cve.bz

CVE-2026-98068 linux linux

no score yetEPSS 0.2%Fix available

linux linux: no CVSS score yet. EPSS 0.2% estimated probability of exploitation activity in the next 30 days.

On this page
In the Linux kernel, the following vulnerability has been resolved: net/rds: don't let rds_conn_shutdown() consume a concurrent drop rds_conn_shutdown() finishes by moving the path from RDS_CONN_DISCONNECTING to RDS_CONN_DOWN, and also accepts RDS_CONN_ERROR as the starting state of that final transition, so that a FIN processed in softirq context during the teardown does not derail the shutdown into a noisy error path. But consuming that RDS_CONN_ERROR also consumes the shutdown pass that came with it: rds_conn_path_drop() sets RDS_CONN_ERROR and then queues cp_down_w, and a pass that starts on a path already in RDS_CONN_DOWN is a no-op. For the FIN case that is harmless - the socket the FIN arrived on is the very socket the teardown just released. It is not harmless for a dropper that attached something to the path first. rds_tcp_accept_one() is such a dropper. Its path claim in rds_tcp_accept_one_path() transitions RDS_CONN_DOWN -> RDS_CONN_CONNECTING, and a concurrent drop - a FIN on a previous socket in softirq context, an administrative reset - can put the path into RDS_CONN_ERROR between that claim and the state check that follows, which accepts RDS_CONN_ERROR. The accept then installs the freshly accepted socket with rds_tcp_set_callbacks() while the queued teardown - which sampled tc->t_sock before this socket existed - is still running. rds_connect_path_complete() fails its transition to RDS_CONN_UP and drops the path again, queueing the pass that should reap the socket it just installed. If the in-flight shutdown's final transition consumes that drop's RDS_CONN_ERROR, the queued pass finds the path in RDS_CONN_DOWN and does nothing. The installed socket is never torn down: it sits established with its callbacks armed and its rds_tcp_connection on rds_tcp_tc_list, the peer sees a connection that nothing ever reads, and the path is wedged in RDS_CONN_DOWN until some later event drops it again. Reproduced with widened race windows as an ever-growing receive queue on a socket owned by a path stuck in RDS_CONN_DOWN, with the peer's send path wedged behind it. Make the final transition only DISCONNECTING -> DOWN. If it fails because the path is in RDS_CONN_ERROR, a drop raced the teardown: cancel the reconnect timer and clear RDS_RECONNECT_PENDING - the one piece of the skipped tail that must not be left behind - and return, letting the pass the drop queued finish the job: it tears down whatever attached to the path in the meantime, completes the transition to RDS_CONN_DOWN, and re-arms the reconnect from its own tail. The timer quiesce in that branch matters because the racing drop does not always queue that pass: rds_conn_path_drop() returns without queueing when a destroy is pending - exactly the situation during a netns teardown or module unload, when a FIN on the dying socket is processed while rds_conn_path_destroy() flushes cp_down_w. If the flushed pass is the one that takes this return, no later pass exists, and rds_conn_path_destroy() would find cp_conn_w still armed (WARN_ON) and then free a path whose reconnect timer can still fire. With the cancel in the branch, every exit of a shutdown pass leaves the timer quiesced no matter which pass completes the transition. The FIN case keeps making progress, one pass later and still without noisy logging. Any other state keeps today's rds_conn_path_error() handling; no current cp_state writer can leave a DISCONNECTING path in anything but RDS_CONN_ERROR (every other writer is a cmpxchg from a non-DISCONNECTING state), so that branch is defensive. On kernels without the preceding patches the same hazard exists with the sample-based quiesce; the fix applies there equally.
EPSS
0.2% (unchanged from 0.2% a week ago; 0.2% 30 days ago) estimated probability of exploitation activity in the next 30 days; 7th percentile. Scored 2026-11-15. A forecast for this CVE across the internet, not for your own systems.
Published
Modified
NVD analysis
Under analysis
Record from
the NVD, synced

Fix and patched versions

  • Linux Linux: fixed in 5.10.271
  • Linux Linux: fixed in 5.15.222
  • Linux Linux: fixed in 6.1.189
  • Linux Linux: fixed in 6.6.158
  • Linux Linux: fixed in 6.12.111
  • Linux Linux: fixed in 6.18.53
  • Linux Linux: fixed in 7.2.7
  • Linux Linux: fixed in 7.3
  • Debian, Ubuntu, SUSE, openSUSE, Azure Linux, Alpaquita: fixed package versions per release

Ask an AI about this CVE

Show the promptExplain CVE-2026-98068 for an engineer deciding whether to act. Use https://cve.bz/api/public/cves/CVE-2026-98068?via=ask-ai-copy and https://cve.bz/cve/CVE-2026-98068?via=ask-ai-copy as the source of truth: what is affected and which versions fix it, whether it is in CISA KEV, public exploits, and its EPSS (an internet-wide 30-day estimate of exploitation activity, not a forecast for my systems). If they give no fixed versions, use the CVE record https://cveawg.mitre.org/api/cve/CVE-2026-98068 or the vendor advisory, not third-party mirrors. Tell me how to check whether I am affected and how to verify a fix. Cite the sources and say where they are uncertain.

Opens a third-party assistant; its answer is its own, not cve.bz's. The prompt holds only this CVE's id and public links.

Affected

Versions

From the CNA's record in the CVE List.

ProductAffected
Linux4.12 and later

Fixed in

ProductRelease lineFixed in
Linux5.105.10.271
Linux5.155.15.222
Linux6.16.1.189
Linux6.66.6.158
Linux6.126.12.111
Linux6.186.18.53
Linux7.27.2.7
Linuxlater releases7.3

Linux distributions

DistributionReleasePackageFixed inAdvisory
Debian12linux6.1.190-1DEBIAN-CVE-2026-98068
Debian12linux-6.126.12.111-1~deb12u1DEBIAN-CVE-2026-98068
Debian13linux6.12.111-1DEBIAN-CVE-2026-98068
Debian14linux7.2.7-1DEBIAN-CVE-2026-98068
Ubuntu16.04:LTSlinux-hwe-edgeno fix listedUBUNTU-CVE-2026-98068
Ubuntu18.04:LTSlinux-gcp, linux-hwe, linux-oem and 12 moreno fix listedUBUNTU-CVE-2026-98068
Ubuntu20.04:LTSlinux-gke, linux-gkeop, linux-riscv and 26 moreno fix listedUBUNTU-CVE-2026-98068
Ubuntu22.04:LTSlinux5.15.0-202.212UBUNTU-CVE-2026-98068 USN-8961-1
Ubuntu22.04:LTSlinux-aws5.15.0-1119.126UBUNTU-CVE-2026-98068 USN-8961-1
Ubuntu22.04:LTSlinux-aws-6.86.8.0-1069.74~22.04.1UBUNTU-CVE-2026-98068 USN-9024-1
Ubuntu22.04:LTSlinux-azure5.15.0-1125.134UBUNTU-CVE-2026-98068 USN-8961-1
Ubuntu22.04:LTSlinux-azure-6.86.8.0-1073.82~22.04.1UBUNTU-CVE-2026-98068 USN-9026-1
Ubuntu22.04:LTSlinux-azure-fde5.15.0-1124.133UBUNTU-CVE-2026-98068 USN-8962-1
Ubuntu22.04:LTSlinux-azure-fde-6.86.8.0-1071.78~22.04.1UBUNTU-CVE-2026-98068 USN-9026-1
Ubuntu22.04:LTSlinux-gcp5.15.0-1120.130UBUNTU-CVE-2026-98068 USN-8961-1
Ubuntu22.04:LTSlinux-gcp-6.86.8.0-1073.81~22.04.1UBUNTU-CVE-2026-98068 USN-9024-1
82 more for Ubuntu in the JSON and on Ubuntu's page.
SUSELinux Enterprise High Availability Extension 16.0kernel-default6.12.0-160000.41.1SUSE-SU-2026:24651-1
SUSELinux Enterprise Server 16.0, Linux Enterprise Server for SAP applications 16.0kernel-64kb, kernel-docs, kernel-syms and 6 more6.12.0-160000.41.1SUSE-SU-2026:24662-1
SUSELinux Enterprise Server 16.0, Linux Enterprise Server for SAP applications 16.0kernel-default-base6.12.0-160000.41.1.160000.2.32SUSE-SU-2026:24662-1
SUSELinux Enterprise Server for NVIDIA 16.1kernel-nvidia64kb, kernel-source-nvidia6.12.0-160100.7.1SUSE-SU-2026:24705-1
SUSELinux Micro 6.0kernel-default-base6.4.0-159998.55.1.159998.21.42SUSE-SU-2026:24550-1
SUSELinux Micro 6.0kernel-source, kernel-default, kernel-kvmsmall6.4.0-159998.55.1SUSE-SU-2026:24550-1
SUSELinux Micro 6.1kernel-default-base6.4.0-159998.55.1.159998.21.41SUSE-SU-2026:24619-1
SUSELinux Micro 6.1kernel-source, kernel-default, kernel-kvmsmall6.4.0-159998.55.1SUSE-SU-2026:24619-1
SUSELinux Micro 6.2kernel-default-base6.12.0-160000.41.1.160000.2.32SUSE-SU-2026:24643-1
SUSELinux Micro 6.2kernel-rt, kernel-64kb, kernel-source and 1 more6.12.0-160000.41.1SUSE-SU-2026:24643-1
SUSELinux Micro Extras 6.2kernel-syms, kernel-obs-build6.12.0-160000.41.1SUSE-SU-2026:24648-1
openSUSELeap 16.0kernel-default-base6.12.0-160000.41.1.160000.2.31openSUSE-SU-2026:22266-1
openSUSELeap 16.0kernel-rt, dtb-aarch64, kernel-64kb and 9 more6.12.0-160000.41.1openSUSE-SU-2026:22266-1
openSUSELeap 16.1kernel-default-base6.12.0-160100.9.1.160100.3.14openSUSE-SU-2026:22309-1
openSUSELeap 16.1kernel-nvidia64kb, kernel-syms-nvidia, kernel-source-nvidia6.12.0-160100.7.1openSUSE-SU-2026:22307-1
openSUSELeap 16.1kernel-rt, dtb-aarch64, kernel-64kb and 9 more6.12.0-160100.9.1openSUSE-SU-2026:22309-1
openSUSETumbleweedkernel-source7.2.9-1.1openSUSE-SU-2026:12074-1
Azure Linux3kernel6.6.159.1-1AZL-110485
Alpaquita23linux-lts6.1.189-r0BELL-CVE-2026-98068
Alpaquita25linux-lts6.12.111-r0BELL-CVE-2026-98068
Alpaquitastreamlinux-lts6.18.53-r0BELL-CVE-2026-98068

From the distributions' data via OSV.dev: the Debian Security Tracker; Ubuntu Security Notices © Canonical Ltd., CC BY-SA 4.0; SUSE security data, CC BY 4.0; Microsoft's Azure Linux vulnerability data, MIT; BellSoft's OSV database, MIT. Modified: reformatted and linked to CVE records. Credits. On the distributions' own pages: Debian · Ubuntu · SUSE · openSUSE · Alpaquita.

Red Hat

Red Hat rates this Moderate.

Red Hat's statement

A moderate-impact flaw was found in the Linux kernel's Reliable Datagram Sockets (RDS) subsystem. A race condition in rds_conn_shutdown() allows the final shutdown state transition to consume an RDS_CONN_ERROR state set by a concurrent connection drop (for example, a FIN processed in softirq context or an administrative reset) racing with teardown or with an incoming connection accepted by rds_tcp_accept_one(). When this happens, the cleanup pass queued by the drop becomes a no-op, so a newly accepted socket can remain established with its callbacks armed while the connection path is wedged in RDS_CONN_DOWN. The socket's receive queue then grows without bound and the peer's send path stalls behind it, resulting in a denial of service. During network namespace teardown or module unload, the same race can leave a reconnect timer armed on a connection path that is subsequently freed, potentially causing a kernel panic. Exploitation depends on winning a narrow timing window between concurrent connection state transitions, which is reflected in the moderate impact rating. No shipped Red Hat product is currently affected: Red Hat Enterprise Linux 7 through 10 (kernel and kernel-rt), Red Hat In-Vehicle OS 2, and OpenShift Container Platform 4 and 5 are marked not affected because the vulnerable code is not present in the shipped kernels, and Red Hat Enterprise Linux 6 Extended Life-cycle Support is out of support scope.

Red Hat's mitigation

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.

ProductPackageStateAdvisory
Red Hat Enterprise Linux 10kernelNot affected
Red Hat Enterprise Linux 6kernelOut of support scope
Red Hat Enterprise Linux 7kernelNot affected
Red Hat Enterprise Linux 7kernel-rtNot affected
Red Hat Enterprise Linux 8kernelNot affected
Red Hat Enterprise Linux 8kernel-rtNot affected
Red Hat Enterprise Linux 9kernelNot affected
Red Hat Enterprise Linux 9kernel-rtNot affected
Red Hat OpenShift Container Platform 4openshift/ose-rhel-coreos-8Not affected
Red Hat OpenShift Container Platform 4openshift/ose-rhel-coreos-9Not affected
Red Hat OpenShift Container Platform 5openshift/ose-rhel-coreos-9Not affected

Contains Red Hat Security Data © Red Hat, Inc., licensed under CC BY 4.0. Source: https://access.redhat.com/security/data. Modified: reformatted. Red Hat's page for CVE-2026-98068.

References

Check your stackJSON