:

  Unable to handle kernel paging request at virtual address 003aaad64b2c8bb9
  [...]
  Internal error: Oops: 0000000096000004 [#1]  SMP
  Modules linked in:
  CPU: 56 UID: 2953037534 PID: 1253231 Comm: php-cgi8.4 Not tainted 6.18.45-i2-ampere #1146 NONE
  [..]
  pc : ceph_mds_check_access+0xd4/0x550
  lr : ceph_mds_check_access+0xc8/0x550
  [...]
  Call trace:
   ceph_mds_check_access+0xd4/0x550 (P)
   ceph_atomic_open+0x138/0xbe8
   path_openat+0xa24/0xfa8
   do_filp_open+0x94/0x158
   do_sys_openat2+0x88/0xf8

Cc: stable@vger.kernel.org
Fixes: 596afb0 ("ceph: add ceph_mds_check_access() helper")
Signed-off-by: Max Kellermann <max.kellermann@ionos.com>
Reviewed-by: Alex Markuze <amarkuze@redhat.com>
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Changed files:
  fs/ceph/mds_client.c
  fs/ceph/mds_client.h

Diff excerpt:

Not included in this email. See the upstream URL for the full patch.

Full patch:
https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=729c1ff670ece6f3d4b1738365a0b2f90d487523

======================================================================
DETAILED REPORT METHODOLOGY
======================================================================

The original Linux kernel CVE announcement for CVE-2026-89990 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-89990

The original CVE announcement normally does not include a security-level
estimate. In particular, it may not contain a CVSS assessment, an impact
level, or enough information to determine whether the reported bug is a
practically relevant security issue. One purpose of this parallel CVE list
is to provide that missing technical and prioritization information.

The original goal of the AL-KERNEL project was to prioritize Linux kernel
CVE analysis automatically before manual review. The system can also help
identify non-security issues that may be suitable for automatic closure.

This report was generated by AL-KERNEL, an AI-assisted Linux kernel
vulnerability analysis system developed by Alexander Larkin.

The first analysis stage combines an autonomous classifier with additional
LLM-based analysis. The autonomous classifier runs locally on a CPU and is
based on a backpropagation neural network. Together, these mechanisms
produce a technical vulnerability description, identify likely weakness
types, estimate CVSS severity, and provide input for ActionableScore.

Two CVSS estimates are retained because incomplete kernel vulnerability
information often permits more than one defensible interpretation:

  Conservative CVSS vector: AV:L/AC:H/PR:L/UI:N/S:U/C:N/I:N/A:H
  The Best / paranoid CVSS vector: AV:L/AC:H/PR:L/UI:N/S:U/C:H/I:H/A:H
  The Best / paranoid CVSS score: 7

The conservative vector represents a lower-impact interpretation.
The Best/paranoid vector intentionally represents a plausible upper-bound
interpretation and should not automatically be treated as demonstrated
real-world impact.

CVSS may also need to be adjusted for a particular Linux deployment,
because actual reachability, privileges, enabled kernel configuration,
hardware, namespaces, exposed device nodes, and other environmental
conditions can differ significantly between systems.

A separate ActionableScore mechanism evaluates practical remediation
urgency. Its analysis may include reachability, attack prerequisites,
subsystem exposure, memory-corruption characteristics, denial-of-service
reliability, and possible confidentiality, integrity, or
privilege-escalation impact.

  Conservative ActionableScore: unknown
  Paranoid ActionableScore: 5

The final base severity is taken directly from the second tab-separated
field of the AL-KERNEL classification result. ActionableScore does not
replace or independently override that final AL-KERNEL decision, and
if ActionableScore adjusted impact level of ALKERNEL, then you would see
self-readable flags above like INCREASED_TO_HIGH_BASED_ON_ACTIONABLESCOREHIGHEREQTHAN7.

For an AL-KERNEL result of MODERATE, this report uses the following
additional presentation split:

  ActionableScore below 5   -> MODERATE REGULAR
  ActionableScore 5 or more -> MODERATE 7.0

The distinction between MODERATE REGULAR and MODERATE 7.0 makes it
possible to identify Moderate issues that should receive manual analysis
and fixes before lower-priority MODERATE REGULAR issues. In many cases,
MODERATE REGULAR fixes may wait for a later rebase or routine update.

There is one override in which MODERATE REGULAR becomes MODERATE 7.0
even when the ActionableScore is below 5. When the AL-KERNEL result
contains the KPANIC flag, a MODERATE result is always presented as
MODERATE 7.0. The KPANIC flag selected with few regexps without
usage of AI at all, so it helps to detect cases when Kernel Crash happens
and similar (to filter False-Negative results from the LLM usage).

KPANIC indicates that a reliable kernel crash, kernel panic, or similarly
serious kernel availability impact was identified by the classification
workflow.

AL-KERNEL base severity for this report: MODERATE
KPANIC detected for this report: YES
Published priority for this report (same as in Subject): MODERATE 7.0

These results are intended to support engineering triage. They are
machine-generated estimates, and cases marked for manual review should
be validated by a human security engineer before final disposition.
For more info read docs linked from here: https://kernelcve.org/
(and you can submit you own patch there to generate such a report
for non-existant CVE-id yet).

Note that in many cases this AI tool selects higher severity, than
real is (means you can expect Importants instead of Moderate 7.0 or
Moderates 7.0 instead of regular Moderates). If you see such cases,
please use reply email interface to add additional manual analyses
info to this particular CVE.
And please, please, let me know when you see Lows instead of Importants
or Important instead of Low (because particular for such cases I
need to tune this AI tool to make it better for this one and next similar).
My contact email for such notifications is alexanjelausa@gmail.com (and both
send reply to CVE record itself too and see "reply" button below for howto reply).