[CVE-2026-98490][MODERATE 7.0] net: openvswitch: conntrack: fix helper UAF due to extensions realloc

[CVE-2026-98490][MODERATE 7.0] net: openvswitch: conntrack: fix helper UAF due to extensions realloc

[AL-KERNEL]

CVE: CVE-2026-98490
Priority: MODERATE 7.0
AL-KERNEL base severity: MODERATE
KPANIC flag: YES
Patch: net: openvswitch: conntrack: fix helper UAF due to extensions realloc
Commit: afcf362625190c5260530b1d6c7daa98831f6b83
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=afcf362625190c5260530b1d6c7daa98831f6b83
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-98490
Analysis date: Mon, 19 Oct 2026 06:13:21 -0400
ActionableScore: 5
ActionableScore lower bound: 4
Actionable bucket: Actionable Moderate / Strong Important candidate requiring manual review
Manual review required: YES

Summary:
A conntrack helper can leave an expectation linked to a reallocated and freed extension area, causing a real active-state UAF and stale hlist metadata access during expectation removal, with kernel crash and unresolved list-corruption consequences for traffic crossing a configured Open vSwitch helper path.

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

The original Linux kernel CVE announcement for CVE-2026-98490 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-98490

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-2026-98490	MODERATE	https://www.kernelcve.org/list/?q=CVE-2026-98490 CHECK WITH IMPACT FROM ORIG NN IMPORTANT | AEGIS IMPACT MODERATE_KPANIC AEGIS_AS=4 LEGACY=MODERATE_KPANIC RESULT=SAME (final reconciliation agrees; DECISION: both MODERATE_KPANIC)	Maybe valid. Check manually. Hints by AL-KERNEL: The best (paranoid) CVSS is 'AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:H/A:H';*CWE-416;CWE-825;*CWE-664;Other CVSS 'AV:A/AC:H/PR:N/UI:N/S:U/C:L/I:L/A:H';BEST CVSS score: '8.1';DESCR 'A use after free can occur in the Open vSwitch conntrack path because a helper may link an expectation to the conntrack extensions area before later extension allocations reallocate that area. The expectations list can then retain a pointer into freed memory and access it when the expectation is removed. A packet processed by a configured conntrack helper may reach this path from a guest, container, or network peer without requiring the attacker to configure Open vSwitch. For the CVSS the PR:N reflects the paranoid scenario where the vulnerable helper and flow rules are already configured and the attacker only needs to provide matching traffic. Network reachability depends on the Open vSwitch topology and helper configuration. The base score uses AV:A for the common virtual or internal network boundary, while the paranoid score uses AV:N for deployments that pass routed remote traffic through the affected helper. Impact is at least denial of service through kernel memory corruption and a possible kernel crash. The stale list pointer may also cause writes through hlist metadata, so confidentiality and integrity impact or privilege escalation cannot be excluded without manual analysis, although a controlled arbitrary write is not established. Automated impact decision path. The original SimpleNet classifier selected IMPORTANT. The primary LLM estimated paranoid CVSS 8.1 and the production ActionableScore returned 5 with recommendation Actionable Moderate / Strong Important candidate requiring manual review. The production rules then evaluated CVSS thresholds and ActionableScore thresholds. Recorded rule transitions were KP and FD specialist checks stage IMPORTANT to MODERATE_KPANIC because a specific evidence based protection or manual review veto was applied and FD false positive review allowed a downgrade. The legacy KP specialist produced a review artifact and its decision was processed by the applicable KP rules. The legacy FD or DH downgrade specialist produced a review artifact and its decision was processed by the applicable downgrade and preservation rules. Exact triggered rule markers were V33_ACTIVE_STATE_BOUNDARY_CANDIDATE and IMPORTANT_CANDIDATE_MANUAL_REVIEW and KP_ONLY_DOWNGRADE_VETO and DECREASED_TO_MODERATE7_BASED_ON_FALSEPOSCHECKOFDH. The independent weaker Aegis shadow evaluated CVSS 3.1 and ActionableScore 4 with primitive class CLASS_B and manual review REQUIRED. Its independently reconciled impact was MODERATE_KPANIC. The shadow agreed with the final production impact and aggregation did not change that impact. The final reported impact is MODERATE_KPANIC.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Actionable Moderate / Strong Important candidate requiring manual review (with actual score 5)	MAYBE	UAF SKB LINUS KPANIC  V33_ACTIVE_STATE_BOUNDARY_CANDIDATE IMPORTANT_CANDIDATE_MANUAL_REVIEW KP_ONLY_DOWNGRADE_VETO DECREASED_TO_MODERATE7_BASED_ON_FALSEPOSCHECKOFDH	-	-	checked

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

ActionableScore=5
ActionableScoreLower=4
PrimitiveClass=CLASS_B
RealUAF=YES
HostAdminRequired=UNKNOWN
NamespaceScopedPrivilege=UNKNOWN
ImportantCandidate=YES
ManualReview=REQUIRED
AutoDowngradeAllowed=NO
RequiresHostUID0=UNKNOWN
AuthorizationDecisionBypass=NO
ProtectedResourceImpact=NONE
B4PendingReview=NO
B4PendingConfirmed=NO

1. ActionableScore

* Raw conservative score: 3.
* Conservative score: 4. REAL UAF FLOOR 4 changes 3 to 4 because an active stale list reference is used after the conntrack extension allocation is freed, while host-admin-equivalent privilege for the packet-side trigger is not established.
* Raw paranoid score: 5.
* Paranoid score: 5. No floor or cap changes the result.
* Final recommended bucket: **Actionable Moderate / Strong Important candidate requiring manual review**::

2. Signal breakdown

Conservative signals:

* +2 Real UAF. `master_help->expectations` is embedded in the conntrack extension area. Later `*_ext_add()` reallocation invalidates the list-head address while an expectation retains a link to it.
* +1 Realistic kernel crash or strong DoS. Removing the expectation later accesses stale list metadata in freed storage and can cause a fault or list corruption.
* Raw conservative total: 3.
* FLOOR 4 applied because this is active structured list state rather than fixed stale storage or a shallow guaranteed-fault pointer.

Additional paranoid signals:

* +1 Structured stale-object exploitation leverage. Expectation removal consumes the stale `hlist` linkage and can update list metadata through an obsolete `pprev` relationship. This is meaningful post-lifetime processing beyond a passive stale read.
* +1 Attacker-influenced active-state lifecycle. In a deployment with the affected helper configured, matching traffic can create expectations and drive connection or expectation teardown repeatedly.
* Raw paranoid total: 5.

V33 candidate analysis:

* Write: normal `hlist` removal can write through linkage derived from the stale expectations-list head in the freed extension area.
* Privileged consumer: kernel conntrack expectation removal consumes and modifies this active list state.
* Trust boundary: traffic from a guest, container, delegated network, or network peer may exercise an already configured Open vSwitch conntrack helper. The complete boundary is credible but unresolved from the supplied patch.
* Result: V33 supports `ImportantCandidate=YES`, `ManualReview=REQUIRED`, and `AutoDowngradeAllowed=NO`. It does not prove host panic, arbitrary write, privilege escalation, or Important severity.

Not awarded:

* No direct remote-reachability +2 because the vulnerable path requires prior Open vSwitch flow and conntrack-helper configuration.
* No local-unprivileged +1 because the complete below-host-admin trigger path is not established.
* No strong corruption +2 because controlled reclaim, attacker-selected destination, attacker-controlled replacement contents, and arbitrary write are not demonstrated.
* No confidentiality or integrity impact points. The stale list write is real, but modification of a specific protected object or disclosure path is not established.
* No extra availability point because the possible crash is already counted.
* No repeated-release or refcount point. The patch does not show two lifetime-mutating puts or releases on the same object.
* NEW52 B5-R does not apply because this is not a retry-accumulated heap OOB write.
* NEW53 B3-PENDING does not apply because no shared or file-backed storage write is involved.
* NEW54 B4-PENDING does not apply because no authorization decision changes from DENY to ALLOW.

3. Reachability analysis

A privileged administrator normally configures the Open vSwitch datapath, conntrack action, and helper. After that setup, matching packets may invoke the helper, create an expectation for an unconfirmed connection, cause additional conntrack extensions to be allocated during commit, and later drive expectation removal.

The supplied evidence does not establish whether the complete trigger is available to an ordinary remote peer, an unprivileged guest, a container with a delegated interface, or only a network-management actor. Therefore `HostAdminRequired=UNKNOWN` rather than YES or NO.

Open vSwitch and conntrack operations can be associated with network namespaces, but the patch does not establish that every configuration and trigger stage is available in an attacker-created namespace. Therefore `NamespaceScopedPrivilege=UNKNOWN`.

Prior privileged setup is a credible prerequisite. It prevents awarding direct remote-reachability points, but it does not prove that the packet sender must possess administrative privileges. No special hardware, boot mode, or timing race is required. The flaw is an ordered allocation and lifetime sequence rather than a narrow race.

4. Primitive and severity interpretation

The exact primitive is a stale pointer into a reallocated conntrack extension block. The expectation list is linked to `master_help->expectations` before subsequent extension additions can replace and free the old extension allocation. Later expectation removal accesses linkage associated with that freed storage.

This is an active structured-object UAF, not fixed stale storage. The later operation is broader than a read-only cleanup dereference because list removal can mutate linkage through stale `hlist` metadata. It is not a repeated-release or refcount UAF.

The demonstrated consequence is post-lifetime list access with realistic crash or list-corruption potential. A controlled reclaim, directed kernel write, disclosure, callback takeover, or privilege escalation is not demonstrated. Those remain speculative.

`PrimitiveClass=CLASS_B` is appropriate because a real active-state lifetime and list-corruption primitive exists, but Class A control is not established. `ImportantCandidate=YES` follows from the active structured state, write-like list consumption, privileged kernel consumer, and credible unresolved guest, namespace, or network boundary. `AutoDowngradeAllowed=NO` preserves this unresolved exploitation question for human review without asserting that the issue is already proven Important.

5. One-sentence report phrase

A conntrack helper can leave an expectation linked to a reallocated and freed extension area, causing a real active-state UAF and stale hlist metadata access during expectation removal, with kernel crash and unresolved list-corruption consequences for traffic crossing a configured Open vSwitch helper path.

6. Manual review recommendation

MANUAL CHECK REQUIRED

Review should determine whether unprivileged guest, container, or remote traffic can reliably create and remove the affected expectations, and whether allocator reuse turns the stale `hlist` update into controlled corruption beyond denial of service.

7. Call-site confidence

Call-site confidence: medium

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

Patch: net: openvswitch: conntrack: fix helper UAF due to extensions realloc
Commit: afcf362625190c5260530b1d6c7daa98831f6b83
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=afcf362625190c5260530b1d6c7daa98831f6b83

Commit description:

While calling the helpers, a raw pointer to the extensions area is
wired into expectations list:

  -> nf_ct_helper()
   -> helper->help()
    -> nf_ct_expect_related_report()
     -> nf_ct_expect_insert()
      -> hlist_add_head_rcu(&exp->lnode, &master_help->expectations)

In case the connection is not confirmed yet, more extensions can be
added afterwards with *_ext_add() calls reallocating the extension
space and leaving the now invalid pointer in the expectations list
that is later accessed while removing the expectation.

Make sure that helpers are called at the end after all the other
extensions are already added.

Note that the helper rejection now leaves the mark and labels set,
but that's not different from how the NAT was handled before or how
the mark and the labels were handled on confirmation failure.  And
there are no atomicity guarantees provided by the API anyway.

Fixes: cae3a26 ("openvswitch: Allow attaching helpers to ct action")
Cc: stable@vger.kernel.org
Reported-by: Axel Mierczuk <axel.mierczuk@1password.com>
Signed-off-by: Ilya Maximets <i.maximets@ovn.org>
Reviewed-by: Aaron Conole <aconole@redhat.com>
Link: https://patch.msgid.link/20260921145655.3167436-4-i.maximets@ovn.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Changed files:
  net/openvswitch/conntrack.c

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=afcf362625190c5260530b1d6c7daa98831f6b83

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

The original Linux kernel CVE announcement for CVE-2026-98490 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-98490

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:A/AC:H/PR:N/UI:N/S:U/C:L/I:L/A:H
  The Best / paranoid CVSS vector: AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:H/A:H
  The Best / paranoid CVSS score: 8.1

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: 4
  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).