GHSA-79rm-9vfm-ggww
Attack ChainAIDerived
Hypothetical attack flow derived from CVE metadata – not a confirmed exploit chain
-
AccessLDAP group resolver enabled on YuniKorn
-
DeliveryPod submitted referencing affected user
-
ExploitResolver reads non-CN memberOf entry
-
ExecutionOut-of-bounds read panics scheduler
-
ImpactScheduler crash denies cluster scheduling
Vulnerability Assessment
| Weakness |
CWE
OWASP Top 10:2025
A10:2025 - Mishandling of Exceptional Conditions
ASVS 5.0
V16 - Security Logging and Error Handling Exception handling completeness
|
| Exploitation Conditions | Exploitation requires that the instance be configured with the non-default LDAP group resolver (the group membership / memberOf resolution feature) - default installations are not affected. The crash is triggered specifically when the LDAP server returns a memberOf group-membership entry for a user named in a submitted pod where that entry does not start with the 'CN=' prefix. Limiting factors: the vulnerable code path only runs for the optional LDAP provider, the attacker must cause a pod referencing an affected user to be processed, and triggering it requires a malformed/non-CN LDAP record that is typically governed by directory contents rather than attacker-controlled input, which is why the vendor rates it AC:H/AT:P. Impact is confined to availability (scheduler crash); no data exposure or modification occurs. |
| Impact | This is a genuinely low-priority issue, and the signals are consistent rather than conflicting. The only authoritative scoring provided is a CVSS 4.0 base of 2.1 (low), with vulnerable-system impact limited to availability (VC:N/VI:N/VA:L) - i.e. a denial-of-service crash of the YuniKorn scheduler, with no confidentiality or integrity impact. The vector's AC:H and AT:P correctly reflect that exploitation is not straightforward: it depends on a non-default LDAP group resolver being configured AND the LDAP directory returning a malformed memberOf entry (one not starting with 'CN='). No EPSS score, no KEV listing, and no public exploit code are present in the provided data, so there is no public exploit identified at time of analysis and no evidence of active exploitation. The practical risk is therefore an operational robustness/availability bug that is far more likely to be triggered accidentally by real-world LDAP data than deliberately weaponized. Note the input provides a CVSS 4.0 vector but no CVSS 3.1 vector, so cross-version comparison relies on an independent assessment. |
| Exploit ScenarioDerived | In a cluster where YuniKorn is configured with the optional LDAP group resolver, a user referenced in a submitted pod has an LDAP memberOf group-membership entry that does not begin with 'CN=' (for example a DN using a lowercase or different attribute prefix). When YuniKorn resolves that user's groups, an out-of-bounds read occurs and the scheduler process crashes, denying scheduling to the whole cluster. No public exploit code is identified at time of analysis, and the condition is as likely to arise from ordinary directory data as from a deliberate attacker. |
Get an e-mail when this CVE is exploited, patched or re-scored. Free account.
Alert meAnalysisAIv1 – Oct 07, 2026
An out-of-bounds read in Apache YuniKorn's non-default LDAP group resolver allows a malformed LDAP group-membership entry to panic the scheduler process in versions 1.8.0 through 1.9.x, making the scheduler unavailable until it is restarted. The crash occurs only when an installation has explicitly configured the LDAP group provider and the LDAP server returns a memberOf entry for a user referenced in a submitted pod that does not begin with the 'CN=' prefix; default installations that use the built-in group resolution are not affected, and successful triggering generally requires an authenticated, low-privileged user able to have a pod processed. This is a low-severity, availability-only defect - CVSS 4.0 base score 2.1 (low) and an independently assessed CVSS 3.1 vector of AV:N/AC:H/PR:L/UI:N/S:U/C:N/I:N/A:L - with no public exploit identified at time of analysis and no confirmed active exploitation; a vendor fix is available in 1.10.0.
RemediationAI
Vendor-released patch: 1.10.0 - upgrade Apache YuniKorn to 1.10.0 (or later), which the vendor states fixes the out-of-bounds read in the LDAP group resolver; the fix is tracked as YUNIKORN-3432 with GitHub PR #1139 and details at https://issues.apache.org/jira/browse/YUNIKORN-3432 and https://yunikorn.apache.org. If an upgrade cannot be scheduled immediately, the most effective compensating control is to disable the non-default LDAP group provider and revert to the built-in group resolver, which removes the vulnerable parsing path entirely at the cost of losing LDAP-backed group-to-queue mapping until the upgrade lands. A secondary option is to sanitize the directory so that every memberOf value returned for users referenced by pods is a well-formed record beginning with 'CN=', which prevents the malformed prefix from reaching the parser, but this is brittle because it depends on directory administrators maintaining that invariant and on the scheduler not processing legacy or externally sourced entries. Because exploitation requires a submitted pod referencing a user whose LDAP membership entry is malformed, tighten who can create pods and workloads on the cluster (the assessed vector assumes only low privileges, PR:L) and review your YuniKorn authentication/authorization configuration so untrusted tenants cannot drive pod submissions. Since the impact is a scheduler process crash rather than data loss or corruption, add process-health alerting and automatic restart for the YuniKorn scheduler to shorten outage windows; note that restarting only restores service and does not prevent recurrence, as the same LDAP entry will re-trigger the panic on the next affected pod.
Affected VersionsENISA
Affected Packages
Source: OSV (Open Source Vulnerabilities)
Affects your stack
Affects 1 package in euvd
Checking this CVE against your own stacks and servers is part of the Basic plan and up.
Technical ContextAI · show details
Apache YuniKorn is a resource scheduler for Kubernetes that can replace the default kube-scheduler, providing hierarchical queues and gang scheduling. To map Kubernetes users to YuniKorn queues and quotas, it supports a pluggable group-resolution layer; the affected code is the optional LDAP group provider, which queries a directory server and reads the memberOf attribute of user entries, parsing each membership record as a distinguished name under the assumption that it starts with the 'CN=' component. When a returned membership value does not carry that prefix, the parser's indexing/slicing of the string runs past the end of the buffer and triggers a Go out-of-bounds read that escalates to a runtime panic, terminating the scheduler component (JIRA component 'core - common', tracked as YUNIKORN-3432 and linked to the related UserGroupCache cleanup issue YUNIKORN-3416). The underlying weakness class is CWE-125 (out-of-bounds read) with an availability-only effect, which the vendor reflects in its AC:H/AT:P rating: the vulnerable path executes only in LDAP-provider configurations, the input originates from directory contents rather than being freely attacker-crafted, and no confidentiality or integrity impact results, only loss of availability while the scheduler is down. The issue was resolved on 27/Aug/26 with target and fix version 1.10.0 and a pull request (GitHub PR #1139) noted in the issue labels.
Know when the next one hits your stack
- Watch products and get an email when something new is published
- Check this CVE against your own stack and servers
- Exploit and PoC references, threat intelligence from 22 sources
No credit card · 7-day full trial
Severity by source
CVSS:4.0/AV:N/AC:H/AT:P/PR:L/UI:P/VC:N/VI:N/VA:L/SC:N/SI:N/SA:L/S:N/AU:N/R:A/V:D/RE:H/U:Green
Network-facing scheduler but AC:H (non-default LDAP provider plus malformed non-CN entry) and PR:L (requires processing a submitted pod); impact is availability-only crash, so C:N/I:N/A:L.
Primary rating from Vendor (CNA).
CVSS VectorVendor
Lifecycle Timeline
3Description PRE-NVD
Executive ViewAI
Apache YuniKorn, the resource scheduler used in some Kubernetes environments, can be crashed and left unavailable in deployments that use its non-default LDAP group resolver, requiring a manual restart to restore scheduling. Only organizations that have explicitly enabled that LDAP provider are exposed, and triggering it requires an authenticated low-privileged user who can get a pod processed, so the practical risk is a repeatable scheduling outage rather than data compromise. Leadership should care because a sustained scheduler outage stalls workload placement across the affected cluster, delaying deployments and service scaling.
Apply vendor patch.
Threat Intelligence Overview
How this decision was reached
The engine reads each signal below and applies a fixed list of rules from top to bottom. The first rule that matches sets the recommended action – like a statement, every line is shown.
What every action level means
SSVC AssessmentENISA
Priority Score
Attack Technique
A buffer overflow occurs when a program writes more data to a memory buffer than it was allocated to hold, causing the excess data to spill into adjacent memory regions.
EUVD-2026-94255
NVD
CVE.org
EPSS
Vulnrichment
Exploit-DB