CVE-2026-58013: GLib: buffer over-read in line reader causes DoS

MEDIUM
Published June 30, 2026
CISO Take

GLib's g_io_channel_read_line_backend() reads past the end of a buffer when a custom line terminator longer than one character is configured. The result is a leak of up to 7 bytes of adjacent memory, or a crash if the read crosses a page boundary. This is a transitive, base-layer flaw. It surfaces in AI infrastructure through Red Hat container images, including RHAIIS vLLM (CUDA and ROCm), the model-opt image and the rhai base images, rather than through AI code itself. Exploitation risk is low: EPSS is 0.85% (top 43%), there is no public exploit or Nuclei template, it is not in CISA KEV, and SSVC says TRACK. The practical exposure depends on whether any network-facing service in the image uses GLib's GIOChannel with a custom multi-character terminator, which is uncommon. Apply the Red Hat errata (RHSA-2026:49512 and later) through your normal image rebuild cycle, and don't treat this as an emergency.

Sources:NVDEPSSRed Hat Security Advisories (access.redhat.com)ATLAS

What is the risk?

Low to medium overall. The CVSS 6.5 score assumes a network-reachable path, but it only applies when an application uses GIOChannel with a custom line terminator longer than one byte, which few do. Confidentiality impact is limited to 7 bytes of adjacent heap data. Availability impact needs a page-boundary crossing, so it is unreliable. EPSS is low, there is no public exploit and it is not in KEV. In AI/ML deployments the risk is further diluted because GLib is a transitive OS-level dependency of container base images, not a component of the model-serving code path.

How does the attack unfold?

Initial Access
Attacker sends crafted input to a network-facing service in an AI container that uses GLib GIOChannel with a multi-character line terminator.
AML.T0049
Exploitation
The line-reading backend runs memcmp past the GString buffer, reading beyond its allocated bounds.
Impact
About 7 bytes of adjacent memory leak, or the process crashes at a page boundary and disrupts the inference service.
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 →
GLib——No patch
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
mingw-glib2——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
6.5 / 10
EPSS
0.7%
chance of exploitation in 30 days
Higher than 54% of all CVEs
Exploitation Status
No known exploitation
Sophistication
Moderate

What is the attack surface?

AVACPRUISCIA
AVNetwork
ACLow
PRNone
UINone
SUnchanged
CLow
INone
ALow

What should I do?

1 step
  1. 1) Apply the relevant Red Hat errata (RHSA-2026:49512, 55440, 57015, 58981, 61766, 61783, 63135, 63138, 63140, 65762) and update glib2 in hosts and images. 2) Rebuild and redeploy AI inference and base images (rhaiis/vllm-*, rhai/base-image-*) from the patched RHEL 9 layers, then re-scan registries. 3) Inventory services that use GIOChannel with custom line terminators (g_io_channel_set_line_term). 4) Monitor inference pods for unexplained crashes or segfaults in libglib. 5) Prioritize internet-exposed workloads first and the rest through normal patch cadence.

What does CISA's SSVC say?

DecisionTrack
Exploitationnone
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-Response to previously unknown or unanticipated risks
OWASP LLM Top 10
LLM03-Supply Chain

Frequently Asked Questions

What is CVE-2026-58013?

GLib's g_io_channel_read_line_backend() reads past the end of a buffer when a custom line terminator longer than one character is configured. The result is a leak of up to 7 bytes of adjacent memory, or a crash if the read crosses a page boundary. This is a transitive, base-layer flaw. It surfaces in AI infrastructure through Red Hat container images, including RHAIIS vLLM (CUDA and ROCm), the model-opt image and the rhai base images, rather than through AI code itself. Exploitation risk is low: EPSS is 0.85% (top 43%), there is no public exploit or Nuclei template, it is not in CISA KEV, and SSVC says TRACK. The practical exposure depends on whether any network-facing service in the image uses GLib's GIOChannel with a custom multi-character terminator, which is uncommon. Apply the Red Hat errata (RHSA-2026:49512 and later) through your normal image rebuild cycle, and don't treat this as an emergency.

Is CVE-2026-58013 actively exploited?

No confirmed active exploitation of CVE-2026-58013 has been reported, but organizations should still patch proactively.

How to fix CVE-2026-58013?

1) Apply the relevant Red Hat errata (RHSA-2026:49512, 55440, 57015, 58981, 61766, 61783, 63135, 63138, 63140, 65762) and update glib2 in hosts and images. 2) Rebuild and redeploy AI inference and base images (rhaiis/vllm-*, rhai/base-image-*) from the patched RHEL 9 layers, then re-scan registries. 3) Inventory services that use GIOChannel with custom line terminators (g_io_channel_set_line_term). 4) Monitor inference pods for unexplained crashes or segfaults in libglib. 5) Prioritize internet-exposed workloads first and the rest through normal patch cadence.

What systems are affected by CVE-2026-58013?

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

What is the CVSS score for CVE-2026-58013?

CVE-2026-58013 has a CVSS v3.1 base score of 6.5 (MEDIUM). The EPSS exploitation probability is 0.71%.

What is the AI security impact?

Affected AI Architectures

model servingLLM inference containersGPU inference podscontainer base imagesKubernetes AI platforms

MITRE ATLAS Techniques

AML.T0029Denial of AI Service
AML.T0049Exploit Public-Facing Application

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 flaw was found in GLib. A buffer over-read can occur in g_io_channel_read_line_backend() in the giochannel.c file when a custom line terminator with a length greater than one is set, causing memcmp to read past the GString buffer. This vulnerability can cause a minor information disclosure of 7 bytes or a denial of service when the buffer over-read crosses a page boundary.

Exploitation Scenario

An attacker identifies an internet-facing service inside an AI inference container that uses GLib's GIOChannel with a multi-character line terminator to parse network input. They send crafted input that makes the terminator comparison run past the GString buffer. Most of the time this leaks about 7 bytes of adjacent heap memory. If the buffer sits at a page boundary, the process segfaults, taking down the inference worker and disrupting availability for downstream applications until it restarts.

Weaknesses (CWE)

CWE-126 — Buffer Over-read: The product reads from a buffer using buffer access mechanisms such as indexes or pointers that reference memory locations after the targeted buffer.

Source: MITRE CWE corpus.

CVSS Vector

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

References

Timeline

Published
June 30, 2026
Last Modified
November 10, 2026
First Seen
September 30, 2026

Related Vulnerabilities