PHP Warns of Critical Security Flaws Affecting All Supported Versions, Fixes Coming November 19 · 5h 10-Year-Old NGINX HTTP/2 Flaw Lets Unauthenticated Attackers Execute Code on Servers · 16h F5 Warns 6 Critical BIG-IP Flaws Can Crash Traffic in Specific Setups, May Allow RCE · 16h CVE-2026-123633 and CVE-2026-123007: Critical NGINX HTTP/2 Vulnerabilities · 17h Critical Apache Roller RCE Flaw Under Active Exploitation, No Authentication Required · 17h Java4Shell Flaw Threatens Global Enterprises, Java Application Ecosystem · 17h How Cloudflare WAF helps protect against Java4Shell (CVE-2026-114938) · 19h Vulnerabilities in LMS software · 21h New Attack Bypasses TPM Lockout to Crack Windows Hello and BitLocker PINs in Days, Not Years · 22h AI Pentesting Checklist for Web Apps: 15 Questions to Ask Before Choosing a Vendor · 22h PHP Warns of Critical Security Flaws Affecting All Supported Versions, Fixes Coming November 19 · 5h 10-Year-Old NGINX HTTP/2 Flaw Lets Unauthenticated Attackers Execute Code on Servers · 16h F5 Warns 6 Critical BIG-IP Flaws Can Crash Traffic in Specific Setups, May Allow RCE · 16h CVE-2026-123633 and CVE-2026-123007: Critical NGINX HTTP/2 Vulnerabilities · 17h Critical Apache Roller RCE Flaw Under Active Exploitation, No Authentication Required · 17h Java4Shell Flaw Threatens Global Enterprises, Java Application Ecosystem · 17h How Cloudflare WAF helps protect against Java4Shell (CVE-2026-114938) · 19h Vulnerabilities in LMS software · 21h New Attack Bypasses TPM Lockout to Crack Windows Hello and BitLocker PINs in Days, Not Years · 22h AI Pentesting Checklist for Web Apps: 15 Questions to Ask Before Choosing a Vendor · 22h
🛡️
ACyber Sentinel Security Assessment Team
← Back to feed
⚡ Intelligence Report

CVE-2026-98432: Critical Linux Kernel Use-After-Free Flaw

CVE-2026-98432: Critical Linux Kernel Use-After-Free Flaw


The Linux kernel sits at the heart of servers, cloud infrastructure, containers and several enterprise platforms. A vulnerability in its networking stack is therefore something administrators cannot afford to overlook, particularly when the flaw can be triggered remotely without authentication.

Study Security

A newly disclosed vulnerability, CVE-2026-98432, exposes a race condition in the Linux kernel’s IPv4 TCP connection-handling mechanism. The issue involves a use-after-free condition that could allow a remote attacker to trigger kernel memory corruption, crash an affected system or potentially execute arbitrary code.

The vulnerability was disclosed on October 15, 2026, and subsequently attracted attention from major Linux vendors, including Red Hat and Amazon Web Services, which published  security guidance and remediation updates.

Understanding CVE-2026-98432

CVE-2026-98432 is a use-after-free vulnerability in the Linux kernel’s IPv4 networking subsystem.

The underlying problem occurs during TCP connection processing, particularly when the kernel handles retransmitted SYN requests concurrently with socket creation and migration.

The vulnerability affects the handling of two internal structures: inet_opt and ireq_opt.

During TCP connection establishment, the function tcp_v4_syn_recv_sock() transfers ownership of ireq->ireq_opt to a newly created child socket without making a separate copy.

At the same time, another CPU can process a retransmitted SYN request for the same request socket. This creates a race in which the networking stack may continue accessing socket options while another execution path has already released the associated memory.

The problem is made worse because the relevant destruction routines use ordinary kfree() rather than deferring memory release through kfree_rcu().

As a result, memory can be freed while another CPU is still accessing it.

This is the core of the use-after-free condition.

What makes the vulnerability dangerous?

The concern is not simply that the kernel may encounter an invalid memory reference. The timing of the race creates an opportunity for memory corruption.

Study Security

Depending on the circumstances, successful exploitation could result in:

  • Kernel crash and denial of service.
  • Kernel memory disclosure.
  • Memory corruption.
  • Potential arbitrary code execution with kernel-level privileges.

The vulnerability is particularly relevant to systems that expose IPv4 TCP services to untrusted networks.

AWS has also reported that researchers demonstrated unauthenticated remote code execution against hosts with a Multipath TCP (MPTCP) listener. This is an important distinction: the demonstrated RCE scenario should not be interpreted as proof that every vulnerable Linux system is equally exploitable.

Audit Security

Affected Linux kernel versions

The vulnerability was introduced in Linux kernel version 4.4 through the following commit:

079096f103faca2dd87342cca6f23d4b34da8871

The Linux kernel security announcement identifies the following upstream fixed releases:

  • Linux kernel 6.12 branch: 6.12.113.
  • Linux kernel 6.18 branch: 6.18.56.
  • Linux kernel 7.2 branch: 7.2.10.
  • Linux kernel 7.3 branch: 7.3-rc7.

The fix has also been backported to supported stable kernel branches. Distribution vendors may provide their own patched package versions, so administrators should not rely exclusively on upstream version numbers when determining exposure.

Compare Security

The affected source files include:

  • net/ipv4/af_inet.c
  • net/ipv4/tcp_ipv4.c

The initial vulnerability concerns IPv4. The Linux kernel announcement explicitly states that IPv6 is not affected by the same first issue because its TCP connection path duplicates the options rather than transferring ownership in the same way.

CVSS severity and vendor assessments

Red Hat has classified CVE-2026-98432 as Critical, with a CVSS v3 score of 9.8.

Study Security

Amazon Linux has assigned a CVSS v3 score of 8.1 in its advisory.

These are vendor-specific assessments, and the difference should be retained rather than presenting one score as universally applicable.

The important operational point is that both vendors recognise the security implications of the underlying kernel flaw.

Vendor response and remediation

Red Hat

Red Hat published security bulletin RHSB-2026-016 on October 30, 2026, identifying the vulnerability as Critical.

The advisory covers potentially affected Red Hat Enterprise Linux products and provides an interim mitigation for systems that cannot immediately receive the kernel update.

Amazon Web Services

AWS published security bulletin 2026-160-AWS on October 30, 2026.

Compare Security

Amazon Linux 2 and Amazon Linux 2023 kernel updates are available through the corresponding security advisories.

AWS recommends updating the kernel package and rebooting affected instances to activate the fix.

For Amazon EKS, updated EKS-optimized Amazon Linux 2023 AMIs are also available for supported Kubernetes versions. Customers should review their node groups and update their worker nodes to patched AMIs.

AWS has also clarified that Lambda workloads are not directly affected because Lambda functions do not expose publicly reachable IPv4 TCP ports that would provide the required attack path.

Mitigation and recommended actions

Patching remains the primary remediation.

Organisations should identify Linux systems running potentially affected kernel builds, with particular attention to:

  • Internet-facing Linux servers.
  • Cloud workloads accepting inbound IPv4 TCP connections.
  • Container hosts and Kubernetes worker nodes.
  • Systems running Multipath TCP listeners.
  • Critical infrastructure relying on affected Linux kernel branches.

Where an immediate kernel update is not possible, temporary mitigations can reduce exposure.

Red Hat recommends filtering incoming IPv4 packets containing IP options using an appropriate firewall rule. AWS has also documented firewall-based mitigation and recommends disabling Multipath TCP as an interim risk-reduction measure where operationally appropriate.

These measures reduce exposure but do not replace the kernel  security update.

Compare Security

After installing the vendor-provided kernel package, administrators should reboot the system or follow the vendor’s supported live-patching procedure, where available. The running kernel must be verified to ensure that the fix is actually active.

Is CVE-2026-98432 being exploited in the wild?

This is an important distinction when assessing the vulnerability.

Study Security

AWS has explicitly stated that it is not aware of exploitation of CVE-2026-98432 in the wild in its November 1 bulletin.

Researchers have demonstrated remote code execution under specific conditions involving an MPTCP listener. However, a demonstrated proof of concept and confirmed exploitation in real-world attacks are two different things.

Organisations should therefore treat this as a critical vulnerability requiring prompt remediation, without incorrectly describing it as an actively exploited zero-day.

Referenced Vulnerabilities

CVE-2026-98432 CVSS 9.8 CRITICAL Received
CVE-2026-98432 - ipv4: free inet_opt and ireq_opt after an RCU grace period
In the Linux kernel, the following vulnerability has been resolved: ipv4: free inet_opt and ireq_opt after an RCU grace period tcp_v4_syn_recv_sock() transfers ownership of ireq->ireq_opt to the child socket (newinet->inet_opt) without copying it. Another cpu can concurrently process a retransmitted SYN for the same request socket, and send a SYNACK from tcp_check_req(). tcp_v4_send_synack() and inet_csk_route_req() read ireq->ireq_opt under rcu_read_lock() only, and ip_build_and_send_pkt() and ip_options_build() then read opt->optlen twice. Note that the SYNACK timer itself is not an issue: it holds a reference on its request socket, and inet_csk_reqsk_queue_drop() calls timer_delete_sync() before the child can be freed. Since commit 079096f103fa ("tcp/dccp: install syn_recv requests into ehash table"), request sockets are processed without holding the listener lock, so nothing prevents the child socket from being freed while the SYNACK is still being built. TCP child sockets do not have SOCK_RCU_FREE, and inet_sock_destruct() frees inet_opt with a plain kfree(), leading to a use-after-free in ip_options_build(). A similar issue exists with request socket migration (net.ipv4.tcp_migrate_req=1, or a BPF_SK_REUSEPORT_SELECT_OR_MIGRATE program). reqsk_timer_handler() clones the request socket with inet_reqsk_clone(), so that the old request socket and its clone share the same ireq_opt, then reqsk_migrate_reset() clears the pointer in the old one. Another cpu holding a reference on the old request socket can still be using these options (sending a SYNACK, or creating a child in tcp_v4_syn_recv_sock()) when the clone is freed, and tcp_v4_reqsk_destructor() also uses a plain kfree(). Readers already use RCU, and other paths replacing inet_opt (do_ip_setsockopt(), cipso_v4_sock_setattr()...) already use kfree_rcu(). Use kfree_rcu() in inet_sock_destruct() and tcp_v4_reqsk_destructor() as well. IPv6 is not affected by the first issue, because tcp_v6_syn_recv_sock() duplicates the options. tcp_v6_reqsk_destructor() has the same migration issue with ipv6_opt, which is only set by CALIPSO. This will be addressed in a separate patch.