Red Hat
Monthly
Stack buffer overflow in GStreamer's H.265/HEVC codec parser (gst-plugins-bad) allows remote unauthenticated attackers to crash GStreamer-based applications by delivering a crafted H.265 video file or stream that a user opens. The root cause is an incorrect loop bound in the buffering period SEI message parser: the parser mistakenly uses cpb_cnt_minus1[i] (the current loop index variable) rather than cpb_cnt_minus1[0] from the referenced Sequence Parameter Set, causing the loop to iterate beyond the bounds of stack-allocated CPB delay arrays and corrupt stack memory. No public exploit has been identified at time of analysis and the vulnerability is not listed in CISA KEV, though the deterministic parser logic makes crash reproduction straightforward.
Out-of-bounds write in GStreamer's H.266/VVC PPS picture partition parser (`gst-plugins-bad`) allows an attacker to crash media-processing applications - and potentially achieve code execution - by delivering a crafted H.266/VVC media file. The flaw in `gst_h266_parser_parse_picture_partition()` (gsth266parser.c) permits unbounded slice index increments across three fixed-size arrays in `GstH266PPS` during multi-slice-in-tile processing. A proof-of-concept demonstrating at least a 4-byte write exists; no public exploit beyond that initial POC or CISA KEV listing has been identified at time of analysis, though the code structure permits larger writes across multiple iterations which elevates downstream risk above a pure DoS assessment.
HTTP Response Splitting via CRLF injection in Apache CXF's OAuth2 module allows an attacker who controls the WWW-Authenticate realm parameter to inject arbitrary HTTP headers or split HTTP responses entirely. Affected deployments include cxf-rt-rs-security-oauth2 versions 4.2.0 before 4.2.2 and all versions before 4.1.7. No public exploit code has been identified at time of analysis and this vulnerability is not listed in CISA KEV, but successful exploitation could enable cache poisoning, header injection, or redirection of downstream HTTP clients processing the malformed response.
Log injection in Apache CXF's OAuth2 module (org.apache.cxf:cxf-rt-rs-security-oauth2) permits remote attackers to forge arbitrary log entries by supplying crafted `clientId` values containing control characters or newline sequences in OAuth2 HTTP requests. Affected are CXF 4.2.0-4.2.1 and all 4.1.x versions before 4.1.7; fixed releases 4.2.2 and 4.1.7 were issued June 10, 2026. No public exploit code or active exploitation has been identified at time of analysis; practical impact is confined to log integrity compromise that could mislead security monitoring and incident response processes.
Authentication bypass in Apache CXF's OAuth2 TokenIntrospectionService allows unauthenticated network access to the token introspection endpoint due to a missing 'throw' keyword in the internal security context check, causing the guard to silently pass rather than reject unauthorized callers. Affected are deployments using cxf-rt-rs-security-oauth2 versions 4.2.0-4.2.1 and all 4.1.x releases before 4.1.7 that relied solely on CXF's built-in check without independent authentication at the container or gateway layer. No public exploit code or active exploitation has been identified; vendor-confirmed patches (4.2.2 and 4.1.7) were released June 10, 2026.
Denial of service in the joi npm validation library (< 18.2.1) is triggered by submitting deeply nested JSON objects against schemas that use recursive `link()` definitions, causing a JavaScript call stack overflow. The untrapped `RangeError` propagates out of joi's link resolver - crashing the Node.js process entirely when `validate()` is invoked without a surrounding `try/catch`, or silently breaking error-handling logic when a `RangeError` is returned where a structured `ValidationError` was expected. No confirmed active exploitation (not in CISA KEV) and no public exploit code identified at time of analysis, but the attack surface is inherently wide given joi's prevalence in Node.js web services.
Host confusion in guzzlehttp/psr7 (all versions prior to 2.10.2) allows unauthenticated network attackers to supply a malformed Host header - such as `trusted.example@evil.example` - causing the library's URI construction logic to reinterpret the value as URI userinfo and a different host, silently replacing the parsed URI host with the attacker-controlled domain. Applications that rely on the resulting PSR-7 URI host for routing, allow-list enforcement, or forwarding decisions are at risk of sending requests and credentials to unintended destinations. No public exploit has been identified at time of analysis and no CISA KEV listing exists, but the conditional impact on forwarding gateways and API proxies built on psr7's server-request parsing functions is concrete.
Artifact pin decay in vLLM (< 0.22.0) allows pinned model deployments to silently load secondary artifacts - dynamic code modules, GGUF files, image processors, retrieval side weights, and same-repository subfolder configs - from unpinned or default HuggingFace Hub revisions, undermining supply-chain integrity guarantees. Operators who supply `--revision` or `--code-revision` to audit and lock a specific model version can unknowingly serve behavior-affecting components that fall outside the reviewed artifact set. No public exploit code has been identified at time of analysis, but seven discrete affected code paths across vllm's model loader and registry subsystems have been confirmed at commit 3795d7acf431980e62e738493f437ae2a51549da, with a fix released in vLLM 0.22.0 via PR #42616.
Cache key collision in OpenFGA's iterator caching mechanism allows two distinct authorization check requests to resolve to the same cache key, causing the engine to return a stale, incorrect authorization result to a subsequent requester. All OpenFGA deployments prior to version 1.16.0 with iterator caching enabled are affected, specifically when using the experimental weighted_graph_check union resolution path. An authenticated attacker who can trigger authorization checks under conditions that produce a colliding cache key may receive an incorrect allow or deny decision, undermining the integrity and confidentiality of access control enforced by the engine. No public exploit identified at time of analysis, and this vulnerability is not listed in the CISA KEV catalog.
Stored cross-site scripting in Red Hat's migration-planner-ui-app (kubev2v project) allows an authenticated attacker who can register a discovery agent to inject a javascript: URI into the agent's credentialUrl field, which executes in another organizational user's browser when they click the resulting link in the UI. No public exploit identified at time of analysis, but the upstream fix (PR #750) and Red Hat Bugzilla entry confirm the issue is real and patched via a safeExternalUrl protocol allowlist. Successful exploitation hijacks the victim's Red Hat SSO session and enables cross-tenant data access and API actions.
Unencrypted secret storage in Jenkins 2.567 and earlier (LTS 2.555.2 and earlier) exposes credentials submitted via POST config.xml to any user holding Item/Extended Read permission or with read access to the Jenkins controller filesystem. Secrets that should be encrypted at rest are written as plaintext into job config.xml files, making them directly readable through Jenkins' built-in permission model or OS-level file access. No public exploit code has been identified and this CVE is not listed in the CISA KEV catalog, but the privileged insider and lateral-movement risk is significant for organizations embedding CI/CD credentials in job configurations.
Open redirect in Jenkins 2.567 and earlier (LTS 2.555.2 and earlier) enables unauthenticated remote attackers to craft login URLs with a malicious external domain in the 'from' parameter under the 'Delegate to servlet container' security realm, silently redirecting authenticated users to attacker-controlled sites post-login for phishing purposes. The CVSS vector (PR:N, UI:R) confirms no attacker authentication is needed, but victim interaction is mandatory - the user must click a crafted link and complete a login. No public exploit identified at time of analysis and the vulnerability is not listed in CISA KEV; vendor-released patches are available as of 2026-06-10.
Missing authorization checks in Jenkins 2.567 and earlier (LTS 2.555.2 and earlier) expose user-specific configuration data to any authenticated user holding the Overall/Read permission. Authenticated low-privilege users can query the timezone setting of any Jenkins user account and enumerate the names of views configured within other users' 'My Views' personal dashboards, constituting an information disclosure weakness. No public exploit code exists and this CVE is not listed in the CISA KEV catalog at time of analysis, placing real-world risk in the low-moderate tier primarily relevant to multi-tenant or shared Jenkins deployments.
Missing authorization enforcement in Jenkins allows low-privileged authenticated users to cancel build queue items they lack permission to view. Affecting Jenkins 2.567 and earlier (LTS 2.555.2 and earlier), this CWE-862 flaw breaks the expected access control invariant that Item/Cancel should be constrained by Item/Read visibility. No public exploit code exists and this vulnerability is not listed in CISA KEV at time of analysis, but the low attack complexity and network accessibility make it a realistic threat in multi-tenant Jenkins deployments where fine-grained RBAC is in use.
Open redirect exploitation in Jenkins allows unauthenticated network attackers to craft login URLs that bypass the post-authentication redirect validation and forward users to attacker-controlled external sites. Affected are Jenkins weekly releases through 2.567 and LTS releases through 2.555.2, where the redirect URL legitimacy check fails when the URL contains relative path segments such as `./` or `../`. No active exploitation is confirmed (not listed in CISA KEV) and no public exploit code has been identified at time of analysis, placing this in a moderate-risk, phishing-enablement category rather than a direct system compromise class.
Denial of service in the image-size Node.js library through version 2.0.2 allows remote unauthenticated attackers to permanently hang the Node.js event loop by supplying a crafted JXL or HEIF image containing a box with a zero-valued size field. Publicly available exploit code exists and an upstream fix has been merged, making this a credible availability threat for any service that accepts user-supplied images and runs them through image-size. No active exploitation has been reported in CISA KEV at time of analysis.
Denial of service in the image-size Node.js library (versions up to and including 2.0.2) allows remote unauthenticated attackers to permanently stall the Node.js event loop by submitting a malformed ICNS image. The flaw stems from an infinite loop in the ICNS parser when an entry length field is zero, and publicly available exploit code exists per VulnCheck and an independent write-up; no public exploit identified at time of analysis indicates active exploitation, but the POC makes weaponization trivial.
Uncontrolled resource consumption in Spring Data Commons exposes applications to a remote Denial of Service condition when two specific features are active simultaneously. An unauthenticated remote attacker can send a specially crafted HTTP request to any endpoint backed by a @ProjectedPayload-annotated controller method, triggering unbounded memory allocation that exhausts the JVM heap and crashes the application. The vulnerability spans eight major release lines (2.7.x through 4.0.x), making the potential blast radius broad across Spring-based Java backends. No public exploit code or CISA KEV listing has been identified at time of analysis, and the CVSS attack complexity rating of High reflects non-trivial preconditions that meaningfully limit opportunistic exploitation.
SpEL (Spring Expression Language) injection in Spring Data KeyValue and Spring Data Redis allows a network-accessible, low-privileged attacker to execute arbitrary SpEL expressions when applications pass unsanitized user-controlled Sort parameters directly to repository query methods delegating to SpelPropertyComparator. Affected versions span eight major release lines from 2.7.x through 4.0.x, making the exposure surface broad across Spring-based Java ecosystems. No public exploit identified at time of analysis and no CISA KEV listing, but the high confidentiality impact rating and network attack vector warrant prompt patching for any application that surfaces Sort query parameters to end users.
Improper certificate validation in Spring AMQP allows a network-positioned attacker to perform a man-in-the-middle attack against applications that configure broker connections via RabbitConnectionFactoryBean.setUri("amqps://...") without explicitly calling setUseSSL(true). Affected are Spring AMQP versions 2.4.0 through 4.0.3 across four release lines. Although TLS encryption is established, the absence of certificate chain validation and hostname verification means a rogue broker can impersonate the legitimate RabbitMQ instance, exposing message content in transit. No public exploit identified at time of analysis, and this CVE is not listed in CISA KEV.
Denial of Service in Spring Data Commons affects applications that parse Sort parameters, triggering a StackOverflowException via crafted input. All major supported and legacy branches from 2.7.x through 4.0.x are affected, making the blast radius exceptionally broad across the Spring ecosystem. CVSS rates this Medium (5.9) with a high attack complexity qualifier (AC:H), unauthenticated network access (PR:N), and full availability impact (A:H); no active exploitation is confirmed and no public exploit code has been identified at time of analysis.
Predictable correlation IDs in Spring AMQP's RabbitTemplate.sendAndReceive() expose request-reply messaging to correlation hijacking when a fixed reply queue is configured. Affected are all widely used Spring AMQP branches from 2.4.x through 4.0.x - a broad install base across Java enterprise applications. A network-positioned attacker with high privileges can exploit the sequential counter to predict future correlation IDs, enabling interception or injection of reply messages into the shared fixed reply queue. No public exploit code exists and this vulnerability is not listed in CISA KEV; CVSS 4.4 Medium reflects real-world limitations from high privilege and complexity requirements despite the changed scope indicator.
Regex parameter binding in Spring Data MongoDB's @Query annotation fails to sufficiently quote user-supplied strings, enabling NoSQL injection that can manipulate query semantics. Unauthenticated remote attackers (PR:N, AV:N) who can influence the bound parameter in applications using this pattern may break out of the intended regular expression context to retrieve unintended documents, resulting in high-impact confidentiality compromise. No confirmed active exploitation or public exploit code has been identified at time of analysis; the CVSS Attack Complexity of High reflects a dependency on a specific application-level coding pattern.
Command injection in Ruby's net-imap gem allows arbitrary IMAP protocol command injection via the Net::IMAP#id and Net::IMAP#enable methods when untrusted input is passed as arguments. The #id method fails to prohibit CRLF sequences in ID field value strings (despite correctly quoting other specials), while #enable passes argument .to_s values verbatim without validating them as legal IMAP atoms. An attacker who can influence these inputs can terminate the current IMAP command and append arbitrary subsequent IMAP commands such as DELETE mailbox. Rated 'moderate' by the maintainer; no public exploit identified at time of analysis and no CISA KEV listing.
CRLF command injection in Ruby's net-imap gem (CVE-2026-47240) enables attackers to inject arbitrary IMAP commands when applications pass unvalidated user-controlled input to raw data arguments of search, fetch, sort, and thread methods. Affected versions (>= 0.6.0 through 0.6.4, and <= 0.5.14) fail to verify that the connected IMAP server supports non-synchronizing literals before sending them, meaning servers without LITERAL+, LITERAL-, or IMAP4rev2 capability may interpret the malformed literal boundary as a pipelined command rather than closing the connection - allowing injected payloads such as DELETE mailbox to execute. No public exploit has been identified at time of analysis and no CISA KEV listing exists, but the advisory is rated moderate severity by the upstream maintainers and patches are available.
Denial-of-service in the Linux kernel's DAMON (Data Access MONitor) subsystem allows a local low-privileged user to crash the kernel by supplying a non-power-of-two `min_region_sz` value through the DAMON sysfs interface and invoking `damon_start()`. The flaw is an incomplete remediation: commit c80f46ac228b fixed the same class of bug for the `damon_commit_ctx()` code path but left `damon_start()` - reachable via DAMON sysfs - without the `is_power_of_2()` guard. No public exploit exists and EPSS sits at 0.02% (4th percentile), indicating negligible real-world exploitation probability; no CISA KEV listing has been issued.
Memory leak in the Linux kernel's drm/nouveau subsystem exposes systems with NVIDIA GPUs using the open-source nouveau driver to local denial of service. When aperture_remove_conflicting_pci_devices() fails during PCI device probe, the error path bypasses the fail_nvkm cleanup label, permanently leaking both the nvkm_device wrapper and the pci_enable_device() reference taken inside nvkm_device_pci_new(). No public exploit exists and EPSS sits at the 5th percentile (0.02%), confirming negligible exploitation probability; however, patched stable releases are available across multiple kernel branches.
Improper end-of-filesystem boundary handling in the Linux kernel EROFS driver for file-backed mounts causes a denial-of-service condition when I/O requests exceed the filesystem image boundary. Affected kernels spanning multiple stable branches fail to zero-fill out-of-bounds I/O regions as expected (analogous to loopback device behavior), producing undefined kernel behavior that results in availability loss. No public exploit has been identified and EPSS is 0.02% (5th percentile), indicating negligible opportunistic exploitation probability; patches are confirmed available across stable branches 6.12.x, 6.18.x, 6.19.x, and 7.0.
Lock resource leak in Linux kernel hugetlbfs subsystem enables local denial of service against systems using huge page memory mappings. Introduced by a faulty commit (ea52cb24cd3f) that incorrectly handled VMA lock allocation during the mmap_prepare stage, the defect allows a failed post-mmap_prepare allocation to leave a VMA lock unreleased, leaking kernel resources. This CVE documents the corrective revert; systems running affected kernel versions (around 6.19 through pre-7.0.12 and pre-7.1-rc6) with hugetlbfs in use are exposed to potential availability impact. No public exploit code exists and EPSS probability is 0.02% (5th percentile), indicating very low real-world exploitation activity.
Double-free vulnerability in OpenSSL's OCSP stapled response verification allows remote attackers with low privilege to cause memory corruption affecting confidentiality, integrity, and availability. CVE-2026-35188 affects OpenSSL 4.0.0 (fixed in 4.0.1) and the 3.6.x branch (fixed in 3.6.3), and is one of seventeen CVEs addressed in the OpenSSL 4.0.1 security patch release published 2026-06-09. No public exploit and no active exploitation (CISA KEV) have been identified at time of analysis; however, the SSVC framework rates technical impact as 'total', flagging potential severity beyond the moderate CVSS base score.
Out-of-bounds read in OpenSSL 4.0.0's `X509_VERIFY_PARAM_set1_email()` function can crash applications performing email-based X.509 certificate verification when processing attacker-influenced email input, resulting in a denial-of-service condition. The vulnerability is scoped to OpenSSL 4.0.0 only and was patched in the June 9, 2026 security release (4.0.1), which bundled fixes for 18 CVEs. No public exploit identified at time of analysis and no CISA KEV listing.
Stale kernel memory disclosure in the Linux kernel's io_uring IORING_OP_WAITID operation allows a local low-privileged user to read arbitrary bytes from reused io_kiocb command storage via the userspace siginfo output buffer. The io_uring prep path for waitid requests fails to zero-initialize the io_waitid::info result field, meaning that when a wait completes without a child event the kernel copies whatever bytes happen to occupy that slab slot into the caller's siginfo_t. No public exploit has been identified at time of analysis; EPSS at 0.02% (5th percentile) reflects low current exploitation probability.
Server-side request forgery (SSRF) in Spring Framework's UriComponentsBuilder affects applications that use this API to parse and validate externally supplied URL strings. Incorrect host parsing allows a remote, unauthenticated attacker - with user interaction - to cause the application server to issue requests to unintended internal or external destinations, exposing low-level confidentiality and integrity impacts. No public exploit identified at time of analysis and no CISA KEV listing; however, SSRF in widely deployed Java frameworks warrants attention in any internet-facing application that processes user-controlled URLs.
Multipart request smuggling in Spring Framework's MVC and WebFlux components exposes applications to HTTP request manipulation via CWE-444. Unauthenticated remote attackers (AV:N/AC:L/PR:N/UI:N per CVSS) can exploit inconsistent multipart boundary parsing to smuggle malformed HTTP requests, achieving low-integrity impact against affected deployments. No public exploit code and no CISA KEV listing have been identified at time of analysis; however, the zero-prerequisite attack profile and broad version coverage across four major Spring branches (5.3.x, 6.1.x, 6.2.x, 7.0.x) make this relevant to any Java shop running Spring MVC or WebFlux with multipart upload handling enabled.
Spring Expression Language (SpEL) evaluation logic in Spring Framework 5.3.x through 7.0.x fails to enforce method invocation restrictions in read-only or restricted contexts, allowing remote unauthenticated attackers (CVSS PR:N, AV:N) to invoke arbitrary zero-argument methods and trigger unintended application logic. Scored LOW (3.7) with only availability impact per CVSS (A:L), though the 'Authentication Bypass' tag and CWE-863 (Incorrect Authorization) suggest the authorization bypass itself may have broader implications not fully reflected in the score. No public exploit identified at time of analysis and no CISA KEV listing, but the wide version range across four active Spring Framework release trains represents significant ecosystem exposure.
Security bypass in Spring WebFlux's Kotlin Router DSL (CWE-284: Improper Access Control) allows remote unauthenticated attackers to circumvent access control rules in Spring Framework 5.3.0 through 5.3.48. The CVSS vector (AV:N/AC:H/PR:N/UI:N) indicates network-reachable exploitation without authentication, though high attack complexity constrains opportunistic exploitation. No public exploit identified at time of analysis, and this vulnerability is not listed in the CISA KEV catalog.
Cross-site scripting (XSS) in Spring Framework's MVC JSP form tags allows unauthenticated remote attackers to inject arbitrary HTML or JavaScript into rendered pages by supplying malicious values through the cssClass, cssErrorClass, or cssStyle tag attributes. Applications across four active Spring Framework release lines (5.3.x through 7.0.x) are affected when they pass user-controlled input directly into these tag attributes. No public exploit code has been identified at time of analysis, and CISA has not listed this CVE in the Known Exploited Vulnerabilities catalog, but the broad installed base of Spring MVC in enterprise Java environments and the high confidentiality impact (session hijacking, credential theft) warrant prompt patching.
Cross-site scripting in Spring Framework versions 5.3.0-5.3.48, 6.1.0-6.1.27, 6.2.0-6.2.18, and 7.0.0-7.0.7 allows remote attackers to inject JavaScript into victim browsers when applications rely on JavaScriptUtils.javaScriptEscape() for output encoding. The flaw stems from incomplete escaping in this utility method, and successful exploitation requires user interaction (UI:R) such as visiting a crafted page. No public exploit has been identified at time of analysis and the issue is not in CISA KEV.
Open redirect in Spring Framework (Spring MVC and Spring WebFlux) across four major version branches enables unauthenticated remote attackers to craft URLs that cause the application to issue a 302 HTTP redirect to an arbitrary attacker-controlled external host. The vulnerability is conditionally exploitable - requiring a catch-all wildcard route mapping without an explicit view name - and demands user interaction to trigger. CVSS rates this 4.2 Medium (AV:N/AC:H/PR:N/UI:R); no public exploit code or CISA KEV listing has been identified at time of analysis.
Denial of Service in Spring WebFlux's multipart request processing allows unauthenticated remote attackers to exhaust server resources across all supported Spring Framework branches. Affects Spring Framework 5.3.x through 7.0.x when applications use the reactive WebFlux stack and expose endpoints that accept multipart data. No public exploit code has been identified at time of analysis, and CISA KEV listing is absent, but the network-reachable, zero-privilege attack surface warrants prompt patching for internet-facing WebFlux deployments.
Session fixation in Spring Framework's WebFlux reactive stack (versions 5.3.x through 7.0.x) enables a remote attacker to hijack an authenticated user's session by leveraging a compromised subdomain - typically via cross-site scripting - to plant a known session ID and exchange it for the victim's authenticated session post-login. The attack is classified as CWE-384 and requires both a prior subdomain compromise and user interaction, placing real-world exploitability well below the headline concern for most deployments. No public exploit code and no CISA KEV listing have been identified at time of analysis.
Unbounded HTTP/2 stream creation in Netty's netty-codec-http2 library exposes any Netty HTTP/2 server running on default configuration to memory exhaustion from a single TCP connection. Because DefaultHttp2Connection.DefaultEndpoint initializes stream limits to Integer.MAX_VALUE and Http2Settings never advertises SETTINGS_MAX_CONCURRENT_STREAMS unless the application explicitly calls initialSettings().maxConcurrentStreams(n), a remote unauthenticated attacker can sustain hundreds of thousands of simultaneous streams, each forcing JVM heap allocations for DefaultStream objects, PropertyMap slots, flow-controller state, and IntObjectHashMap entries. This misconfiguration is also the structural precondition for CVE-2023-44487-style HTTP/2 Rapid Reset amplification. No public exploit has been identified at time of analysis; vendor-released patches are available in 4.1.135.Final and 4.2.15.Final.
DNS cache poisoning in Netty's resolver component (io.netty:netty-resolver-dns) enables remote unauthenticated attackers to redirect downstream application traffic to attacker-controlled IPs through a Kaminsky-style spoofing attack. The vulnerability combines two compounding weaknesses present in the default configuration: DNS transaction IDs shuffled with a mathematically predictable Linear Congruential Generator (ThreadLocalRandom), and a static UDP source port resulting from the default ChannelPerResolver channel strategy. No public exploit has been identified at time of analysis and the issue is not listed in CISA KEV, but the Netty project rated it high severity and released fixes in versions 4.1.135.Final and 4.2.15.Final.
File descriptor exhaustion in Netty's Unix domain socket native transport allows a local peer to leak two file descriptors per crafted SCM_RIGHTS message into the receiving process, with neither FD ever closed. Affected are applications explicitly configured with DomainSocketReadMode.FILE_DESCRIPTORS on Epoll or KQueue DomainSocketChannel - a non-default opt-in setting. Sustained message flooding by a local socket peer can exhaust the Netty process's file descriptor table, ultimately causing denial of service. No public exploit identified at time of analysis and not listed in CISA KEV; CVSS scores this at 4.0 (Low) reflecting the local-only attack surface and low availability impact.
Unauthenticated open redirect in Authlib's OAuth 2.0 authorization endpoint allows any remote attacker to weaponize a trusted authorization server domain as a redirector - no client registration, no authenticated session, and no prior state required. The flaw affects all deployments of pip/authlib < 1.6.10 and == 1.7.0 running an authorization endpoint via the Flask or Django integrations. Publicly available exploit code exists and was dynamically confirmed against the live HEAD (v1.6.6-104-g68e6ab3f); no public exploit identified at time of analysis as confirmed actively exploited (CISA KEV status absent), but the trivial, zero-prerequisite exploit path makes phishing and domain-allowlist bypass practical at scale.
Infinite loop denial-of-service in the Linux kernel's drm/v3d GPU driver allows a local low-privileged user to permanently peg a CPU core by submitting a crafted self-referential ioctl extension structure. The vulnerability exists in v3d_get_extensions(), which traverses a userspace-controlled linked list without bounding chain length; when both in_sync_count and out_sync_count are zero, the duplicate-extension guard silently passes on every iteration, trapping the kernel thread indefinitely. No public exploit or active exploitation has been identified at time of analysis; EPSS stands at 0.02%, consistent with the local-access-only attack surface.
Error pointer dereference in the Linux kernel's Intel IPU6 media driver crashes the kernel during a failed probe, resulting in local denial of service. The flaw resides in `ipu6_pci_probe()` at `drivers/media/pci/intel/ipu6/ipu6.c:690`, where `isp->psys` holds an ERR_PTR value rather than NULL in a specific error path, causing it to be dereferenced during cleanup rather than handled safely. No public exploit exists and EPSS sits at 0.02% (5th percentile), reflecting negligible real-world exploitation interest.
Kernel warning (WARN_ON) triggerable in the Linux videobuf2 DMA scatter-gather subsystem on Apple Silicon systems allows a low-privileged local user to destabilize video capture pipelines by performing mmap() on an imported DMA-buf. The vb2_dma_sg_mmap function fails to set VM_DONTEXPAND and VM_DONTDUMP VMA flags - protections that vb2_dma_contig correctly applies - causing drm_gem_mmap_obj() to hit its assertion guard. No public exploit has been identified at time of analysis; EPSS is 0.02% (5th percentile), consistent with the narrow hardware-specific trigger surface.
NULL pointer dereference in the Linux kernel's Renesas VSP1 media driver crashes the kernel during module unload on Generation 4 Renesas SoC hardware. The defect exists because the cleanup path unconditionally invokes vsp1_drm_cleanup() rather than vsp1_vspx_cleanup() for gen 4 IP versions, while the initialization path correctly checks the IP version - an asymmetry introduced when gen 4 support was added. With an EPSS of 0.02% (4th percentile), no KEV listing, and no public exploit code, real-world exploitation risk is very low and confined to niche embedded and automotive platforms; vendor-released patches are available in stable kernel branches.
Local denial-of-service in the Linux kernel's SELinux subsystem allows any low-privileged local process to monopolize the `/sys/fs/selinux/policy` file descriptor, blocking all other processes from reading the active kernel policy. The root cause is a `policy_opened` flag that enforced a single-open restriction on the SELinux policy pseudofile - originally intended to prevent inconsistent reads or unbounded kernel memory allocation, but which also created a trivial resource-starvation primitive. No active exploitation has been identified (not in CISA KEV), and the EPSS score of 0.02% places this in the 5th percentile for exploitation likelihood.
Deadlock vulnerability in the Linux kernel's pseries/papr-hvpipe subsystem allows a local low-privileged attacker on IBM pSeries Power Architecture systems to cause a kernel deadlock by triggering a race between the ioctl or release handler and a hardware interrupt firing on the same CPU. Exploitation is architecture-specific, requiring IBM PAPR hardware with the hvpipe driver loaded, and results in an availability-only denial-of-service. No public exploit exists and EPSS probability stands at 0.02%, consistent with a narrow attack surface; vendor-released patches are confirmed available for Linux 7.0.7, 7.1-rc3, and the 6.18 stable branch.
Improper interrupt registration in the Linux kernel's libwx network driver causes a kernel WARNING when VF (Virtual Function) misc interrupts are initialized, resulting in high availability impact on affected systems. The libwx driver incorrectly calls request_threaded_irq() with a primary handler, a NULL threaded handler, and the IRQF_ONESHOT flag - an invalid combination flagged as a WARNING by the kernel interrupt management subsystem since commit aef30c8d569c. Exploitation requires local access with low privileges on a system equipped with a Wangxun VF NIC using the libwx driver; no public exploit code exists and EPSS is negligible at 0.02%.
NULL pointer dereference in the Linux kernel's s3c64xx SPI driver triggers a kernel crash on driver unbind, allowing a local low-privileged attacker to cause a denial of service. The flaw affects systems equipped with Samsung S3C64xx-series SoC hardware running unpatched kernel versions from 6.0 onward. No public exploit code exists and EPSS probability sits at 0.02%, but the crash is deterministic once the driver unbind sequence is triggered on vulnerable hardware. No active exploitation (CISA KEV) has been identified.
KVM/x86 on Linux exposes a race condition in interrupt request register (IRR) scanning that causes incorrect reporting of the highest pending interrupt in nested virtualization scenarios. Specifically, when PID.ON is set by a sender vCPU after the target vCPU has already harvested and cleared the PIR, a second call to vmx_sync_pir_to_irr() observes an empty PIR and returns max_irr=-1 without scanning the IRR, producing a spurious kernel WARNING in vmx_check_nested_events() and a wasted L2 VM-Enter/VM-Exit cycle. The interrupt itself is not lost - it resides in the IRR from the first sync - but the incorrect max_irr triggers the assertion. No public exploit exists and EPSS is 0.02%, consistent with the highly localized, timing-dependent nature of this defect.
Missing resource cleanup in the Linux kernel's generic power domain (genpd) subsystem causes runtime PM to remain enabled for virtual devices after detach, leading to a NULL pointer dereference in genpd_runtime_suspend() and local denial of service. The flaw affects systems where drivers use genpd_dev_pm_attach_by_id() - predominantly ARM and SoC-based platforms - since the balancing pm_runtime_disable() call is absent from genpd_dev_pm_detach(). No public exploit exists and EPSS at 0.02% (5th percentile) confirms negligible exploitation probability; this is not listed in CISA KEV.
Sensitive HMAC key material is exposed through kernel debug logging in the Linux kernel's CAAM (Cryptographic Acceleration and Assurance Module) crypto driver. On systems with CONFIG_DYNAMIC_DEBUG enabled and debug output activated for the caam module at runtime, calls to hash_digest_key() emit raw HMAC key bytes via hex dump into the kernel log, allowing any local attacker with debug log read access to extract cryptographic secrets. No public exploit code has been identified and EPSS probability is 0.02%, though the confidentiality impact is severe if exploitation conditions are met. The provided CVSS vector (C:N/A:H) appears to misclassify this as an availability issue rather than a confidentiality issue - see confidence notes.
Linux kernel x86 EFI runtime service fault handling panics instead of recovering gracefully on systems with buggy EFI firmware, following the FPU softirq changes introduced in commit d02198550423. When kernel_fpu_begin() began using local_bh_disable() in place of preempt_disable(), SOFTIRQ_OFFSET is set in preempt_count for the duration of EFI runtime service calls, causing in_interrupt() to return true in normal task context - which causes efi_crash_gracefully_on_page_fault() to bail unconditionally, escalating to panic(). On hardware where EFI firmware (e.g., GetTime()) accesses unmapped memory, this produces an unrecoverable system freeze. No public exploit identified at time of analysis; EPSS is 0.02%, confirming no observed exploitation activity.
Array out-of-bounds read in the Linux kernel's Qualcomm Light Pulse Generator (qcom-lpg) LED driver allows a local low-privileged user to crash the kernel via denial of service on affected Qualcomm SoC-based systems. The flaw stems from using FIELD_GET() to extract a 3-bit register value (range 0-7) as an index into an array with only 5 elements, enabling reads beyond array bounds and subsequent hardware misconfiguration or kernel panic. No public exploit has been identified and EPSS stands at the 5th percentile, placing this firmly in lower-priority triage despite the technically High availability impact per CVSS.
NULL pointer dereference in the Linux kernel's HugeTLB early boot parameter parser causes a system crash before the OS fully initializes. Specifically, `hugetlb_add_param()` in `mm/hugetlb` dereferences a NULL pointer via `strlen()` when `hugepages`, `hugepagesz`, or `default_hugepagesz` kernel command line parameters are supplied without a '=' separator, producing a boot-time kernel panic. The impact is a complete denial of service - the system fails to boot until the malformed parameter is corrected. No public exploit or CISA KEV listing exists; EPSS stands at 0.02% (5th percentile), reflecting very low observed exploitation activity. Patches are available in Linux 6.18.27, 7.0.4, and 7.1-rc1, with Ubuntu distributing fixes via USN-8489-1 and USN-8488-1.
Sensitive cryptographic material-HMAC session keys, nonces, and passphrase data held in struct tpm2_auth-is left in freed slab memory when a TPM device is released via tpm_dev_release() in Linux kernels from 6.10 onward, because that code path uses kfree() rather than kfree_sensitive(). Every other tear-down path in the same TPM driver (tpm2_end_auth_session() and tpm_buf_check_hmac_response()) already calls kfree_sensitive(), making this an isolated inconsistency. No public exploit exists and EPSS is 0.02% (5th percentile), but in high-assurance environments where TPM-backed key confidentiality is mandatory the residual key material is a meaningful exposure until overwritten by subsequent allocations.
NULL pointer dereference in the Linux kernel's admv1013 IIO frequency driver crashes the kernel when device_property_read_string() fails during device initialization, leaving a garbage pointer subsequently passed to strcmp(). A local low-privileged user on a system with ADMV1013 microwave upconverter hardware can reliably trigger a kernel panic, resulting in a complete denial of service. No public exploit exists and EPSS sits at 0.02% (5th percentile), but vendor-released patches are available across multiple stable kernel branches including 6.12.86 and 7.0.4.
NULL pointer dereference in the Linux kernel's drm/imagination (PowerVR GPU) driver crashes the kernel when a local user writes to the ftrace mask debugfs entry. Affected kernels range from the introduction of the drm/imagination driver at commit a331631496a0af9a6f4e7e1860983afd8b1bb013 through Linux 7.0, with fixes landed in 7.0.4 and 7.1-rc2. An attacker with local access and write permission to the debugfs interface can trigger a kernel Oops, resulting in a full system crash and denial of service. No public exploit exists and EPSS is 0.02%, consistent with the narrow hardware scope (Imagination Technologies PowerVR GPUs) and local-only attack surface.
Kernel panic in the Linux amdgpu DRM driver exposes systems running RDNA4 hardware (e.g., AMD RX 9070 XT) to a denial-of-service condition during driver initialization, but only when the kernel is compiled with CONFIG_DRM_DEBUG_MM enabled. The RDNA4 architecture (GFX 12) removed the GDS, GWS, and OA on-chip memory resources entirely; however, amdgpu_ttm_init() unconditionally invokes amdgpu_ttm_init_on_chip() for these resources regardless of size, ultimately calling drm_mm_init(mm, 0, 0), which triggers DRM_MM_BUG_ON and crashes the kernel at modprobe time. No public exploit exists and EPSS sits at 0.02% (7th percentile), consistent with a hardware-specific, debug-config-dependent crash rather than a remotely triggerable attack surface.
Open redirect in WebOb (pip/webob <= 1.8.9) enables unauthenticated network attackers to redirect victims to arbitrary attacker-controlled domains by bypassing the prior CVE-2024-42353 patch. The bypass exploits Python 3.10+'s silent stripping of ASCII control characters (tab, CR, LF) from URLs before parsing, so a crafted path like /\t/attacker.com passes the previous // guard, then gets normalized by urlsplit into a protocol-relative URL pointing to attacker.com. No public exploit has been identified at time of analysis beyond the advisory's own proof-of-concept code, but the bypass technique is trivial and the full exploit path is published in the GitHub advisory GHSA-fh3h-vg37-cc95.
Path traversal in Docling's LaTeX backend (pip/docling versions 2.73.0 through 2.90.x) allows an attacker who supplies a crafted LaTeX document to read arbitrary files accessible to the conversion process via the \includegraphics, \input, and \include command handlers. With a high confidentiality impact (C:H) but local attack vector and required user interaction (AV:L/UI:R per CVSS), the practical risk is concentrated in automated document-processing pipelines or services that ingest untrusted .tex files. No public exploit identified at time of analysis, and the vulnerability is not listed in CISA KEV. A vendor-released patch exists as of version 2.91.0.
Improper authorization in Apache Kafka 4.0.0-4.3.0 arises from a discrepancy between the documented ACL requirement and the actual runtime behavior of the CONSUMER_GROUP_DESCRIBE (API key 69) endpoint. The API checks for DESCRIBE permission on the GROUP resource at runtime, while official Kafka documentation and KIP-848 specify READ as the required operation - causing administrators who followed the documentation to configure ACLs that either over-grant READ access to users who should only observe group metadata, or under-restrict DESCRIBE-only users who can nonetheless access sensitive consumer group state. No public exploit has been identified at time of analysis, and EPSS exploitation probability is 0.02% (4th percentile), indicating negligible opportunistic exploitation risk.
Denial-of-service in Spring Cloud Function allows a low-privileged actor with physical access and required user interaction to exhaust JVM heap memory by registering an unbounded number of functions into the Function Registry, triggering an Out-of-Memory (OOM) crash. Five actively maintained release lines (3.2.x through 5.0.x) plus all unsupported versions are affected. No public exploit code exists and the vulnerability is not listed in CISA KEV at time of analysis; the CVSS score of 5.7 reflects the physical access vector and user interaction dependency that substantially constrain real-world exploitability compared to the High availability impact alone.
Uncontrolled recursion in Spring Cloud Function's routing layer triggers an Out-of-Memory (OOM) condition, resulting in a denial of service across all active release trains (3.2.x through 5.0.x). The vulnerability is rooted in CWE-674 (Uncontrolled Recursion), where the routing layer lacks depth limits or cycle detection, allowing recursive request-handling to exhaust JVM heap space. No public exploit code or CISA KEV listing has been identified at time of analysis; however, the CVSS physical attack vector (AV:P) appears inconsistent with a routing-layer vulnerability and warrants verification against the Spring/VMware advisory.
OpenShift Container Platform exposes a resource exhaustion path where low-privileged authenticated users with pod creation rights can trigger cluster-wide API server degradation by exploiting two quota enforcement gaps: completed pods with restartPolicy:Never are excluded from ResourceQuota pod limits, and Kubernetes events are not quota-scoped. By repeatedly creating such short-lived pods, an attacker causes Kubernetes events to accumulate unboundedly in etcd, degrading API server performance across the entire cluster - a scope change (CVSS S:C) that extends impact well beyond the attacker's own namespace. No public exploit has been identified at time of analysis, and the vulnerability is not listed in the CISA KEV catalog.
Unauthenticated information disclosure in Apache ActiveMQ Broker allows remote attackers to enumerate all durable topic subscriptions - including client identifiers, subscription names, topic destinations, and JMS selector expressions - by sending a BrokerInfo command to a broker with syncDurableSubs enabled on a network connector. The broker incorrectly skips authentication before servicing the BrokerInfo request, exposing sensitive messaging infrastructure metadata. No public exploit identified at time of analysis, and EPSS stands at 0.02% (6th percentile), indicating very low current exploitation probability despite network-reachable attack vector.
Incomplete authorization in Apache ActiveMQ Broker allows authenticated low-privilege users to remove messaging destinations (queues and topics) beyond their granted permissions, causing targeted availability disruption to broker-connected producers and consumers. Affected versions span the 5.x line before 5.19.7 and the 6.x line from 6.0.0 before 6.2.6, covering ActiveMQ Broker and ActiveMQ All distributions. With an EPSS of 0.01% (3rd percentile), no CISA KEV listing, and no public exploit code identified, this is a low-urgency but genuine authorization control gap that is most relevant to environments with multiple untrusted authenticated broker accounts.
Server-Side Request Forgery in Clair's fetcher component exposes internal network services and cloud metadata endpoints to unauthenticated remote attackers who can submit container image manifests with crafted layer descriptor URIs. Affected deployments are those where PSK (Pre-Shared Key) authentication is not configured - an opt-in control that is not enforced by default - meaning standalone or custom Clair installations without PSK are directly exploitable over the network with no credentials. Reflective SSRF behavior leaks up to 256 bytes of internal error body content per request via CheckResponse error messages, enabling network reconnaissance; no public exploit has been identified at time of analysis and the vulnerability is not listed in CISA KEV.
Panic-induced denial of service in the golang.org/x/image/bmp decoder allows remote unauthenticated attackers to crash any Go application that processes untrusted BMP files. When a paletted BMP containing an out-of-range palette index is decoded, the image package accesses invalid memory and triggers a Go runtime panic, terminating the affected goroutine or process. No public exploit exists at time of analysis, but the attack is automatable per SSVC and requires no privileges or user interaction (CVSS AV:N/AC:L/PR:N/UI:N), making any internet-exposed BMP processing endpoint an accessible target.
Excessive RBAC permissions in the Metal3 ip-address-manager (IPAM) controller's ClusterRole expose Kubernetes Secrets to post-compromise abuse. The metal3-ipam-controller-manager-role granted full CRUD on core/v1 Secrets - yet PR diff evidence confirms the controller never accesses Secrets in normal operation, making these permissions entirely unnecessary. If the controller pod is compromised via a supply chain attack or container escape, an attacker inheriting the controller's service account token can read, modify, or delete any Secret in the namespace, turning a narrow container compromise into cluster-wide credential exfiltration. No public exploit exists and no CISA KEV listing, but the confidentiality impact is rated High (C:H) by NVD.
Host process data leakage in vm2 (npm/vm2 <= 3.11.3) allows untrusted JavaScript executing inside a NodeVM sandbox to read sensitive host-process state - including HTTP Authorization headers, session tokens, and AsyncLocalStorage request context - by requiring process-wide observability builtins that were incorrectly absent from the DANGEROUS_BUILTINS denylist. The affected modules (diagnostics_channel, async_hooks, perf_hooks) are Node.js process-wide singletons, not sandbox-local constructs, so granting sandbox access to them exposes all host process data flowing through those channels. A publicly available exploit code exists confirming extraction of Bearer tokens and x-session-token headers from live host HTTP requests; this is not confirmed actively exploited (not in CISA KEV).
CPU and log amplification in opentelemetry-go's baggage parser allows remote unauthenticated attackers to trigger denial-of-service by submitting oversized or malformed W3C baggage headers to any instrumented Go service. PR #7880 inadvertently removed the upfront raw-length check and per-member size guard from `baggage.Parse`, meaning the parser now fully tokenizes and percent-decodes every member of an arbitrarily large input rather than rejecting it early. A publicly available proof-of-concept exists (GHSA-5wrp-cwcj-q835); the vulnerability is not yet listed in CISA KEV, and impact is bounded to availability with no confirmed confidentiality or integrity consequence.
Uncontrolled resource consumption in PyJWT 2.8.0-2.12.1 exposes any service that verifies detached JWS tokens to unauthenticated denial-of-service. When the unencoded-payload extension (b64=false, RFC 7797) is in use, PyJWT unnecessarily Base64URL-decodes the compact-serialization payload segment before discarding it in favor of the caller-supplied detached payload - turning that segment into an attacker-controlled amplifier for CPU and memory exhaustion regardless of signature validity. No public exploit has been identified at time of analysis, but the CVSS vector (AV:N/AC:L/PR:N/UI:N) confirms fully unauthenticated remote exploitation against any affected endpoint using this feature.
PyJWKClient in PyJWT prior to 2.13.0 passes attacker-influenced URIs directly to Python's urllib.request.urlopen() without restricting URI schemes, enabling Server-Side Request Forgery (SSRF) across file://, FTP, and data-URI schemes against applications that accept untrusted jku values. Affected deployments include any Python application using PyJWKClient where the jku URL originates from a JWT header, OAuth flow parameter, or externally influenced configuration. No public exploit exists and no CISA KEV listing is present; real-world exploitation is constrained by a CVSS-confirmed high attack complexity (AC:H) and required user interaction (UI:R), making opportunistic mass exploitation unlikely.
Memory exhaustion in pypdf prior to 6.12.0 allows an attacker who supplies a crafted PDF to cause large memory consumption in any application that processes it using layout mode text extraction. The vulnerability is triggered by PDFs containing text positioning operators with abnormally large x- or y-coordinate offsets, causing the library to allocate unbounded whitespace and newline characters during rendering. No confirmed active exploitation exists (not in CISA KEV), and SSVC rates this as non-automatable with partial technical impact, placing it in a lower operational priority tier despite the straightforward exploitation mechanic.
Denial-of-service via algorithmic complexity in pypdf before 6.12.0 allows an attacker who can supply a crafted PDF file to cause excessive processing time during cross-reference stream parsing. The vulnerability is triggered by crafting a PDF with /W [0 0 0] field values in a cross-reference stream combined with a large /Size value, which causes the library to perform unbounded iteration over zero-byte entries. No public exploit code has been identified at time of analysis, and this vulnerability is not listed in the CISA KEV catalog; however, any application that processes untrusted PDF input using pypdf is exposed.
Memory exhaustion in pypdf's XMP metadata parser allows denial of service via specially crafted PDF files containing oversized or element-dense XMP blocks, affecting all versions prior to 6.12.1. The vulnerability stems from an absence of input limits in the XML-based XMP parsing subsystem (CWE-770), meaning processing a malicious PDF can consume unbounded system memory. No public exploit code has been identified at time of analysis, and no confirmed active exploitation exists; however, the patch diff is publicly visible on GitHub, making trivial exploit construction feasible.
Out-of-bounds write in bzip2's bzip2recover utility allows a local attacker to supply a specially crafted file that triggers an off-by-one error, corrupting a global buffer and crashing the process. Per the CVSS 4.0 vector (AV:L/AC:L/AT:N/PR:N/UI:N), the attack requires no privileges and no user interaction beyond the utility being invoked against a malicious file. Impact is strictly denial of service against the bzip2recover process - no confidentiality or integrity exposure - and the CVSS 4.0 score of 5.1 (Medium) reflects this constrained scope. No public exploit or active exploitation has been identified at time of analysis.
Runtime PM reference count leak in the Linux kernel's OmniVision OV5647 camera sensor driver (media/i2c/ov5647) causes availability loss for systems equipped with this camera hardware. The s_ctrl function's handling of three V4L2 controls - AUTOGAIN, EXPOSURE_AUTO, and ANALOGUE_GAIN - returns early without invoking pm_runtime_put(), allowing unpaired runtime PM get/put calls to accumulate. A local user with access to the camera device node can trigger this imbalance repeatedly, exhausting the PM reference count and preventing the device from entering low-power states, ultimately making it unavailable. No public exploit has been identified; EPSS is 0.02% (5th percentile) and this vulnerability does not appear in CISA KEV.
Availability impact in the Linux kernel's xbox_remote media/rc driver allows a local, low-privileged user with access to the affected device to crash the kernel via a DMA coherency violation. The IO buffer used for USB transfers is embedded directly within the device structure, which violates DMA coherency rules and can trigger memory corruption leading to kernel panic. No public exploit has been identified at time of analysis, and the EPSS score of 0.02% reflects minimal observed exploitation activity. Vendor-released patches are available across multiple stable kernel branches.
Null pointer dereference in the Linux kernel's saa7164 media driver can crash the kernel when ioremap() fails during PCI device initialization. Systems with SAA7164-based PCIe capture cards (e.g., Phillips/NXP SAA7164) running unpatched kernel versions from 2.6.32 through stable branches prior to 6.6.140, 6.12.90, 6.18.32, and 7.0.9 are affected. A local, low-privileged attacker who can trigger driver initialization under memory pressure conditions may cause a kernel oops or panic, resulting in denial of service. No public exploit exists and EPSS is 0.02% (5th percentile), indicating negligible real-world exploitation probability at this time.
NULL pointer dereference in the Linux kernel's batman-adv Bridge Loop Avoidance (BLA) subsystem enables local denial of service via kernel panic. The race condition exists in batadv_bla_purge_claims(), which traverses the claim hash list under rcu_read_lock() without accounting for concurrent claim releases - when batadv_claim_release() NULLs the backbone_gw pointer mid-traversal, a subsequent call to batadv_bla_claim_get_backbone_gw() on the partially-freed claim triggers a NULL dereference. Exploitation requires local low-privilege access on systems actively running batman-adv with BLA enabled; no active exploitation is confirmed and EPSS stands at 0.02% (5th percentile).
Reference counting failure in the Linux kernel's batman-adv Bridge Loop Avoidance (BLA) subsystem allows a local low-privileged user to cause a slow kernel memory leak and eventual denial of service. Specifically, batadv_bla_add_claim() omits the required batadv_backbone_gw_put() call on the error path when a claim hash insertion fails, preventing the backbone_gw object's reference count from ever reaching zero. The vulnerability has been present since at least kernel 4.7 and is patched across multiple stable branches; no public exploit exists and EPSS is extremely low at 0.02% (5th percentile), placing this firmly in the low-urgency tier except for environments actively running batman-adv with BLA.
Stale VRAM exposure in the Linux kernel's amdkfd driver (drm/amdkfd) allows local authenticated users to observe prior occupants' GPU memory contents and crash multi-GPU compute workloads. The KFD allocation path omitted the AMDGPU_GEM_CREATE_VRAM_CLEARED flag - present in every other userspace GEM allocation path - leaving freshly allocated VRAM buffers populated with data from previous processes. The primary documented availability impact is deterministic crashes in RCCL peer-to-peer transport when stale values corrupt the ptrExchange, head, and tail protocol fields; a secondary information-disclosure risk exists because stale page-table remnants are observable by unprivileged compute kernels. No public exploit identified at time of analysis; EPSS is 0.02% (5th percentile).
Memory leak in the Linux kernel's spi: ch341 USB-to-SPI driver allows a local user to degrade system availability by triggering driver unbind without physical device disconnection. The defect lies in incorrect devres (device-managed resource) lifetime scoping - resources are anchored to the parent USB device rather than the USB interface, so they are not released when the driver unbinds during events such as probe deferral or runtime configuration changes. No public exploit has been identified at time of analysis, and an EPSS score of 0.02% (4th percentile) indicates negligible exploitation probability in the near term.
Improper driver teardown ordering in the Linux kernel's Freescale (FSL) SPI controller driver allows a local low-privileged user to trigger a denial-of-service condition during driver unbind. The SPI controller is not deregistered before underlying resources such as DMA are released, creating a window where the controller may reference freed memory. No public exploit has been identified at time of analysis, and EPSS sits at 0.02% (5th percentile), reflecting low real-world exploitation probability.
Stack buffer overflow in GStreamer's H.265/HEVC codec parser (gst-plugins-bad) allows remote unauthenticated attackers to crash GStreamer-based applications by delivering a crafted H.265 video file or stream that a user opens. The root cause is an incorrect loop bound in the buffering period SEI message parser: the parser mistakenly uses cpb_cnt_minus1[i] (the current loop index variable) rather than cpb_cnt_minus1[0] from the referenced Sequence Parameter Set, causing the loop to iterate beyond the bounds of stack-allocated CPB delay arrays and corrupt stack memory. No public exploit has been identified at time of analysis and the vulnerability is not listed in CISA KEV, though the deterministic parser logic makes crash reproduction straightforward.
Out-of-bounds write in GStreamer's H.266/VVC PPS picture partition parser (`gst-plugins-bad`) allows an attacker to crash media-processing applications - and potentially achieve code execution - by delivering a crafted H.266/VVC media file. The flaw in `gst_h266_parser_parse_picture_partition()` (gsth266parser.c) permits unbounded slice index increments across three fixed-size arrays in `GstH266PPS` during multi-slice-in-tile processing. A proof-of-concept demonstrating at least a 4-byte write exists; no public exploit beyond that initial POC or CISA KEV listing has been identified at time of analysis, though the code structure permits larger writes across multiple iterations which elevates downstream risk above a pure DoS assessment.
HTTP Response Splitting via CRLF injection in Apache CXF's OAuth2 module allows an attacker who controls the WWW-Authenticate realm parameter to inject arbitrary HTTP headers or split HTTP responses entirely. Affected deployments include cxf-rt-rs-security-oauth2 versions 4.2.0 before 4.2.2 and all versions before 4.1.7. No public exploit code has been identified at time of analysis and this vulnerability is not listed in CISA KEV, but successful exploitation could enable cache poisoning, header injection, or redirection of downstream HTTP clients processing the malformed response.
Log injection in Apache CXF's OAuth2 module (org.apache.cxf:cxf-rt-rs-security-oauth2) permits remote attackers to forge arbitrary log entries by supplying crafted `clientId` values containing control characters or newline sequences in OAuth2 HTTP requests. Affected are CXF 4.2.0-4.2.1 and all 4.1.x versions before 4.1.7; fixed releases 4.2.2 and 4.1.7 were issued June 10, 2026. No public exploit code or active exploitation has been identified at time of analysis; practical impact is confined to log integrity compromise that could mislead security monitoring and incident response processes.
Authentication bypass in Apache CXF's OAuth2 TokenIntrospectionService allows unauthenticated network access to the token introspection endpoint due to a missing 'throw' keyword in the internal security context check, causing the guard to silently pass rather than reject unauthorized callers. Affected are deployments using cxf-rt-rs-security-oauth2 versions 4.2.0-4.2.1 and all 4.1.x releases before 4.1.7 that relied solely on CXF's built-in check without independent authentication at the container or gateway layer. No public exploit code or active exploitation has been identified; vendor-confirmed patches (4.2.2 and 4.1.7) were released June 10, 2026.
Denial of service in the joi npm validation library (< 18.2.1) is triggered by submitting deeply nested JSON objects against schemas that use recursive `link()` definitions, causing a JavaScript call stack overflow. The untrapped `RangeError` propagates out of joi's link resolver - crashing the Node.js process entirely when `validate()` is invoked without a surrounding `try/catch`, or silently breaking error-handling logic when a `RangeError` is returned where a structured `ValidationError` was expected. No confirmed active exploitation (not in CISA KEV) and no public exploit code identified at time of analysis, but the attack surface is inherently wide given joi's prevalence in Node.js web services.
Host confusion in guzzlehttp/psr7 (all versions prior to 2.10.2) allows unauthenticated network attackers to supply a malformed Host header - such as `trusted.example@evil.example` - causing the library's URI construction logic to reinterpret the value as URI userinfo and a different host, silently replacing the parsed URI host with the attacker-controlled domain. Applications that rely on the resulting PSR-7 URI host for routing, allow-list enforcement, or forwarding decisions are at risk of sending requests and credentials to unintended destinations. No public exploit has been identified at time of analysis and no CISA KEV listing exists, but the conditional impact on forwarding gateways and API proxies built on psr7's server-request parsing functions is concrete.
Artifact pin decay in vLLM (< 0.22.0) allows pinned model deployments to silently load secondary artifacts - dynamic code modules, GGUF files, image processors, retrieval side weights, and same-repository subfolder configs - from unpinned or default HuggingFace Hub revisions, undermining supply-chain integrity guarantees. Operators who supply `--revision` or `--code-revision` to audit and lock a specific model version can unknowingly serve behavior-affecting components that fall outside the reviewed artifact set. No public exploit code has been identified at time of analysis, but seven discrete affected code paths across vllm's model loader and registry subsystems have been confirmed at commit 3795d7acf431980e62e738493f437ae2a51549da, with a fix released in vLLM 0.22.0 via PR #42616.
Cache key collision in OpenFGA's iterator caching mechanism allows two distinct authorization check requests to resolve to the same cache key, causing the engine to return a stale, incorrect authorization result to a subsequent requester. All OpenFGA deployments prior to version 1.16.0 with iterator caching enabled are affected, specifically when using the experimental weighted_graph_check union resolution path. An authenticated attacker who can trigger authorization checks under conditions that produce a colliding cache key may receive an incorrect allow or deny decision, undermining the integrity and confidentiality of access control enforced by the engine. No public exploit identified at time of analysis, and this vulnerability is not listed in the CISA KEV catalog.
Stored cross-site scripting in Red Hat's migration-planner-ui-app (kubev2v project) allows an authenticated attacker who can register a discovery agent to inject a javascript: URI into the agent's credentialUrl field, which executes in another organizational user's browser when they click the resulting link in the UI. No public exploit identified at time of analysis, but the upstream fix (PR #750) and Red Hat Bugzilla entry confirm the issue is real and patched via a safeExternalUrl protocol allowlist. Successful exploitation hijacks the victim's Red Hat SSO session and enables cross-tenant data access and API actions.
Unencrypted secret storage in Jenkins 2.567 and earlier (LTS 2.555.2 and earlier) exposes credentials submitted via POST config.xml to any user holding Item/Extended Read permission or with read access to the Jenkins controller filesystem. Secrets that should be encrypted at rest are written as plaintext into job config.xml files, making them directly readable through Jenkins' built-in permission model or OS-level file access. No public exploit code has been identified and this CVE is not listed in the CISA KEV catalog, but the privileged insider and lateral-movement risk is significant for organizations embedding CI/CD credentials in job configurations.
Open redirect in Jenkins 2.567 and earlier (LTS 2.555.2 and earlier) enables unauthenticated remote attackers to craft login URLs with a malicious external domain in the 'from' parameter under the 'Delegate to servlet container' security realm, silently redirecting authenticated users to attacker-controlled sites post-login for phishing purposes. The CVSS vector (PR:N, UI:R) confirms no attacker authentication is needed, but victim interaction is mandatory - the user must click a crafted link and complete a login. No public exploit identified at time of analysis and the vulnerability is not listed in CISA KEV; vendor-released patches are available as of 2026-06-10.
Missing authorization checks in Jenkins 2.567 and earlier (LTS 2.555.2 and earlier) expose user-specific configuration data to any authenticated user holding the Overall/Read permission. Authenticated low-privilege users can query the timezone setting of any Jenkins user account and enumerate the names of views configured within other users' 'My Views' personal dashboards, constituting an information disclosure weakness. No public exploit code exists and this CVE is not listed in the CISA KEV catalog at time of analysis, placing real-world risk in the low-moderate tier primarily relevant to multi-tenant or shared Jenkins deployments.
Missing authorization enforcement in Jenkins allows low-privileged authenticated users to cancel build queue items they lack permission to view. Affecting Jenkins 2.567 and earlier (LTS 2.555.2 and earlier), this CWE-862 flaw breaks the expected access control invariant that Item/Cancel should be constrained by Item/Read visibility. No public exploit code exists and this vulnerability is not listed in CISA KEV at time of analysis, but the low attack complexity and network accessibility make it a realistic threat in multi-tenant Jenkins deployments where fine-grained RBAC is in use.
Open redirect exploitation in Jenkins allows unauthenticated network attackers to craft login URLs that bypass the post-authentication redirect validation and forward users to attacker-controlled external sites. Affected are Jenkins weekly releases through 2.567 and LTS releases through 2.555.2, where the redirect URL legitimacy check fails when the URL contains relative path segments such as `./` or `../`. No active exploitation is confirmed (not listed in CISA KEV) and no public exploit code has been identified at time of analysis, placing this in a moderate-risk, phishing-enablement category rather than a direct system compromise class.
Denial of service in the image-size Node.js library through version 2.0.2 allows remote unauthenticated attackers to permanently hang the Node.js event loop by supplying a crafted JXL or HEIF image containing a box with a zero-valued size field. Publicly available exploit code exists and an upstream fix has been merged, making this a credible availability threat for any service that accepts user-supplied images and runs them through image-size. No active exploitation has been reported in CISA KEV at time of analysis.
Denial of service in the image-size Node.js library (versions up to and including 2.0.2) allows remote unauthenticated attackers to permanently stall the Node.js event loop by submitting a malformed ICNS image. The flaw stems from an infinite loop in the ICNS parser when an entry length field is zero, and publicly available exploit code exists per VulnCheck and an independent write-up; no public exploit identified at time of analysis indicates active exploitation, but the POC makes weaponization trivial.
Uncontrolled resource consumption in Spring Data Commons exposes applications to a remote Denial of Service condition when two specific features are active simultaneously. An unauthenticated remote attacker can send a specially crafted HTTP request to any endpoint backed by a @ProjectedPayload-annotated controller method, triggering unbounded memory allocation that exhausts the JVM heap and crashes the application. The vulnerability spans eight major release lines (2.7.x through 4.0.x), making the potential blast radius broad across Spring-based Java backends. No public exploit code or CISA KEV listing has been identified at time of analysis, and the CVSS attack complexity rating of High reflects non-trivial preconditions that meaningfully limit opportunistic exploitation.
SpEL (Spring Expression Language) injection in Spring Data KeyValue and Spring Data Redis allows a network-accessible, low-privileged attacker to execute arbitrary SpEL expressions when applications pass unsanitized user-controlled Sort parameters directly to repository query methods delegating to SpelPropertyComparator. Affected versions span eight major release lines from 2.7.x through 4.0.x, making the exposure surface broad across Spring-based Java ecosystems. No public exploit identified at time of analysis and no CISA KEV listing, but the high confidentiality impact rating and network attack vector warrant prompt patching for any application that surfaces Sort query parameters to end users.
Improper certificate validation in Spring AMQP allows a network-positioned attacker to perform a man-in-the-middle attack against applications that configure broker connections via RabbitConnectionFactoryBean.setUri("amqps://...") without explicitly calling setUseSSL(true). Affected are Spring AMQP versions 2.4.0 through 4.0.3 across four release lines. Although TLS encryption is established, the absence of certificate chain validation and hostname verification means a rogue broker can impersonate the legitimate RabbitMQ instance, exposing message content in transit. No public exploit identified at time of analysis, and this CVE is not listed in CISA KEV.
Denial of Service in Spring Data Commons affects applications that parse Sort parameters, triggering a StackOverflowException via crafted input. All major supported and legacy branches from 2.7.x through 4.0.x are affected, making the blast radius exceptionally broad across the Spring ecosystem. CVSS rates this Medium (5.9) with a high attack complexity qualifier (AC:H), unauthenticated network access (PR:N), and full availability impact (A:H); no active exploitation is confirmed and no public exploit code has been identified at time of analysis.
Predictable correlation IDs in Spring AMQP's RabbitTemplate.sendAndReceive() expose request-reply messaging to correlation hijacking when a fixed reply queue is configured. Affected are all widely used Spring AMQP branches from 2.4.x through 4.0.x - a broad install base across Java enterprise applications. A network-positioned attacker with high privileges can exploit the sequential counter to predict future correlation IDs, enabling interception or injection of reply messages into the shared fixed reply queue. No public exploit code exists and this vulnerability is not listed in CISA KEV; CVSS 4.4 Medium reflects real-world limitations from high privilege and complexity requirements despite the changed scope indicator.
Regex parameter binding in Spring Data MongoDB's @Query annotation fails to sufficiently quote user-supplied strings, enabling NoSQL injection that can manipulate query semantics. Unauthenticated remote attackers (PR:N, AV:N) who can influence the bound parameter in applications using this pattern may break out of the intended regular expression context to retrieve unintended documents, resulting in high-impact confidentiality compromise. No confirmed active exploitation or public exploit code has been identified at time of analysis; the CVSS Attack Complexity of High reflects a dependency on a specific application-level coding pattern.
Command injection in Ruby's net-imap gem allows arbitrary IMAP protocol command injection via the Net::IMAP#id and Net::IMAP#enable methods when untrusted input is passed as arguments. The #id method fails to prohibit CRLF sequences in ID field value strings (despite correctly quoting other specials), while #enable passes argument .to_s values verbatim without validating them as legal IMAP atoms. An attacker who can influence these inputs can terminate the current IMAP command and append arbitrary subsequent IMAP commands such as DELETE mailbox. Rated 'moderate' by the maintainer; no public exploit identified at time of analysis and no CISA KEV listing.
CRLF command injection in Ruby's net-imap gem (CVE-2026-47240) enables attackers to inject arbitrary IMAP commands when applications pass unvalidated user-controlled input to raw data arguments of search, fetch, sort, and thread methods. Affected versions (>= 0.6.0 through 0.6.4, and <= 0.5.14) fail to verify that the connected IMAP server supports non-synchronizing literals before sending them, meaning servers without LITERAL+, LITERAL-, or IMAP4rev2 capability may interpret the malformed literal boundary as a pipelined command rather than closing the connection - allowing injected payloads such as DELETE mailbox to execute. No public exploit has been identified at time of analysis and no CISA KEV listing exists, but the advisory is rated moderate severity by the upstream maintainers and patches are available.
Denial-of-service in the Linux kernel's DAMON (Data Access MONitor) subsystem allows a local low-privileged user to crash the kernel by supplying a non-power-of-two `min_region_sz` value through the DAMON sysfs interface and invoking `damon_start()`. The flaw is an incomplete remediation: commit c80f46ac228b fixed the same class of bug for the `damon_commit_ctx()` code path but left `damon_start()` - reachable via DAMON sysfs - without the `is_power_of_2()` guard. No public exploit exists and EPSS sits at 0.02% (4th percentile), indicating negligible real-world exploitation probability; no CISA KEV listing has been issued.
Memory leak in the Linux kernel's drm/nouveau subsystem exposes systems with NVIDIA GPUs using the open-source nouveau driver to local denial of service. When aperture_remove_conflicting_pci_devices() fails during PCI device probe, the error path bypasses the fail_nvkm cleanup label, permanently leaking both the nvkm_device wrapper and the pci_enable_device() reference taken inside nvkm_device_pci_new(). No public exploit exists and EPSS sits at the 5th percentile (0.02%), confirming negligible exploitation probability; however, patched stable releases are available across multiple kernel branches.
Improper end-of-filesystem boundary handling in the Linux kernel EROFS driver for file-backed mounts causes a denial-of-service condition when I/O requests exceed the filesystem image boundary. Affected kernels spanning multiple stable branches fail to zero-fill out-of-bounds I/O regions as expected (analogous to loopback device behavior), producing undefined kernel behavior that results in availability loss. No public exploit has been identified and EPSS is 0.02% (5th percentile), indicating negligible opportunistic exploitation probability; patches are confirmed available across stable branches 6.12.x, 6.18.x, 6.19.x, and 7.0.
Lock resource leak in Linux kernel hugetlbfs subsystem enables local denial of service against systems using huge page memory mappings. Introduced by a faulty commit (ea52cb24cd3f) that incorrectly handled VMA lock allocation during the mmap_prepare stage, the defect allows a failed post-mmap_prepare allocation to leave a VMA lock unreleased, leaking kernel resources. This CVE documents the corrective revert; systems running affected kernel versions (around 6.19 through pre-7.0.12 and pre-7.1-rc6) with hugetlbfs in use are exposed to potential availability impact. No public exploit code exists and EPSS probability is 0.02% (5th percentile), indicating very low real-world exploitation activity.
Double-free vulnerability in OpenSSL's OCSP stapled response verification allows remote attackers with low privilege to cause memory corruption affecting confidentiality, integrity, and availability. CVE-2026-35188 affects OpenSSL 4.0.0 (fixed in 4.0.1) and the 3.6.x branch (fixed in 3.6.3), and is one of seventeen CVEs addressed in the OpenSSL 4.0.1 security patch release published 2026-06-09. No public exploit and no active exploitation (CISA KEV) have been identified at time of analysis; however, the SSVC framework rates technical impact as 'total', flagging potential severity beyond the moderate CVSS base score.
Out-of-bounds read in OpenSSL 4.0.0's `X509_VERIFY_PARAM_set1_email()` function can crash applications performing email-based X.509 certificate verification when processing attacker-influenced email input, resulting in a denial-of-service condition. The vulnerability is scoped to OpenSSL 4.0.0 only and was patched in the June 9, 2026 security release (4.0.1), which bundled fixes for 18 CVEs. No public exploit identified at time of analysis and no CISA KEV listing.
Stale kernel memory disclosure in the Linux kernel's io_uring IORING_OP_WAITID operation allows a local low-privileged user to read arbitrary bytes from reused io_kiocb command storage via the userspace siginfo output buffer. The io_uring prep path for waitid requests fails to zero-initialize the io_waitid::info result field, meaning that when a wait completes without a child event the kernel copies whatever bytes happen to occupy that slab slot into the caller's siginfo_t. No public exploit has been identified at time of analysis; EPSS at 0.02% (5th percentile) reflects low current exploitation probability.
Server-side request forgery (SSRF) in Spring Framework's UriComponentsBuilder affects applications that use this API to parse and validate externally supplied URL strings. Incorrect host parsing allows a remote, unauthenticated attacker - with user interaction - to cause the application server to issue requests to unintended internal or external destinations, exposing low-level confidentiality and integrity impacts. No public exploit identified at time of analysis and no CISA KEV listing; however, SSRF in widely deployed Java frameworks warrants attention in any internet-facing application that processes user-controlled URLs.
Multipart request smuggling in Spring Framework's MVC and WebFlux components exposes applications to HTTP request manipulation via CWE-444. Unauthenticated remote attackers (AV:N/AC:L/PR:N/UI:N per CVSS) can exploit inconsistent multipart boundary parsing to smuggle malformed HTTP requests, achieving low-integrity impact against affected deployments. No public exploit code and no CISA KEV listing have been identified at time of analysis; however, the zero-prerequisite attack profile and broad version coverage across four major Spring branches (5.3.x, 6.1.x, 6.2.x, 7.0.x) make this relevant to any Java shop running Spring MVC or WebFlux with multipart upload handling enabled.
Spring Expression Language (SpEL) evaluation logic in Spring Framework 5.3.x through 7.0.x fails to enforce method invocation restrictions in read-only or restricted contexts, allowing remote unauthenticated attackers (CVSS PR:N, AV:N) to invoke arbitrary zero-argument methods and trigger unintended application logic. Scored LOW (3.7) with only availability impact per CVSS (A:L), though the 'Authentication Bypass' tag and CWE-863 (Incorrect Authorization) suggest the authorization bypass itself may have broader implications not fully reflected in the score. No public exploit identified at time of analysis and no CISA KEV listing, but the wide version range across four active Spring Framework release trains represents significant ecosystem exposure.
Security bypass in Spring WebFlux's Kotlin Router DSL (CWE-284: Improper Access Control) allows remote unauthenticated attackers to circumvent access control rules in Spring Framework 5.3.0 through 5.3.48. The CVSS vector (AV:N/AC:H/PR:N/UI:N) indicates network-reachable exploitation without authentication, though high attack complexity constrains opportunistic exploitation. No public exploit identified at time of analysis, and this vulnerability is not listed in the CISA KEV catalog.
Cross-site scripting (XSS) in Spring Framework's MVC JSP form tags allows unauthenticated remote attackers to inject arbitrary HTML or JavaScript into rendered pages by supplying malicious values through the cssClass, cssErrorClass, or cssStyle tag attributes. Applications across four active Spring Framework release lines (5.3.x through 7.0.x) are affected when they pass user-controlled input directly into these tag attributes. No public exploit code has been identified at time of analysis, and CISA has not listed this CVE in the Known Exploited Vulnerabilities catalog, but the broad installed base of Spring MVC in enterprise Java environments and the high confidentiality impact (session hijacking, credential theft) warrant prompt patching.
Cross-site scripting in Spring Framework versions 5.3.0-5.3.48, 6.1.0-6.1.27, 6.2.0-6.2.18, and 7.0.0-7.0.7 allows remote attackers to inject JavaScript into victim browsers when applications rely on JavaScriptUtils.javaScriptEscape() for output encoding. The flaw stems from incomplete escaping in this utility method, and successful exploitation requires user interaction (UI:R) such as visiting a crafted page. No public exploit has been identified at time of analysis and the issue is not in CISA KEV.
Open redirect in Spring Framework (Spring MVC and Spring WebFlux) across four major version branches enables unauthenticated remote attackers to craft URLs that cause the application to issue a 302 HTTP redirect to an arbitrary attacker-controlled external host. The vulnerability is conditionally exploitable - requiring a catch-all wildcard route mapping without an explicit view name - and demands user interaction to trigger. CVSS rates this 4.2 Medium (AV:N/AC:H/PR:N/UI:R); no public exploit code or CISA KEV listing has been identified at time of analysis.
Denial of Service in Spring WebFlux's multipart request processing allows unauthenticated remote attackers to exhaust server resources across all supported Spring Framework branches. Affects Spring Framework 5.3.x through 7.0.x when applications use the reactive WebFlux stack and expose endpoints that accept multipart data. No public exploit code has been identified at time of analysis, and CISA KEV listing is absent, but the network-reachable, zero-privilege attack surface warrants prompt patching for internet-facing WebFlux deployments.
Session fixation in Spring Framework's WebFlux reactive stack (versions 5.3.x through 7.0.x) enables a remote attacker to hijack an authenticated user's session by leveraging a compromised subdomain - typically via cross-site scripting - to plant a known session ID and exchange it for the victim's authenticated session post-login. The attack is classified as CWE-384 and requires both a prior subdomain compromise and user interaction, placing real-world exploitability well below the headline concern for most deployments. No public exploit code and no CISA KEV listing have been identified at time of analysis.
Unbounded HTTP/2 stream creation in Netty's netty-codec-http2 library exposes any Netty HTTP/2 server running on default configuration to memory exhaustion from a single TCP connection. Because DefaultHttp2Connection.DefaultEndpoint initializes stream limits to Integer.MAX_VALUE and Http2Settings never advertises SETTINGS_MAX_CONCURRENT_STREAMS unless the application explicitly calls initialSettings().maxConcurrentStreams(n), a remote unauthenticated attacker can sustain hundreds of thousands of simultaneous streams, each forcing JVM heap allocations for DefaultStream objects, PropertyMap slots, flow-controller state, and IntObjectHashMap entries. This misconfiguration is also the structural precondition for CVE-2023-44487-style HTTP/2 Rapid Reset amplification. No public exploit has been identified at time of analysis; vendor-released patches are available in 4.1.135.Final and 4.2.15.Final.
DNS cache poisoning in Netty's resolver component (io.netty:netty-resolver-dns) enables remote unauthenticated attackers to redirect downstream application traffic to attacker-controlled IPs through a Kaminsky-style spoofing attack. The vulnerability combines two compounding weaknesses present in the default configuration: DNS transaction IDs shuffled with a mathematically predictable Linear Congruential Generator (ThreadLocalRandom), and a static UDP source port resulting from the default ChannelPerResolver channel strategy. No public exploit has been identified at time of analysis and the issue is not listed in CISA KEV, but the Netty project rated it high severity and released fixes in versions 4.1.135.Final and 4.2.15.Final.
File descriptor exhaustion in Netty's Unix domain socket native transport allows a local peer to leak two file descriptors per crafted SCM_RIGHTS message into the receiving process, with neither FD ever closed. Affected are applications explicitly configured with DomainSocketReadMode.FILE_DESCRIPTORS on Epoll or KQueue DomainSocketChannel - a non-default opt-in setting. Sustained message flooding by a local socket peer can exhaust the Netty process's file descriptor table, ultimately causing denial of service. No public exploit identified at time of analysis and not listed in CISA KEV; CVSS scores this at 4.0 (Low) reflecting the local-only attack surface and low availability impact.
Unauthenticated open redirect in Authlib's OAuth 2.0 authorization endpoint allows any remote attacker to weaponize a trusted authorization server domain as a redirector - no client registration, no authenticated session, and no prior state required. The flaw affects all deployments of pip/authlib < 1.6.10 and == 1.7.0 running an authorization endpoint via the Flask or Django integrations. Publicly available exploit code exists and was dynamically confirmed against the live HEAD (v1.6.6-104-g68e6ab3f); no public exploit identified at time of analysis as confirmed actively exploited (CISA KEV status absent), but the trivial, zero-prerequisite exploit path makes phishing and domain-allowlist bypass practical at scale.
Infinite loop denial-of-service in the Linux kernel's drm/v3d GPU driver allows a local low-privileged user to permanently peg a CPU core by submitting a crafted self-referential ioctl extension structure. The vulnerability exists in v3d_get_extensions(), which traverses a userspace-controlled linked list without bounding chain length; when both in_sync_count and out_sync_count are zero, the duplicate-extension guard silently passes on every iteration, trapping the kernel thread indefinitely. No public exploit or active exploitation has been identified at time of analysis; EPSS stands at 0.02%, consistent with the local-access-only attack surface.
Error pointer dereference in the Linux kernel's Intel IPU6 media driver crashes the kernel during a failed probe, resulting in local denial of service. The flaw resides in `ipu6_pci_probe()` at `drivers/media/pci/intel/ipu6/ipu6.c:690`, where `isp->psys` holds an ERR_PTR value rather than NULL in a specific error path, causing it to be dereferenced during cleanup rather than handled safely. No public exploit exists and EPSS sits at 0.02% (5th percentile), reflecting negligible real-world exploitation interest.
Kernel warning (WARN_ON) triggerable in the Linux videobuf2 DMA scatter-gather subsystem on Apple Silicon systems allows a low-privileged local user to destabilize video capture pipelines by performing mmap() on an imported DMA-buf. The vb2_dma_sg_mmap function fails to set VM_DONTEXPAND and VM_DONTDUMP VMA flags - protections that vb2_dma_contig correctly applies - causing drm_gem_mmap_obj() to hit its assertion guard. No public exploit has been identified at time of analysis; EPSS is 0.02% (5th percentile), consistent with the narrow hardware-specific trigger surface.
NULL pointer dereference in the Linux kernel's Renesas VSP1 media driver crashes the kernel during module unload on Generation 4 Renesas SoC hardware. The defect exists because the cleanup path unconditionally invokes vsp1_drm_cleanup() rather than vsp1_vspx_cleanup() for gen 4 IP versions, while the initialization path correctly checks the IP version - an asymmetry introduced when gen 4 support was added. With an EPSS of 0.02% (4th percentile), no KEV listing, and no public exploit code, real-world exploitation risk is very low and confined to niche embedded and automotive platforms; vendor-released patches are available in stable kernel branches.
Local denial-of-service in the Linux kernel's SELinux subsystem allows any low-privileged local process to monopolize the `/sys/fs/selinux/policy` file descriptor, blocking all other processes from reading the active kernel policy. The root cause is a `policy_opened` flag that enforced a single-open restriction on the SELinux policy pseudofile - originally intended to prevent inconsistent reads or unbounded kernel memory allocation, but which also created a trivial resource-starvation primitive. No active exploitation has been identified (not in CISA KEV), and the EPSS score of 0.02% places this in the 5th percentile for exploitation likelihood.
Deadlock vulnerability in the Linux kernel's pseries/papr-hvpipe subsystem allows a local low-privileged attacker on IBM pSeries Power Architecture systems to cause a kernel deadlock by triggering a race between the ioctl or release handler and a hardware interrupt firing on the same CPU. Exploitation is architecture-specific, requiring IBM PAPR hardware with the hvpipe driver loaded, and results in an availability-only denial-of-service. No public exploit exists and EPSS probability stands at 0.02%, consistent with a narrow attack surface; vendor-released patches are confirmed available for Linux 7.0.7, 7.1-rc3, and the 6.18 stable branch.
Improper interrupt registration in the Linux kernel's libwx network driver causes a kernel WARNING when VF (Virtual Function) misc interrupts are initialized, resulting in high availability impact on affected systems. The libwx driver incorrectly calls request_threaded_irq() with a primary handler, a NULL threaded handler, and the IRQF_ONESHOT flag - an invalid combination flagged as a WARNING by the kernel interrupt management subsystem since commit aef30c8d569c. Exploitation requires local access with low privileges on a system equipped with a Wangxun VF NIC using the libwx driver; no public exploit code exists and EPSS is negligible at 0.02%.
NULL pointer dereference in the Linux kernel's s3c64xx SPI driver triggers a kernel crash on driver unbind, allowing a local low-privileged attacker to cause a denial of service. The flaw affects systems equipped with Samsung S3C64xx-series SoC hardware running unpatched kernel versions from 6.0 onward. No public exploit code exists and EPSS probability sits at 0.02%, but the crash is deterministic once the driver unbind sequence is triggered on vulnerable hardware. No active exploitation (CISA KEV) has been identified.
KVM/x86 on Linux exposes a race condition in interrupt request register (IRR) scanning that causes incorrect reporting of the highest pending interrupt in nested virtualization scenarios. Specifically, when PID.ON is set by a sender vCPU after the target vCPU has already harvested and cleared the PIR, a second call to vmx_sync_pir_to_irr() observes an empty PIR and returns max_irr=-1 without scanning the IRR, producing a spurious kernel WARNING in vmx_check_nested_events() and a wasted L2 VM-Enter/VM-Exit cycle. The interrupt itself is not lost - it resides in the IRR from the first sync - but the incorrect max_irr triggers the assertion. No public exploit exists and EPSS is 0.02%, consistent with the highly localized, timing-dependent nature of this defect.
Missing resource cleanup in the Linux kernel's generic power domain (genpd) subsystem causes runtime PM to remain enabled for virtual devices after detach, leading to a NULL pointer dereference in genpd_runtime_suspend() and local denial of service. The flaw affects systems where drivers use genpd_dev_pm_attach_by_id() - predominantly ARM and SoC-based platforms - since the balancing pm_runtime_disable() call is absent from genpd_dev_pm_detach(). No public exploit exists and EPSS at 0.02% (5th percentile) confirms negligible exploitation probability; this is not listed in CISA KEV.
Sensitive HMAC key material is exposed through kernel debug logging in the Linux kernel's CAAM (Cryptographic Acceleration and Assurance Module) crypto driver. On systems with CONFIG_DYNAMIC_DEBUG enabled and debug output activated for the caam module at runtime, calls to hash_digest_key() emit raw HMAC key bytes via hex dump into the kernel log, allowing any local attacker with debug log read access to extract cryptographic secrets. No public exploit code has been identified and EPSS probability is 0.02%, though the confidentiality impact is severe if exploitation conditions are met. The provided CVSS vector (C:N/A:H) appears to misclassify this as an availability issue rather than a confidentiality issue - see confidence notes.
Linux kernel x86 EFI runtime service fault handling panics instead of recovering gracefully on systems with buggy EFI firmware, following the FPU softirq changes introduced in commit d02198550423. When kernel_fpu_begin() began using local_bh_disable() in place of preempt_disable(), SOFTIRQ_OFFSET is set in preempt_count for the duration of EFI runtime service calls, causing in_interrupt() to return true in normal task context - which causes efi_crash_gracefully_on_page_fault() to bail unconditionally, escalating to panic(). On hardware where EFI firmware (e.g., GetTime()) accesses unmapped memory, this produces an unrecoverable system freeze. No public exploit identified at time of analysis; EPSS is 0.02%, confirming no observed exploitation activity.
Array out-of-bounds read in the Linux kernel's Qualcomm Light Pulse Generator (qcom-lpg) LED driver allows a local low-privileged user to crash the kernel via denial of service on affected Qualcomm SoC-based systems. The flaw stems from using FIELD_GET() to extract a 3-bit register value (range 0-7) as an index into an array with only 5 elements, enabling reads beyond array bounds and subsequent hardware misconfiguration or kernel panic. No public exploit has been identified and EPSS stands at the 5th percentile, placing this firmly in lower-priority triage despite the technically High availability impact per CVSS.
NULL pointer dereference in the Linux kernel's HugeTLB early boot parameter parser causes a system crash before the OS fully initializes. Specifically, `hugetlb_add_param()` in `mm/hugetlb` dereferences a NULL pointer via `strlen()` when `hugepages`, `hugepagesz`, or `default_hugepagesz` kernel command line parameters are supplied without a '=' separator, producing a boot-time kernel panic. The impact is a complete denial of service - the system fails to boot until the malformed parameter is corrected. No public exploit or CISA KEV listing exists; EPSS stands at 0.02% (5th percentile), reflecting very low observed exploitation activity. Patches are available in Linux 6.18.27, 7.0.4, and 7.1-rc1, with Ubuntu distributing fixes via USN-8489-1 and USN-8488-1.
Sensitive cryptographic material-HMAC session keys, nonces, and passphrase data held in struct tpm2_auth-is left in freed slab memory when a TPM device is released via tpm_dev_release() in Linux kernels from 6.10 onward, because that code path uses kfree() rather than kfree_sensitive(). Every other tear-down path in the same TPM driver (tpm2_end_auth_session() and tpm_buf_check_hmac_response()) already calls kfree_sensitive(), making this an isolated inconsistency. No public exploit exists and EPSS is 0.02% (5th percentile), but in high-assurance environments where TPM-backed key confidentiality is mandatory the residual key material is a meaningful exposure until overwritten by subsequent allocations.
NULL pointer dereference in the Linux kernel's admv1013 IIO frequency driver crashes the kernel when device_property_read_string() fails during device initialization, leaving a garbage pointer subsequently passed to strcmp(). A local low-privileged user on a system with ADMV1013 microwave upconverter hardware can reliably trigger a kernel panic, resulting in a complete denial of service. No public exploit exists and EPSS sits at 0.02% (5th percentile), but vendor-released patches are available across multiple stable kernel branches including 6.12.86 and 7.0.4.
NULL pointer dereference in the Linux kernel's drm/imagination (PowerVR GPU) driver crashes the kernel when a local user writes to the ftrace mask debugfs entry. Affected kernels range from the introduction of the drm/imagination driver at commit a331631496a0af9a6f4e7e1860983afd8b1bb013 through Linux 7.0, with fixes landed in 7.0.4 and 7.1-rc2. An attacker with local access and write permission to the debugfs interface can trigger a kernel Oops, resulting in a full system crash and denial of service. No public exploit exists and EPSS is 0.02%, consistent with the narrow hardware scope (Imagination Technologies PowerVR GPUs) and local-only attack surface.
Kernel panic in the Linux amdgpu DRM driver exposes systems running RDNA4 hardware (e.g., AMD RX 9070 XT) to a denial-of-service condition during driver initialization, but only when the kernel is compiled with CONFIG_DRM_DEBUG_MM enabled. The RDNA4 architecture (GFX 12) removed the GDS, GWS, and OA on-chip memory resources entirely; however, amdgpu_ttm_init() unconditionally invokes amdgpu_ttm_init_on_chip() for these resources regardless of size, ultimately calling drm_mm_init(mm, 0, 0), which triggers DRM_MM_BUG_ON and crashes the kernel at modprobe time. No public exploit exists and EPSS sits at 0.02% (7th percentile), consistent with a hardware-specific, debug-config-dependent crash rather than a remotely triggerable attack surface.
Open redirect in WebOb (pip/webob <= 1.8.9) enables unauthenticated network attackers to redirect victims to arbitrary attacker-controlled domains by bypassing the prior CVE-2024-42353 patch. The bypass exploits Python 3.10+'s silent stripping of ASCII control characters (tab, CR, LF) from URLs before parsing, so a crafted path like /\t/attacker.com passes the previous // guard, then gets normalized by urlsplit into a protocol-relative URL pointing to attacker.com. No public exploit has been identified at time of analysis beyond the advisory's own proof-of-concept code, but the bypass technique is trivial and the full exploit path is published in the GitHub advisory GHSA-fh3h-vg37-cc95.
Path traversal in Docling's LaTeX backend (pip/docling versions 2.73.0 through 2.90.x) allows an attacker who supplies a crafted LaTeX document to read arbitrary files accessible to the conversion process via the \includegraphics, \input, and \include command handlers. With a high confidentiality impact (C:H) but local attack vector and required user interaction (AV:L/UI:R per CVSS), the practical risk is concentrated in automated document-processing pipelines or services that ingest untrusted .tex files. No public exploit identified at time of analysis, and the vulnerability is not listed in CISA KEV. A vendor-released patch exists as of version 2.91.0.
Improper authorization in Apache Kafka 4.0.0-4.3.0 arises from a discrepancy between the documented ACL requirement and the actual runtime behavior of the CONSUMER_GROUP_DESCRIBE (API key 69) endpoint. The API checks for DESCRIBE permission on the GROUP resource at runtime, while official Kafka documentation and KIP-848 specify READ as the required operation - causing administrators who followed the documentation to configure ACLs that either over-grant READ access to users who should only observe group metadata, or under-restrict DESCRIBE-only users who can nonetheless access sensitive consumer group state. No public exploit has been identified at time of analysis, and EPSS exploitation probability is 0.02% (4th percentile), indicating negligible opportunistic exploitation risk.
Denial-of-service in Spring Cloud Function allows a low-privileged actor with physical access and required user interaction to exhaust JVM heap memory by registering an unbounded number of functions into the Function Registry, triggering an Out-of-Memory (OOM) crash. Five actively maintained release lines (3.2.x through 5.0.x) plus all unsupported versions are affected. No public exploit code exists and the vulnerability is not listed in CISA KEV at time of analysis; the CVSS score of 5.7 reflects the physical access vector and user interaction dependency that substantially constrain real-world exploitability compared to the High availability impact alone.
Uncontrolled recursion in Spring Cloud Function's routing layer triggers an Out-of-Memory (OOM) condition, resulting in a denial of service across all active release trains (3.2.x through 5.0.x). The vulnerability is rooted in CWE-674 (Uncontrolled Recursion), where the routing layer lacks depth limits or cycle detection, allowing recursive request-handling to exhaust JVM heap space. No public exploit code or CISA KEV listing has been identified at time of analysis; however, the CVSS physical attack vector (AV:P) appears inconsistent with a routing-layer vulnerability and warrants verification against the Spring/VMware advisory.
OpenShift Container Platform exposes a resource exhaustion path where low-privileged authenticated users with pod creation rights can trigger cluster-wide API server degradation by exploiting two quota enforcement gaps: completed pods with restartPolicy:Never are excluded from ResourceQuota pod limits, and Kubernetes events are not quota-scoped. By repeatedly creating such short-lived pods, an attacker causes Kubernetes events to accumulate unboundedly in etcd, degrading API server performance across the entire cluster - a scope change (CVSS S:C) that extends impact well beyond the attacker's own namespace. No public exploit has been identified at time of analysis, and the vulnerability is not listed in the CISA KEV catalog.
Unauthenticated information disclosure in Apache ActiveMQ Broker allows remote attackers to enumerate all durable topic subscriptions - including client identifiers, subscription names, topic destinations, and JMS selector expressions - by sending a BrokerInfo command to a broker with syncDurableSubs enabled on a network connector. The broker incorrectly skips authentication before servicing the BrokerInfo request, exposing sensitive messaging infrastructure metadata. No public exploit identified at time of analysis, and EPSS stands at 0.02% (6th percentile), indicating very low current exploitation probability despite network-reachable attack vector.
Incomplete authorization in Apache ActiveMQ Broker allows authenticated low-privilege users to remove messaging destinations (queues and topics) beyond their granted permissions, causing targeted availability disruption to broker-connected producers and consumers. Affected versions span the 5.x line before 5.19.7 and the 6.x line from 6.0.0 before 6.2.6, covering ActiveMQ Broker and ActiveMQ All distributions. With an EPSS of 0.01% (3rd percentile), no CISA KEV listing, and no public exploit code identified, this is a low-urgency but genuine authorization control gap that is most relevant to environments with multiple untrusted authenticated broker accounts.
Server-Side Request Forgery in Clair's fetcher component exposes internal network services and cloud metadata endpoints to unauthenticated remote attackers who can submit container image manifests with crafted layer descriptor URIs. Affected deployments are those where PSK (Pre-Shared Key) authentication is not configured - an opt-in control that is not enforced by default - meaning standalone or custom Clair installations without PSK are directly exploitable over the network with no credentials. Reflective SSRF behavior leaks up to 256 bytes of internal error body content per request via CheckResponse error messages, enabling network reconnaissance; no public exploit has been identified at time of analysis and the vulnerability is not listed in CISA KEV.
Panic-induced denial of service in the golang.org/x/image/bmp decoder allows remote unauthenticated attackers to crash any Go application that processes untrusted BMP files. When a paletted BMP containing an out-of-range palette index is decoded, the image package accesses invalid memory and triggers a Go runtime panic, terminating the affected goroutine or process. No public exploit exists at time of analysis, but the attack is automatable per SSVC and requires no privileges or user interaction (CVSS AV:N/AC:L/PR:N/UI:N), making any internet-exposed BMP processing endpoint an accessible target.
Excessive RBAC permissions in the Metal3 ip-address-manager (IPAM) controller's ClusterRole expose Kubernetes Secrets to post-compromise abuse. The metal3-ipam-controller-manager-role granted full CRUD on core/v1 Secrets - yet PR diff evidence confirms the controller never accesses Secrets in normal operation, making these permissions entirely unnecessary. If the controller pod is compromised via a supply chain attack or container escape, an attacker inheriting the controller's service account token can read, modify, or delete any Secret in the namespace, turning a narrow container compromise into cluster-wide credential exfiltration. No public exploit exists and no CISA KEV listing, but the confidentiality impact is rated High (C:H) by NVD.
Host process data leakage in vm2 (npm/vm2 <= 3.11.3) allows untrusted JavaScript executing inside a NodeVM sandbox to read sensitive host-process state - including HTTP Authorization headers, session tokens, and AsyncLocalStorage request context - by requiring process-wide observability builtins that were incorrectly absent from the DANGEROUS_BUILTINS denylist. The affected modules (diagnostics_channel, async_hooks, perf_hooks) are Node.js process-wide singletons, not sandbox-local constructs, so granting sandbox access to them exposes all host process data flowing through those channels. A publicly available exploit code exists confirming extraction of Bearer tokens and x-session-token headers from live host HTTP requests; this is not confirmed actively exploited (not in CISA KEV).
CPU and log amplification in opentelemetry-go's baggage parser allows remote unauthenticated attackers to trigger denial-of-service by submitting oversized or malformed W3C baggage headers to any instrumented Go service. PR #7880 inadvertently removed the upfront raw-length check and per-member size guard from `baggage.Parse`, meaning the parser now fully tokenizes and percent-decodes every member of an arbitrarily large input rather than rejecting it early. A publicly available proof-of-concept exists (GHSA-5wrp-cwcj-q835); the vulnerability is not yet listed in CISA KEV, and impact is bounded to availability with no confirmed confidentiality or integrity consequence.
Uncontrolled resource consumption in PyJWT 2.8.0-2.12.1 exposes any service that verifies detached JWS tokens to unauthenticated denial-of-service. When the unencoded-payload extension (b64=false, RFC 7797) is in use, PyJWT unnecessarily Base64URL-decodes the compact-serialization payload segment before discarding it in favor of the caller-supplied detached payload - turning that segment into an attacker-controlled amplifier for CPU and memory exhaustion regardless of signature validity. No public exploit has been identified at time of analysis, but the CVSS vector (AV:N/AC:L/PR:N/UI:N) confirms fully unauthenticated remote exploitation against any affected endpoint using this feature.
PyJWKClient in PyJWT prior to 2.13.0 passes attacker-influenced URIs directly to Python's urllib.request.urlopen() without restricting URI schemes, enabling Server-Side Request Forgery (SSRF) across file://, FTP, and data-URI schemes against applications that accept untrusted jku values. Affected deployments include any Python application using PyJWKClient where the jku URL originates from a JWT header, OAuth flow parameter, or externally influenced configuration. No public exploit exists and no CISA KEV listing is present; real-world exploitation is constrained by a CVSS-confirmed high attack complexity (AC:H) and required user interaction (UI:R), making opportunistic mass exploitation unlikely.
Memory exhaustion in pypdf prior to 6.12.0 allows an attacker who supplies a crafted PDF to cause large memory consumption in any application that processes it using layout mode text extraction. The vulnerability is triggered by PDFs containing text positioning operators with abnormally large x- or y-coordinate offsets, causing the library to allocate unbounded whitespace and newline characters during rendering. No confirmed active exploitation exists (not in CISA KEV), and SSVC rates this as non-automatable with partial technical impact, placing it in a lower operational priority tier despite the straightforward exploitation mechanic.
Denial-of-service via algorithmic complexity in pypdf before 6.12.0 allows an attacker who can supply a crafted PDF file to cause excessive processing time during cross-reference stream parsing. The vulnerability is triggered by crafting a PDF with /W [0 0 0] field values in a cross-reference stream combined with a large /Size value, which causes the library to perform unbounded iteration over zero-byte entries. No public exploit code has been identified at time of analysis, and this vulnerability is not listed in the CISA KEV catalog; however, any application that processes untrusted PDF input using pypdf is exposed.
Memory exhaustion in pypdf's XMP metadata parser allows denial of service via specially crafted PDF files containing oversized or element-dense XMP blocks, affecting all versions prior to 6.12.1. The vulnerability stems from an absence of input limits in the XML-based XMP parsing subsystem (CWE-770), meaning processing a malicious PDF can consume unbounded system memory. No public exploit code has been identified at time of analysis, and no confirmed active exploitation exists; however, the patch diff is publicly visible on GitHub, making trivial exploit construction feasible.
Out-of-bounds write in bzip2's bzip2recover utility allows a local attacker to supply a specially crafted file that triggers an off-by-one error, corrupting a global buffer and crashing the process. Per the CVSS 4.0 vector (AV:L/AC:L/AT:N/PR:N/UI:N), the attack requires no privileges and no user interaction beyond the utility being invoked against a malicious file. Impact is strictly denial of service against the bzip2recover process - no confidentiality or integrity exposure - and the CVSS 4.0 score of 5.1 (Medium) reflects this constrained scope. No public exploit or active exploitation has been identified at time of analysis.
Runtime PM reference count leak in the Linux kernel's OmniVision OV5647 camera sensor driver (media/i2c/ov5647) causes availability loss for systems equipped with this camera hardware. The s_ctrl function's handling of three V4L2 controls - AUTOGAIN, EXPOSURE_AUTO, and ANALOGUE_GAIN - returns early without invoking pm_runtime_put(), allowing unpaired runtime PM get/put calls to accumulate. A local user with access to the camera device node can trigger this imbalance repeatedly, exhausting the PM reference count and preventing the device from entering low-power states, ultimately making it unavailable. No public exploit has been identified; EPSS is 0.02% (5th percentile) and this vulnerability does not appear in CISA KEV.
Availability impact in the Linux kernel's xbox_remote media/rc driver allows a local, low-privileged user with access to the affected device to crash the kernel via a DMA coherency violation. The IO buffer used for USB transfers is embedded directly within the device structure, which violates DMA coherency rules and can trigger memory corruption leading to kernel panic. No public exploit has been identified at time of analysis, and the EPSS score of 0.02% reflects minimal observed exploitation activity. Vendor-released patches are available across multiple stable kernel branches.
Null pointer dereference in the Linux kernel's saa7164 media driver can crash the kernel when ioremap() fails during PCI device initialization. Systems with SAA7164-based PCIe capture cards (e.g., Phillips/NXP SAA7164) running unpatched kernel versions from 2.6.32 through stable branches prior to 6.6.140, 6.12.90, 6.18.32, and 7.0.9 are affected. A local, low-privileged attacker who can trigger driver initialization under memory pressure conditions may cause a kernel oops or panic, resulting in denial of service. No public exploit exists and EPSS is 0.02% (5th percentile), indicating negligible real-world exploitation probability at this time.
NULL pointer dereference in the Linux kernel's batman-adv Bridge Loop Avoidance (BLA) subsystem enables local denial of service via kernel panic. The race condition exists in batadv_bla_purge_claims(), which traverses the claim hash list under rcu_read_lock() without accounting for concurrent claim releases - when batadv_claim_release() NULLs the backbone_gw pointer mid-traversal, a subsequent call to batadv_bla_claim_get_backbone_gw() on the partially-freed claim triggers a NULL dereference. Exploitation requires local low-privilege access on systems actively running batman-adv with BLA enabled; no active exploitation is confirmed and EPSS stands at 0.02% (5th percentile).
Reference counting failure in the Linux kernel's batman-adv Bridge Loop Avoidance (BLA) subsystem allows a local low-privileged user to cause a slow kernel memory leak and eventual denial of service. Specifically, batadv_bla_add_claim() omits the required batadv_backbone_gw_put() call on the error path when a claim hash insertion fails, preventing the backbone_gw object's reference count from ever reaching zero. The vulnerability has been present since at least kernel 4.7 and is patched across multiple stable branches; no public exploit exists and EPSS is extremely low at 0.02% (5th percentile), placing this firmly in the low-urgency tier except for environments actively running batman-adv with BLA.
Stale VRAM exposure in the Linux kernel's amdkfd driver (drm/amdkfd) allows local authenticated users to observe prior occupants' GPU memory contents and crash multi-GPU compute workloads. The KFD allocation path omitted the AMDGPU_GEM_CREATE_VRAM_CLEARED flag - present in every other userspace GEM allocation path - leaving freshly allocated VRAM buffers populated with data from previous processes. The primary documented availability impact is deterministic crashes in RCCL peer-to-peer transport when stale values corrupt the ptrExchange, head, and tail protocol fields; a secondary information-disclosure risk exists because stale page-table remnants are observable by unprivileged compute kernels. No public exploit identified at time of analysis; EPSS is 0.02% (5th percentile).
Memory leak in the Linux kernel's spi: ch341 USB-to-SPI driver allows a local user to degrade system availability by triggering driver unbind without physical device disconnection. The defect lies in incorrect devres (device-managed resource) lifetime scoping - resources are anchored to the parent USB device rather than the USB interface, so they are not released when the driver unbinds during events such as probe deferral or runtime configuration changes. No public exploit has been identified at time of analysis, and an EPSS score of 0.02% (4th percentile) indicates negligible exploitation probability in the near term.
Improper driver teardown ordering in the Linux kernel's Freescale (FSL) SPI controller driver allows a local low-privileged user to trigger a denial-of-service condition during driver unbind. The SPI controller is not deregistered before underlying resources such as DMA are released, creating a window where the controller may reference freed memory. No public exploit has been identified at time of analysis, and EPSS sits at 0.02% (5th percentile), reflecting low real-world exploitation probability.