CVE-2026-15588: GLib GDBus: unbounded auth line enables local DoS

MEDIUMCISA: TRACK*
Published July 20, 2026
CISO Take

GLib's GDBus authentication code does not cap the length of lines read from a client, so an attacker can send an oversized stream and exhaust memory and CPU until the process crashes or hangs. The CVSS vector (AV:L/PR:L) makes this a local, low-privilege issue and not a remote one, despite the description's wording. Exploitation risk is low: EPSS is 0.48% (61st percentile), it is not in CISA KEV, no public exploit or Nuclei template exists, and CISA SSVC says TRACK*. The AI relevance is indirect. GLib ships in Red Hat base images for AI workloads, including the RHAIIS vLLM CUDA/ROCm images, the rhai base images (CPU, CUDA, ROCm, TPU, Spyre) and OpenShift CoreOS nodes. Apply the Red Hat advisories (for example RHSA-2026:39985) through normal patch cycles, prioritizing shared multi-tenant GPU hosts where untrusted local users or containers can reach a D-Bus socket.

Sources:NVDEPSSRed Hat Security AdvisoriesATLASCISA SSVC

What is the risk?

Low to medium overall. The base score is 5.3 with local access and low privileges required, with partial confidentiality, integrity and availability impact. Exploitation is unlikely in the near term (EPSS 0.48%, no KEV, no public exploit). The main exposure is availability on shared AI/ML hosts, such as multi-tenant GPU nodes or notebook servers, where a low-privileged user can degrade a D-Bus service. Internet exposure is not expected, because GDBus normally listens on local sockets. Confidence in this assessment is MEDIUM: the vendor advisories list affected products without version ranges, and the description's mention of remote attackers conflicts with the local vector.

How does the attack unfold?

Initial Access
A low-privileged local user or compromised container reaches a GDBus-enabled service over a local socket.
AML.T0049
Exploitation
The attacker sends an excessively long data line during GDBus authentication, which has no length limit.
Resource Exhaustion
The service buffers the input and consumes large amounts of memory and CPU until it crashes or hangs.
AML.T0029
Impact
Co-located AI inference or training workloads on the node are degraded or killed, causing an availability outage.
AML.T0029

What systems are affected?

PackageEcosystemVulnerable RangePatched
vLLMpip—No patch
96.9K93 dependentsPushed 3d ago26% patched~37d to patchFull package profile →
vLLMpip—No patch
96.9K93 dependentsPushed 3d ago26% patched~37d to patchFull package profile →
cert-manager/cert-manager-istio-csr-rhel9——No patch
cert-manager/cert-manager-operator-rhel9——No patch
cert-manager/cert-manager-trust-manager-rhel9——No patch
cert-manager/jetstack-cert-manager-acmesolver-rhel9——No patch
cert-manager/jetstack-cert-manager-rhel9——No patch
discovery/discovery-server-rhel9——No patch
discovery/discovery-ui-rhel9——No patch
glib2——No patch
glib2-main——No patch
mingw-glib2——No patch
openshift/ose-rhel-coreos-8——No patch
openshift/ose-rhel-coreos-9——No patch
rhai/base-image-cpu-rhel9——No patch
rhai/base-image-cuda-rhel9——No patch
rhai/base-image-rocm-rhel9——No patch
rhai/base-image-spyre-rhel9——No patch
rhai/base-image-tpu-rhel9——No patch
rhaiis/model-opt-cuda-rhel9——No patch
rhui5/cds-kubernetes-rhel9——No patch
rhui5/cds-kubernetes-tp-rhel9——No patch
rhui5/cds-rhel9——No patch
rhui5/haproxy-rhel9——No patch
rhui5/installer-rhel9——No patch
rhui5/installer-tp-rhel9——No patch
rhui5/rhua-rhel9——No patch
rhui5/rhua-tp-rhel9——No patch

How severe is it?

CVSS 3.1
5.3 / 10
EPSS
0.6%
chance of exploitation in 30 days
Higher than 48% of all CVEs
Exploitation Status
Exploit Available
Exploitation: MEDIUM
Sophistication
Trivial
Exploitation Confidence
medium
○CISA SSVC: Public PoC
Composite signal derived from CISA KEV, VulnCheck KEV, CISA SSVC, EPSS, Metasploit, Exploit-DB, trickest/cve, Nuclei templates, and inthewild.io exploitation reports.

What is the attack surface?

AVACPRUISCIA
AVLocal
ACLow
PRLow
UINone
SUnchanged
CLow
ILow
ALow

What should I do?

1 step
  1. 1) Apply the Red Hat errata (RHSA-2026:39985, 40485, 42329, 55440, 57015, 58981, 61766, 61783, 63135, 63138) and update glib2 on hosts. 2) Rebuild and redeploy AI/ML container images from the patched RHEL 9 base images, including the vLLM and rhai base images. 3) Update OpenShift nodes through the normal machine config and CoreOS update path. 4) Set memory and CPU limits (cgroups, Kubernetes requests/limits) on inference and notebook workloads so one runaway process cannot take down a node. 5) Restrict access to D-Bus sockets and avoid exposing them to untrusted users or containers. 6) Detect abuse by monitoring for abnormal memory growth or CPU spikes in dbus-daemon and GDBus-using processes, plus OOM-killer events on GPU nodes. 7) Scan image inventories for the affected glib2 builds.

What does CISA's SSVC say?

DecisionTrack*
Exploitationpoc
AutomatableNo
Technical Impactpartial

Source: CISA Vulnrichment (SSVC v2.0). Decision based on the CISA Coordinator decision tree.

How is it classified?

Which compliance frameworks are affected?

This CVE is relevant to:

ISO 42001
A.6.2.6-AI system operation and monitoring
NIST AI RMF
MANAGE 2.3-Responses to AI risks are managed and documented
OWASP LLM Top 10
LLM03-Supply Chain

Frequently Asked Questions

What is CVE-2026-15588?

GLib's GDBus authentication code does not cap the length of lines read from a client, so an attacker can send an oversized stream and exhaust memory and CPU until the process crashes or hangs. The CVSS vector (AV:L/PR:L) makes this a local, low-privilege issue and not a remote one, despite the description's wording. Exploitation risk is low: EPSS is 0.48% (61st percentile), it is not in CISA KEV, no public exploit or Nuclei template exists, and CISA SSVC says TRACK*. The AI relevance is indirect. GLib ships in Red Hat base images for AI workloads, including the RHAIIS vLLM CUDA/ROCm images, the rhai base images (CPU, CUDA, ROCm, TPU, Spyre) and OpenShift CoreOS nodes. Apply the Red Hat advisories (for example RHSA-2026:39985) through normal patch cycles, prioritizing shared multi-tenant GPU hosts where untrusted local users or containers can reach a D-Bus socket.

Is CVE-2026-15588 actively exploited?

No confirmed active exploitation of CVE-2026-15588 has been reported, but organizations should still patch proactively.

How to fix CVE-2026-15588?

1) Apply the Red Hat errata (RHSA-2026:39985, 40485, 42329, 55440, 57015, 58981, 61766, 61783, 63135, 63138) and update glib2 on hosts. 2) Rebuild and redeploy AI/ML container images from the patched RHEL 9 base images, including the vLLM and rhai base images. 3) Update OpenShift nodes through the normal machine config and CoreOS update path. 4) Set memory and CPU limits (cgroups, Kubernetes requests/limits) on inference and notebook workloads so one runaway process cannot take down a node. 5) Restrict access to D-Bus sockets and avoid exposing them to untrusted users or containers. 6) Detect abuse by monitoring for abnormal memory growth or CPU spikes in dbus-daemon and GDBus-using processes, plus OOM-killer events on GPU nodes. 7) Scan image inventories for the affected glib2 builds.

What systems are affected by CVE-2026-15588?

This vulnerability affects the following AI/ML architecture patterns: model serving, LLM inference containers, GPU base images, Kubernetes/OpenShift AI platforms, training pipelines.

What is the CVSS score for CVE-2026-15588?

CVE-2026-15588 has a CVSS v3.1 base score of 5.3 (MEDIUM). The EPSS exploitation probability is 0.58%.

What is the AI security impact?

Affected AI Architectures

model servingLLM inference containersGPU base imagesKubernetes/OpenShift AI platformstraining pipelines

MITRE ATLAS Techniques

AML.T0010.001AI Software
AML.T0029Denial of AI Service

Compliance Controls Affected

ISO 42001: A.6.2.6
NIST AI RMF: MANAGE 2.3
OWASP LLM Top 10: LLM03

What are the technical details?

Original Advisory

A denial-of-service and resource exhaustion vulnerability exists within the `GDBus` component of GLib. The `gdbusauth` authentication mechanism fails to enforce proper length limitations on data lines read from a client. An unauthenticated local or remote attacker can exploit this lack of input validation by sending excessively long streams of data, causing the application to consume massive amounts of system memory and CPU, potentially leading to a crash or system hang.

Exploitation Scenario

A low-privileged user on a shared GPU host, or a compromised container with access to a D-Bus socket, connects to a GDBus-based service and sends an endless or oversized line during the authentication handshake. Because the line length is not limited, the service buffers it until it exhausts memory or burns CPU. The service crashes or hangs, and co-located vLLM or other inference workloads suffer degraded performance or are killed by the OOM killer. This results in an availability outage with no privilege escalation or data theft.

Weaknesses (CWE)

CWE-770 — Allocation of Resources Without Limits or Throttling: The product allocates a reusable resource or group of resources on behalf of an actor without imposing any intended restrictions on the size or number of resources that can be allocated.

  • [Requirements] Clearly specify the minimum and maximum expectations for capabilities, and dictate which behaviors are acceptable when resource allocation reaches limits.
  • [Architecture and Design] Limit the amount of resources that are accessible to unprivileged users. Set per-user limits for resources. Allow the system administrator to define these limits. Be careful to avoid CWE-410.

Source: MITRE CWE corpus.

CVSS Vector

CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:L/I:L/A:L

References

Timeline

Published
July 20, 2026
Last Modified
November 6, 2026
First Seen
September 30, 2026

Related Vulnerabilities