Btrfs transaction aborts in the Linux kernel allow local low-privileged users to crash the filesystem by triggering a logic defect in DUP chunk allocation that generates overlapping physical address ranges in the chunk map. Systems running btrfs with DUP metadata profiles - the default for single-device btrfs deployments - can encounter EEXIST (-17) errors in insert_dev_extents() during btrfs_create_pending_block_groups(), causing the transaction to abort and the filesystem to enter an error state. No public exploit or active exploitation (CISA KEV) has been identified; with an EPSS of 0.02% (4th percentile), this is a kernel reliability defect of operational concern to btrfs operators rather than a traditional attack vector.
NULL pointer dereference in the Linux kernel's ovpn (in-kernel OpenVPN) TCP socket handling causes a local denial of service via kernel crash. The race condition - between keepalive-driven peer release and concurrent userspace socket closure via tcp_close() - allows a low-privileged local user to trigger a kernel crash when ovpn attempts to dereference a NULL sk->sk_socket pointer during socket detachment. No public exploit has been identified and EPSS stands at 0.02% (4th percentile), reflecting narrow real-world exploitability constrained by the specific configuration and timing required.
Memory leak in the Linux kernel's AMD XDnA accelerator driver (accel/amdxdna) allows a local low-privileged user to degrade system availability by exhausting kernel memory. The amdxdna_ubuf_map() function fails to release previously allocated scatter-gather (sg) and internal sg table memory when error paths are taken during sg_alloc_table_from_pages or dma_map_sgtable operations. No active exploitation is confirmed (absent from CISA KEV), EPSS stands at 0.02% (4th percentile), and impact is strictly limited to availability - no confidentiality or integrity exposure exists.
Deadlock in the Linux kernel mlx5e Mellanox/NVIDIA Ethernet driver allows a low-privileged local attacker to hang the system by triggering network health reporter recovery paths that acquire locks in the wrong order. Specifically, work handlers acquire the netdev instance lock before invoking devlink_health_report, which then attempts to acquire the devlink lock - reversing the mandated devlink → rtnl → netdev ordering and producing an ABBA deadlock. The vulnerability affects systems equipped with Mellanox/NVIDIA ConnectX NICs; no public exploit exists and EPSS sits at 0.02% (4th percentile), consistent with a kernel-internal locking race rather than an externally triggerable flaw.
Divide-by-zero kernel panic (Oops) in the Linux kernel MPTCP subsystem's `mptcp_rcvbuf_grow()` function can be triggered by a local authenticated user under a rare but specific race condition involving concurrent out-of-order packet arrival and receive buffer initialization. The vulnerability also causes a secondary effect where the MPTCP receive buffer slowly drifts toward the `tcp_rmem[2]` maximum, degrading system performance on MPTCP-heavy workloads. No public exploit identified at time of analysis; EPSS is 0.02% (4th percentile), consistent with its local-only, race-dependent nature.
Memory leak in the Linux kernel's af_unix subsystem allows a local low-privileged user to exhaust kernel memory by repeatedly triggering a failure path in unix_stream_connect(). When prepare_peercred() fails after unix_create1() has already allocated a new socket object (newsk), the error path omits the required unix_release_sock() call, leaving kernel memory permanently unreleased. No public exploit has been identified at time of analysis; EPSS at 0.02% (4th percentile) reflects negligible widespread exploitation likelihood, consistent with the local-only attack vector.
Error handling failure in the Linux kernel's arm64 Guarded Control Stack (GCS) subsystem allows a local low-privileged user on ARMv9 hardware to trigger a kernel denial of service by exploiting an incorrect NULL check in arch_set_shadow_stack_status(). Because alloc_gcs() propagates do_mmap() failures as error-encoded pointers rather than NULL, the existing guard is bypassed and the kernel proceeds to use an invalid GCS address, risking a kernel panic. No public exploit exists and EPSS sits at the 4th percentile, but the vulnerability is confirmed patched in Linux 6.18.14, 6.19.4, and 7.0.
Kernel panic in the Linux inside-secure/eip93 hardware crypto driver occurs when the driver teardown routine unconditionally unregisters all cryptographic algorithms, including those never registered because the underlying EIP93 silicon does not implement them. Platforms with partial EIP93 silicon support - where only a subset of algorithms are burned into hardware - trigger the panic during module removal or system shutdown. No public exploit exists and EPSS sits at the 4th percentile (0.02%), indicating this is primarily a stability defect with negligible exploitation interest rather than a targeted attack surface.
Null pointer dereference in Wireshark's ROHC protocol dissector causes application crashes across two active release branches, constituting a denial-of-service condition. Affected versions span Wireshark 4.6.0 through 4.6.5 and 4.4.0 through 4.4.15; patched releases 4.6.6 and 4.4.16 are available per the vendor advisory wnpa-sec-2026-51. The attack vector is local with required user interaction (CVSS AV:L/UI:R), meaning exploitation requires a victim to open a specially crafted packet capture file - no remote or automated exploitation path exists, and no public exploit code or active exploitation has been identified at time of analysis.
Denial of service in IBM Db2 11.5.x and 12.1.x allows a low-privileged local user to crash the database engine by executing a specially crafted query against range partitioned tables. The vulnerability stems from uncontrolled resource allocation (CWE-770) during query processing, resulting in complete availability loss with no impact to confidentiality or integrity. No public exploit code exists and this vulnerability has not been listed in the CISA KEV catalog at time of analysis.
Denial of service in IBM Db2 versions 11.5.0-11.5.9 and 12.1.0-12.1.4 allows a locally authenticated, low-privileged user to crash the database service by executing a specially crafted SQL query against an instance configured with a small statement heap. The vulnerability stems from uncontrolled resource consumption (CWE-400) during query processing, resulting in high availability impact with no confidentiality or integrity exposure. No public exploit code and no active exploitation have been identified at time of analysis; SSVC classifies exploitation status as none.
Sensitive information disclosure in IBM App Connect Enterprise 13.0.1.0 through 13.0.7.0 exposes potentially sensitive data via log files accessible to local users. The CVSS vector (AV:L/PR:L) confirms exploitation requires local, low-privileged authenticated access, limiting the attack surface to users already present on the system. No public exploit has been identified and CISA SSVC rates exploitation as none, but the confidentiality impact is rated High, meaning successful access to log files could yield significant sensitive data.
Cross-site scripting in IBM Cognos Analytics and IBM Cognos Transformer allows a remote authenticated attacker to inject arbitrary JavaScript into the web user interface, executing in the browser context of other users within a trusted session. Affected versions span IBM Cognos Analytics 11.2.0 through 12.1.0 and IBM Cognos Transformer 11.2.4 through 12.1.0. The primary risk is credential disclosure - an attacker who can plant a payload could harvest session tokens or credentials from other authenticated users. No public exploit code exists and CISA SSVC rates exploitation as none at time of analysis.
Stored cross-site scripting in creatorsofcode's simplephp admin panel allows authenticated low-privileged users to inject persistent malicious scripts via the /admin/config-module.php configuration endpoint. When an administrator or privileged user subsequently views the affected page, the stored payload executes in their browser context, enabling session hijacking, credential theft, or unauthorized administrative actions. A proof-of-concept exists per SSVC intelligence; this CVE is not currently listed in CISA KEV.
Stored Cross-Site Scripting in ShopLentor (WordPress plugin, versions ≤ 3.3.8) allows authenticated contributors to permanently embed malicious JavaScript into WordPress pages via the 'blockUniqId' attribute of Product Grid blocks. Any user who subsequently visits an injected page triggers script execution in their browser, enabling session hijacking, credential theft, or unauthorized actions on behalf of the victim. EPSS is negligible at 0.03% (9th percentile), no CISA KEV listing exists, and no public exploit has been identified at time of analysis.
Incorrect authorization in DFIR-IRIS before version 2.4.28 allows authenticated low-privileged users to falsely attribute security alerts to arbitrary customers, corrupting case integrity in digital forensics and incident response workflows. Discovered by SBA Research and disclosed via oss-security on 2026-05-19, this flaw enables manipulation of alert-to-customer linkage without proper authorization checks. No active exploitation has been identified at time of analysis, and a patched release (2.4.28) is available.
Open redirect in WeGIA before version 3.7.3 enables authenticated attackers to weaponize the trusted WeGIA domain for phishing, credential harvesting, and malware distribution by manipulating the unvalidated `nextPage` parameter at the `/WeGIA/controle/control.php` endpoint. Affected deployments include any WeGIA instance running versions prior to 3.7.3 where the control endpoint is accessible to low-privileged authenticated users. No public exploit code and no CISA KEV listing have been identified at time of analysis, but the social engineering abuse potential against users who trust the institution's domain is the primary real-world risk.
Cross-site scripting in Synology Contacts before version 1.0.10-20659 allows authenticated remote users to read or write specific files containing non-sensitive information by injecting malicious input through the contact functionality. The CVSS scope change (S:C) confirms the injected script executes in a context beyond the originating application, affecting any victim who views the crafted contact entry. No public exploit identified at time of analysis, and CISA has not listed this in the Known Exploited Vulnerabilities catalog.
Unauthenticated statistics reset in WP Promoter plugin (WordPress, versions ≤1.3) allows any remote attacker to permanently delete promotional bar and popup campaign analytics by exploiting a missing capability check on the reset_stats() function. The function is registered on the wp_ajax_nopriv_wpp-reset_stats action hook - WordPress's mechanism for unauthenticated AJAX access - with no nonce validation, capability check, or authentication enforcement of any kind, making the destructive operation trivially invocable via a single HTTP POST request. No public exploit code has been identified at time of analysis, EPSS is 0.06% (18th percentile), and SSVC rates exploitation as none, indicating no observed active exploitation.
Uncontrolled resource consumption in the Simply Schedule Appointments WordPress plugin (all versions ≤ 1.6.11.5) enables unauthenticated remote attackers to exhaust PHP-FPM or mod_php worker processes, effectively rendering the WordPress site unavailable to legitimate users. The attack surface is a publicly accessible REST endpoint (/wp-json/ssa/v1/async) that directly passes a caller-controlled delay parameter into PHP's native sleep() function with no rate limiting or input sanitization. No public exploit code has been identified at time of analysis and EPSS is very low (0.05%, 15th percentile), suggesting limited opportunistic interest so far, though the trivially low attack complexity means any actor can attempt this with no tooling.
Prototype-chain property leakage in LiquidJS (npm/liquidjs <= 10.25.7) allows unauthenticated remote attackers to read prototype-chain properties of JavaScript objects passed to a {% render %} partial, even when the caller explicitly invoked parseAndRender() with { ownPropertyOnly: true } to lock down the render. The root cause is Context.spawn() failing to propagate the resolved per-render ownPropertyOnly value to child contexts, silently discarding a documented security override. A publicly available proof-of-concept exploit exists demonstrating that top-level {{ user.passwordHash }} is correctly blocked while the identical expression inside a {% render %} partial returns the sensitive value; no vendor-released patch is available at time of analysis.
Denial-of-service exposure in IBM OpenBMC firmware versions FW1110.00 through FW1110.11 allows unauthenticated remote attackers to partially degrade system availability by sending specially crafted network requests exploiting improper input quantity validation (CWE-1284). The attack requires no authentication, no user interaction, and low complexity, making it fully automatable per SSVC assessment - though no public exploit code has been identified at time of analysis. Because BMCs operate independently of the host OS and remain network-accessible even when servers are powered down, disrupting this layer carries operational risk disproportionate to the CVSS 5.3 Medium score alone.
SQL injection in uzy-ssm-mall v1.1.0 exposes sensitive database information to unauthenticated remote attackers via unsanitized input passed through the ProductMapper.xml MyBatis mapper and OrderUtil.java components. The vulnerability requires no authentication or user interaction, making it trivially automatable according to the SSVC framework. No public exploit identified at time of analysis, and EPSS sits at 0.04% (12th percentile), indicating low current exploitation pressure despite the permissive attack surface.
We are tasked with synthesizing data from multiple intelligence sources about CVE-2026-44838 into a comprehensive analysis, returned as a JSON object. The instructions are very specific: every field must be a plain text string, a coherent paragraph, not an array or object. We need to cross-reference all provided data. Let's extract key information: - Product: RabbitMQ (from description: "RabbitMQ is a messaging and streaming broker. From 4.2.0 to before 4.2.4, RabbitMQ's MQTT plugin allows..."). In CPE: cpe:2.3:a:broadcom:rabbitmq_server:*:*:*:*:*:*:*:* so product_name: "RabbitMQ" (as the broker). Or "RabbitMQ Server"? The description says "RabbitMQ" but product_name should be 1-3 words only, specific product. Use "RabbitMQ". I'll use "RabbitMQ". - Summary: Needs to be 2-3 sentence plain-language synthesis. Must not start with "A vulnerability..." etc. Start with specific impact verb and product. So: "Authorization bypass in RabbitMQ MQTT plugin allows authenticated users to escape regex topic restrictions." The vulnerability: regex injection due to unescaped client_id. Impact: bypass topic-level authorization, read/write to topics they shouldn't. Affected versions: 4.2.0 to before 4.2.4. Fixed in 4.2.4 and 4.3.0. No active exploitation indicated (KEV not listed, only SSVC exploitation none). EPSS 0.04% low. Patch available. So summary: "Authorization bypass in RabbitMQ's MQTT plugin (versions 4.2.0 to 4.2.3) enables authenticated MQTT users to inject regex operators into topic authorization patterns, gaining unauthorized access to messaging topics. Exploitation is not yet observed in the wild (CISA KEV not listed), and EPSS indicates low probability, but a patch is available in versions 4.2.4 and 4.3.0." That's good, mentions specific impact, product, versions, exploitation, patch. - technical_context: Underlying technology: RabbitMQ MQTT plugin, regex pattern matching with variable substitution for topic authorization. CWE-863: Incorrect Authorization, so root c
Kirby CMS's content-locking feature leaks authenticated users' email addresses and internal identifiers to low-privilege Panel users who are explicitly prohibited from seeing those users under role-based `users.access` or `users.list` permission restrictions. Any low-privilege authenticated Panel user on an affected site can harvest admin email addresses and user IDs during active content lock windows (default 10 minutes) simply by triggering an edit conflict or inspecting Panel view payloads. No public exploit has been identified at time of analysis, and exploitation is bounded to sites with non-default user-visibility restrictions, but the harvested data directly enables downstream phishing, credential stuffing, and admin account enumeration.
Authenticated cross-client stale result replay in Microsoft UFO's WebSocket task handling allows a low-privileged attacker to retrieve another user's completed automation session output. The framework accepts client-supplied session_id values without verifying ownership, so a requester who knows or can predict a prior session's identifier can hijack its stored result via the normal send_task_end() callback path. No public exploit has been identified at time of analysis, and KEV listing is absent, but the High confidentiality impact (C:H) is significant given UFO orchestrates device automation tasks that may capture sensitive screen content, documents, or credentials.
IBM SDI 7.2.0.0 through 7.2.0.14 and IBM Security Directory Integrator 10.0.0.0 through 10.0.0.2 could allow a remote attacker to obtain sensitive information when a detailed technical error message. Rated medium severity (CVSS 5.3), this vulnerability is remotely exploitable, no authentication required, low attack complexity. No vendor patch available.
Stored cross-site scripting in Agent Zero before version 1.15 enables arbitrary JavaScript execution in the application origin by exploiting the image_get API endpoint's failure to set Content-Security-Policy, X-Content-Type-Options, or Content-Disposition headers when serving SVG files. An unauthenticated attacker (CVSS PR:N) who can write a crafted SVG to any filesystem path readable by the agent-zero process can then socially engineer an authenticated user into visiting the endpoint, causing the browser to execute embedded scripts, exfiltrate the csrf_token cookie, and issue unauthorized API calls on the victim's behalf. No public exploit has been identified at time of analysis, and SSVC classifies exploitation as none with automatable set to no, reflecting the mandatory user-interaction prerequisite.
Cryptographic signature verification bypass in Northern.tech Mender Client 5 (before 5.0.4) allows unauthenticated remote attackers to circumvent JWT-based authentication controls. Tagged as both an Authentication Bypass and a JWT Attack, this CWE-347 flaw means the client fails to properly validate cryptographic signatures on tokens, potentially enabling unauthorized access to protected functionality or data. No public exploit code has been identified at time of analysis, and the low EPSS score of 0.02% (5th percentile) suggests exploitation is not currently widespread.
CSRF vulnerability in ZTE ZXUniPOS NDS-LTE enables an attacker to forge authenticated cross-site requests that modify system configuration data on behalf of a high-privilege user. The CVSS vector (PR:H/UI:R/AC:H) tightly constrains exploitation: a high-privilege administrator must be actively tricked into visiting attacker-controlled content while an authenticated session is live. No public exploit code exists and no KEV listing is present; EPSS at 0.02% (4th percentile) and SSVC Exploitation=none collectively signal negligible observed real-world exploitation activity.
Unauthorized private project enumeration in GitLab CE/EE exposes confidential project metadata to unauthenticated network attackers due to incorrect authorization checks (CWE-863). All GitLab installations running versions from 18.2 through the patched releases are affected - both Community and Enterprise editions. While the direct impact is limited to information disclosure (project enumeration rather than content access), exposed project names and IDs can facilitate targeted follow-on attacks against otherwise hidden repositories. No public exploit code has been identified at time of analysis, and this CVE is not listed in the CISA KEV catalog.
Use-after-free in TeamSpeak 3 Server versions 3.13.0 through 3.13.7 allows a low-privileged remote attacker to corrupt server memory via the process_resend_queue function within Connection State Management, resulting in limited integrity and availability impact. Discovered and disclosed by modzero.com (advisory MZ-26-01) and acknowledged by TeamSpeak via official security advisory TS-SA-2026-001, the vendor has released version 3.13.8 as the fix. No public exploit code exists and no active exploitation has been identified at time of analysis.
Server-Side Request Forgery in Budibase's VectorDB configuration endpoint (versions prior to 3.35.3) allows authenticated builder-level users to supply arbitrary host values - including cloud metadata addresses such as 169.254.169.254 or localhost - causing the server to initiate outbound TCP connections to internal network resources on the attacker's behalf. No public exploit has been identified at time of analysis, though SSVC classifies exploitation status as 'poc', indicating proof-of-concept material exists. In cloud-hosted deployments, the real-world impact exceeds what the CVSS 5.3 score implies, as metadata endpoint access can expose instance credentials and enable privilege escalation.
Missing authorization in ElementsKit Elementor addons Lite (WordPress plugin by Wpmet) through version 3.9.6 allows unauthenticated remote attackers to exploit incorrectly configured access control, resulting in limited unauthorized read access to protected data or functionality. The CVSS vector confirms network-based, zero-interaction exploitation with no authentication required, and SSVC classifies it as automatable - meaning attackers can scan and exploit at scale without manual intervention. No public exploit or CISA KEV listing exists at time of analysis, but the unauthenticated, low-complexity nature of the flaw makes it a realistic target for automated WordPress scanning campaigns.
pam_usb prior to 0.9.0 crashes under memory pressure due to assert()-based OOM guards in src/mem.c that are silently stripped by standard distribution build flags, enabling a local denial-of-service against authentication subsystems. Any allocation failure in xmalloc(), xrealloc(), or xstrdup() returns NULL, which every caller then dereferences unconditionally - the intended abort-before-dereference guarantee exists only in debug builds, not in Debian, Fedora, or Arch Linux packages that define -DNDEBUG via CFLAGS. A local attacker who can induce memory pressure at authentication time causes the PAM module to crash, locking all users out of sudo and login for the duration of the crash. No public exploit has been identified at time of analysis and the vulnerability is not listed in the CISA KEV catalog.
Out-of-bounds read in libusb's parse_iad_array() function (descriptor.c) affects all releases before 1.0.30, enabling local attackers in virtualized environments with USB passthrough to crash libusb-dependent processes via a crafted USB descriptor. The off-by-one error causes the bounds check to evaluate against the original total buffer size rather than the remaining unparsed size, allowing a one-byte read past the end of the malloc allocation when a descriptor's bLength is set to exactly (total_size - 1). No public exploit code exists and the vulnerability is absent from CISA KEV; a vendor-released patch is confirmed in v1.0.30.
Server-side request forgery in Budibase's automation engine (versions prior to 3.39.0) allows high-privileged authenticated users to redirect outbound HTTP requests from the Budibase server to attacker-controlled internal destinations by supplying an unvalidated queryId in the executeQuery automation step. When paired with a REST datasource pointed at internal infrastructure, automation execution causes the server to fetch arbitrary internal URLs and return their responses to the caller, potentially leaking sensitive internal service data. A proof-of-concept exists per SSVC assessment; no confirmed active exploitation or CISA KEV listing at time of analysis.
Sensitive information disclosure in IBM MQ Operator and IBM-supplied MQ Advanced container images exposes potentially sensitive data written to log files, readable by local users on the host or container system. Affected versions span three release tracks (LTS, CD, SC2) across both the MQ Operator (v2.0.0 through v3.9.1) and a broad range of container image releases from 9.3.x through 9.4.x. The CVSS score of 5.1 with a local attack vector and high complexity rating confines exploitation to users with existing local or container runtime access, and no public exploit has been identified at time of analysis.
Insufficiently protected credentials vulnerability in IPSpeaker component in Synology Surveillance Station before 9.2.2-11575 and 9.2.2-9575 allows remote authenticated users with administrator. Rated medium severity (CVSS 4.9), this vulnerability is remotely exploitable, low attack complexity.
Missing authorization vulnerability in AddOns functionality in Synology Surveillance Station before 9.2.2-11575 and 9.2.2-9575 allows remote authenticated users with administrator privileges to. Rated medium severity (CVSS 4.9), this vulnerability is remotely exploitable, low attack complexity.
Time-based blind SQL Injection in the EnvíaloSimple: Email Marketing y Newsletters WordPress plugin (all versions through 2.4.5) allows authenticated administrators to extract sensitive data from the underlying database. The vulnerability is in the 'orderby' parameter, which is insufficiently escaped and passed into existing SQL queries without adequate preparation, enabling an attacker with administrator-level WordPress credentials to append arbitrary SQL and enumerate database contents. EPSS is very low (0.03%, 8th percentile), no public exploit has been identified, and the vulnerability is not listed in CISA KEV, suggesting limited real-world exploitation pressure at this time.
Cleartext transmission of sensitive information vulnerability in Export Key functionality in Synology Surveillance Station before 9.2.2-11575 and 9.2.2-9575 allows remote authenticated users with. Rated medium severity (CVSS 4.9), this vulnerability is remotely exploitable, low attack complexity.
Memory exhaustion in IBM WebSphere Application Server (Liberty 19.0.0.7-26.0.0.5, traditional WAS 8.5 and 9.0) allows an adjacent-network, low-privileged attacker to trigger uncontrolled memory consumption by sending a specially crafted request. The attack requires both network adjacency and high complexity conditions, constraining the realistic threat surface significantly compared to the High availability impact rating. No public exploit code exists and CISA SSVC rates exploitation as 'none' with technical impact classified as 'partial', placing this vulnerability in a lower operational priority tier despite the A:H component impact.
Stored Cross-Site Scripting in the myLinksDump WordPress plugin (all versions ≤1.6) allows authenticated administrators to permanently inject arbitrary JavaScript into pages via the unsanitized 'link_title' parameter, executing in any victim's browser upon page access. Exploitation is constrained to WordPress multi-site environments or single-site installs with unfiltered_html disabled, and requires administrator-level credentials plus victim interaction. EPSS is 0.03% (9th percentile) and SSVC confirms no known exploitation, placing this firmly in a low-priority tier despite the stored XSS class.
Stored Cross-Site Scripting in the rexCrawler WordPress plugin (versions ≤ 1.0.15) allows authenticated administrators to inject persistent malicious scripts into settings pages, which then execute in the browsers of any user who accesses those pages. The vulnerability originates in admin_main.php at two distinct injection points (lines 108 and 239) and is constrained to multi-site WordPress environments or single-site installs where the unfiltered_html capability has been explicitly disabled. With an EPSS of 0.02% (7th percentile), no CISA KEV listing, and SSVC exploitation status of 'none', this represents a low-urgency finding despite its network-accessible attack vector. No public exploit code has been identified at time of analysis.
Open redirect in DFIR-IRIS before version 2.4.28 allows unauthenticated remote attackers to craft URLs that silently redirect authenticated users to arbitrary external domains. The CVSS Scope:Changed (S:C) component confirms the vulnerability crosses the application's trust boundary, making it a viable vector for phishing and credential harvesting campaigns targeting incident responders and forensic analysts who use the platform. No public exploit code or active exploitation has been identified at time of analysis, and no CISA KEV listing is present.
Race condition in the Linux kernel's XFRM ICMP route lookup path causes a kernel WARN_ON that can crash affected systems. Within `icmp_route_lookup()`, a TOCTOU window between a locality check and a subsequent `ip_route_input()` call allows a concurrently executing `ip addr add` to return a LOCAL route whose `dst.output` is set to `ip_rt_bug()` - a debugging stub that fires `WARN_ON` when invoked during ICMP error transmission. Exploitation requires local access, active XFRM/IPsec policy, and precise race-window timing; no active exploitation is confirmed and EPSS sits at 0.02%, though a public reproducer exists that requires kernel modification to reliably trigger.
Race condition in the Linux kernel's accel/amdxdna driver allows a local low-privileged user to cause device availability impact on systems equipped with AMD XDna AI accelerator hardware. The flaw in the rpm_on flag check permits command submissions to the accelerator during a narrow autosuspend transition window before the device has fully resumed, producing undefined hardware behavior or driver crashes. No public exploit exists and EPSS is at the 6th percentile (0.02%), indicating negligible real-world exploitation probability; no active exploitation is confirmed.
Race condition in the Linux kernel memory management subsystem during large-folio migration can cause kernel availability disruption on SMP/NUMA systems. The flaw in migrate_folio_move() causes a destination folio to become visible to concurrent rmap-removal paths before being requeued onto the deferred split queue, triggering a kernel WARN in deferred_split_folio() or silently losing a folio from split_queue when the shrinker races the migration lock. With no public exploit, no CISA KEV listing, and an EPSS of 0.02%, this is a low real-world risk issue primarily relevant to HPC, virtualization, and database workloads with heavy NUMA migration activity.
Deadlock in Linux kernel DAMON (Data Access Monitor) subsystem allows a local low-privileged user or kernel code path to cause an indefinite thread hang in the mm/damon/core module via a race condition between damos_walk() request registration and kdamond_fn() exit sequencing. Systems running Linux kernels from commit bf0eaba0ff9c9c8e6fd58ddfa1a8b6df4b813f61 through the patch commits are affected, with availability as the sole impact (CVSS C:N/I:N/A:H). No public exploit identified at time of analysis, and EPSS probability is 0.02% (5th percentile), indicating negligible real-world exploitation interest.