RHSA-2026:98239HighCVSS 8.8

Red Hat Security Advisory: kernel security, bug fix, and enhancement update

Published
November 11, 2026
Last Modified
November 12, 2026

🔗 CVE IDs covered (32)

CVE-2026-63829 →CVE-2026-63884 →CVE-2026-63889 →CVE-2026-45878 →CVE-2026-52912 →CVE-2026-68159 →CVE-2026-53329 →CVE-2026-74669 →CVE-2026-89972 →CVE-2026-74569 →CVE-2026-80714 →CVE-2026-53356 →CVE-2026-53374 →CVE-2026-63994 →CVE-2026-64219 →CVE-2026-68155 →CVE-2026-90227 →CVE-2026-43206 →CVE-2026-53254 →CVE-2026-63992 →CVE-2026-53136 →CVE-2026-89481 →CVE-2026-46259 →CVE-2026-53143 →CVE-2026-68432 →CVE-2026-72052 →CVE-2026-72255 →CVE-2026-63879 →CVE-2026-68156 →CVE-2026-68157 →CVE-2026-68476 →CVE-2026-97417 →

📋 Description

CVE-2026-43206 — kernel: drm/amdkfd: Fix out-of-bounds write in kfd_event_page_set() CVE-2026-45878 — kernel: drm/amdkfd: Fix watch_id bounds checking in debug address watch v2 CVE-2026-46259 — kernel: procfs: fix missing RCU protection when reading real_parent in do_task_stat() CVE-2026-52912 — kernel: netfilter: nf_queue: hold bridge skb->dev while queued CVE-2026-53136 — kernel: drm/amd/display: Clamp VBIOS HDMI retimer register count to array size CVE-2026-53143 — kernel: drm/amdkfd: Fix buffer overflow in SDMA queue checkpoint/restore on GFX11 CVE-2026-53254 — kernel: Bluetooth: RFCOMM: validate skb length in MCC handlers CVE-2026-53329 — kernel: drm/amd/display: Use krealloc_array() in dal_vector_reserve() CVE-2026-53356 — kernel: drm/i915/gem: Fix phys BO pread/pwrite with offset CVE-2026-53374 — kernel: drm/amdgpu: zero-initialize GART table on allocation CVE-2026-63829 — kernel: net: ip_gre: require CAP_NET_ADMIN in the device netns for changelink CVE-2026-63879 — kernel: drm/amdgpu: fix amdgpu_hmm_range_get_pages CVE-2026-63884 — kernel: drm/i915: Fix potential UAF in TTM object purge CVE-2026-63889 — kernel: scsi: scsi_transport_fc: Widen FPIN pname walker counter to u32 CVE-2026-63992 — kernel: tunnels: do not assume transport header in iptunnel_pmtud_check_icmp() CVE-2026-63994 — kernel: tunnels: load network headers after skb_cow() in iptunnel_pmtud_build_icmpv6 CVE-2026-64219 — kernel: drm/amd/display: Validate payload length and link_index in dc_process_dmub_aux_transfer_async CVE-2026-68155 — kernel: Linux kernel (libceph): Denial of Service due to malformed monitor maps CVE-2026-68156 — kernel: libceph: refresh auth->authorizer_buf{,_len} after authorizer update CVE-2026-68157 — kernel: Linux kernel: libceph null pointer dereference leads to denial of service CVE-2026-68159 — kernel: Linux kernel: libceph stack out-of-bounds write via crafted OSDMap CVE-2026-68432 — kernel: vxlan: require CAP_NET_ADMIN in the device netns for changelink CVE-2026-68476 — kernel: ipvs: reload ip header after head reallocation CVE-2026-72052 — kernel: net: ip6_gre: require CAP_NET_ADMIN in the device netns for changelink CVE-2026-72255 — kernel: netfilter: nf_queue: pin bridge device while NFQUEUE holds fake dst CVE-2026-74569 — kernel: netfilter: nf_conntrack_sip: widen NAT rewrite delta to s32 in sip_help_tcp() CVE-2026-74669 — kernel: ipvs: clear IPv4 options after rebasing tunnel ICMP errors CVE-2026-80714 — kernel: ipvs: do not propagate one-packet flag to synced conns CVE-2026-89481 — kernel: nvme-tcp: fix host memory disclosure on R2T for a read command CVE-2026-89972 — kernel: nvme: add missing SRCU grace period in error path CVE-2026-90227 — kernel: nvme/ioctl: check SUBMIT_IO with nvme_cmd_allowed() CVE-2026-97417 — kernel: netfilter: nf_conntrack: use get_unaligned_be32() in tcp_sack()

🎯 Affected products200

  • Red Hat CodeReady Linux Builder EUS (v.9.6)
  • Red Hat Enterprise Linux AppStream EUS (v.9.6)
  • Red Hat Enterprise Linux BaseOS EUS (v.9.6)
  • Red Hat Enterprise Linux Real Time EUS (v.9.6)
  • Red Hat Enterprise Linux Real Time for NFV EUS (v.9.6)
  • kernel-0:5.14.0-570.155.1.el9_6.aarch64 as a component of Red Hat Enterprise Linux BaseOS EUS (v.9.6)
  • kernel-0:5.14.0-570.155.1.el9_6.ppc64le as a component of Red Hat Enterprise Linux BaseOS EUS (v.9.6)
  • kernel-0:5.14.0-570.155.1.el9_6.s390x as a component of Red Hat Enterprise Linux BaseOS EUS (v.9.6)
  • kernel-0:5.14.0-570.155.1.el9_6.src as a component of Red Hat Enterprise Linux BaseOS EUS (v.9.6)
  • kernel-0:5.14.0-570.155.1.el9_6.x86_64 as a component of Red Hat Enterprise Linux BaseOS EUS (v.9.6)
  • kernel-64k-0:5.14.0-570.155.1.el9_6.aarch64 as a component of Red Hat Enterprise Linux BaseOS EUS (v.9.6)
  • kernel-64k-core-0:5.14.0-570.155.1.el9_6.aarch64 as a component of Red Hat Enterprise Linux BaseOS EUS (v.9.6)
  • kernel-64k-debug-0:5.14.0-570.155.1.el9_6.aarch64 as a component of Red Hat Enterprise Linux BaseOS EUS (v.9.6)
  • kernel-64k-debug-core-0:5.14.0-570.155.1.el9_6.aarch64 as a component of Red Hat Enterprise Linux BaseOS EUS (v.9.6)
  • kernel-64k-debug-debuginfo-0:5.14.0-570.155.1.el9_6.aarch64 as a component of Red Hat CodeReady Linux Builder EUS (v.9.6)
  • kernel-64k-debug-debuginfo-0:5.14.0-570.155.1.el9_6.aarch64 as a component of Red Hat Enterprise Linux AppStream EUS (v.9.6)
  • kernel-64k-debug-debuginfo-0:5.14.0-570.155.1.el9_6.aarch64 as a component of Red Hat Enterprise Linux BaseOS EUS (v.9.6)
  • kernel-64k-debug-debuginfo-0:5.14.0-570.155.1.el9_6.aarch64 as a component of Red Hat Enterprise Linux Real Time EUS (v.9.6)
  • kernel-64k-debug-devel-0:5.14.0-570.155.1.el9_6.aarch64 as a component of Red Hat Enterprise Linux AppStream EUS (v.9.6)
  • kernel-64k-debug-devel-matched-0:5.14.0-570.155.1.el9_6.aarch64 as a component of Red Hat Enterprise Linux AppStream EUS (v.9.6)
  • kernel-64k-debug-modules-0:5.14.0-570.155.1.el9_6.aarch64 as a component of Red Hat Enterprise Linux BaseOS EUS (v.9.6)
  • kernel-64k-debug-modules-core-0:5.14.0-570.155.1.el9_6.aarch64 as a component of Red Hat Enterprise Linux BaseOS EUS (v.9.6)
  • kernel-64k-debug-modules-extra-0:5.14.0-570.155.1.el9_6.aarch64 as a component of Red Hat Enterprise Linux BaseOS EUS (v.9.6)
  • kernel-64k-debuginfo-0:5.14.0-570.155.1.el9_6.aarch64 as a component of Red Hat CodeReady Linux Builder EUS (v.9.6)
  • kernel-64k-debuginfo-0:5.14.0-570.155.1.el9_6.aarch64 as a component of Red Hat Enterprise Linux AppStream EUS (v.9.6)
  • kernel-64k-debuginfo-0:5.14.0-570.155.1.el9_6.aarch64 as a component of Red Hat Enterprise Linux BaseOS EUS (v.9.6)
  • kernel-64k-debuginfo-0:5.14.0-570.155.1.el9_6.aarch64 as a component of Red Hat Enterprise Linux Real Time EUS (v.9.6)
  • kernel-64k-devel-0:5.14.0-570.155.1.el9_6.aarch64 as a component of Red Hat Enterprise Linux AppStream EUS (v.9.6)
  • kernel-64k-devel-matched-0:5.14.0-570.155.1.el9_6.aarch64 as a component of Red Hat Enterprise Linux AppStream EUS (v.9.6)
  • kernel-64k-modules-0:5.14.0-570.155.1.el9_6.aarch64 as a component of Red Hat Enterprise Linux BaseOS EUS (v.9.6)
  • +170 more not shown

✅ Remediation

For details on how to apply this update, which includes the changes described in this advisory, refer to: https://access.redhat.com/articles/11258 The system must be rebooted for this update to take effect. Red Hat recommends treating all kernel errata as security-relevant. Given the kernel's fundamental role, any bug has a higher chance of impacting system security, even if that impact only becomes clear after a fix is published. Therefore, Red Hat prioritizes delivering fixes that improve our customers' overall security posture. Because of this proactive approach, a patch may be associated with a CVE assignment at a future date. Retroactive CVE assignments are always documented in the corresponding errata and on Red Hat's CVE pages. We strongly advise against delaying updates, as doing so may leave your system exposed when protections are already available. Workaround: To mitigate this issue, prevent the `amdkfd` kernel module from loading if it is not required. This can be achieved by blacklisting the module. 1. Create a new file `/etc/modprobe.d/blacklist-amdkfd.conf` with the following content: `blacklist amdkfd` 2. Regenerate the initramfs to ensure the blacklist is applied during boot: `sudo dracut -f -v` (for RHEL 8/9) `sudo mkinitrd -f -v /boot/initramfs-$(uname -r).img $(uname -r)` (for RHEL 7) 3. Reboot the system for the changes to take effect. Note: Blacklisting the `amdkfd` module may impact functionality that relies on the AMD KFD driver, such as certain AMD GPU compute features. Evaluate the necessity of this module before applying the mitigation. Workaround: To mitigate this vulnerability, prevent the amdgpu kernel module from loading if AMD GPU hardware is not required on the system. 1. Create a configuration file in `/etc/modprobe.d/` to blacklist the module: ```bash echo "blacklist amdgpu" > /etc/modprobe.d/amdgpu-blacklist.conf echo "install amdgpu /bin/true" >> /etc/modprobe.d/amdgpu-blacklist.conf ``` 2. If the module is already loaded, unload it or reboot the machine: ```bash rmmod amdgpu ``` Caveats: Disabling the `amdgpu` driver will prevent graphical output, video acceleration, and compute functionality on systems equipped with AMD Radeon graphics hardware. Warning: Unloading active kernel modules or applying configuration changes may require a full system restart to take effect if the driver is currently in use by the display server or running applications. Workaround: Mitigation for this issue is either not available or the currently available options do not meet the Red Hat Product Security criteria comprising ease of use and deployment, applicability to widespread installation base, or stability. Workaround: Mitigation for this issue is either not available or the currently available options don't meet the Red Hat Product Security criteria comprising ease of use and deployment, applicability to widespread installation base or stability. Workaround: To mitigate this vulnerability, prevent the `ceph` kernel module from loading if Ceph client functionality is not in use. Create a file `/etc/modprobe.d/ceph.conf` with `install ceph /bin/true`. If the module is currently loaded, unload it using `rmmod ceph`. A system reboot is required for the changes to take full effect and ensure the module is not reloaded. Unloading the `ceph` module will impact systems actively using Ceph client features. Workaround: If the system does not require the `libceph` module for Ceph cluster connectivity, prevent the `ceph` kernel module from loading by blacklisting it. Create a file `/etc/modprobe.d/ceph.conf` with the content `blacklist ceph`. After creating the file, regenerate the initramfs using `dracut -f -v` and reboot the system for the changes to take effect. If `libceph` is in use, restrict network access to Ceph monitors to trusted hosts through firewall rules to reduce the attack surface. Workaround: If IPv6 GRE or IP-ERSpan tunnels are not in use, their respective kernel modules (`ip6_gre` and `ip6erspan`) can be blacklisted to prevent exploitation. Create a file such as `/etc/modprobe.d/disable-ip6gre.conf` with the following content: ``` install ip6_gre /bin/true install ip6erspan /bin/true ``` After creating the file, regenerate the initramfs and reboot the system for the changes to take effect. This mitigation may impact network functionality if these tunnel types are actively used. Workaround: To mitigate this issue, prevent the `ip_vs` kernel module from loading. This can be achieved by creating a blacklist rule. 1. Unload the module if currently loaded: `sudo modprobe -r ip_vs` 2. Create a blacklist configuration file: `echo "blacklist ip_vs" | sudo tee /etc/modprobe.d/disable-ip_vs.conf` 3. Rebuild the initial RAM disk to ensure the module is not loaded at boot: `sudo dracut -f` A system reboot is required for the changes to take full effect. Disabling this module will prevent the use of IP Virtual Server (IPVS) functionality. Workaround: To mitigate this issue, prevent the `ip_vs` kernel module from loading. This can be achieved by creating a blacklist rule. Note that this may impact systems utilizing IPVS for load balancing. 1. Create a configuration file to blacklist the module: `echo "blacklist ip_vs" > /etc/modprobe.d/disable-ip_vs.conf` 2. Regenerate the initramfs to ensure the blacklist is applied at boot: `dracut -f -v` 3. Reboot the system for the changes to take effect: `reboot` Alternatively, if the module is currently loaded, it can be unloaded (if no dependencies exist): `rmmod ip_vs` However, a system reboot is recommended to ensure the module is not loaded on subsequent boots. Workaround: Prevent the nvme-tcp kernel module from loading if NVMe over TCP storage is not required on the system. 1. Add an entry to /etc/modprobe.d to prevent the module from loading: # echo "install nvme-tcp /bin/true" >> /etc/modprobe.d/disable-nvme-tcp.conf 2. If the module is currently loaded and not in use by any storage target, unload it: # modprobe -r nvme-tcp Caveats: Disabling this module prevents the system from connecting to NVMe over TCP block storage devices. Warning: Do not attempt to unload this module if active NVMe over TCP storage is mounted or in use, as doing so will cause I/O failure and potential data loss. If the module cannot be cleanly unloaded, a system restart is required for the change to take full effect. Additionally, restrict NVMe-oF connections to verified, trusted storage targets over isolated or encrypted network fabrics. Workaround: Systems that do not require stateful firewall, NAT, or connection tracking can eliminate exposure by unloading the nf_conntrack kernel module. On systems where conntrack is required, exposure can be reduced by applying NOTRACK rules in the raw table to bypass connection tracking on untrusted or internet-facing interfaces, ensuring the vulnerable TCP SACK parsing code path in tcp_sack() is not reached for attacker-controlled traffic.

🔗 References (35)