Use-after-free in PHP's per-request global state handling lets unauthenticated remote attackers achieve arbitrary code execution against persistent workers that reuse memory across requests - PHP 8.2 before 8.2.35, 8.3 before 8.3.36, 8.4 before 8.4.27 and 8.5 before 8.5.12 are affected. Exploitation is high complexity (CVSS:4.0 9.2 with AC:H/AT:P): it needs a 'producer' operation that leaves request-owned state published after an interruption such as a fatal error, uncaught exception, timeout or OOM, followed by a later request to the same worker whose early operation consumes the stale pointer before normal initialization resets it. The vendor documents a confirmed mod_php RCE path (PHP 8.5 with fatal_error_backtraces enabled) and a confirmed session upload progress lifecycle; a public PoC reference exists but is not verified as working exploit code, CISA SSVC records exploitation as none, and the priority decision is act_now because affected PHP branches are patched and technical impact is total.
Unauthenticated remote code execution in PHP 8.2.x before 8.2.35, 8.3.x before 8.3.36, 8.4.x before 8.4.27, and 8.5.x before 8.5.12 arises when an attacker drives a worker into an out-of-memory condition while request memory is only partially initialized, leaving freed/reused structures that the request shutdown path then processes. Exploitation is confirmed practical against Apache mod_php with default input and memory limits in normal multi-worker prefork setups, and publicly available exploit code exists; the vendor and distros (Ubuntu USN-8948-1) have released fixed builds. With a 9.2 CVSS score, a public PoC, and an act_now priority decision, this warrants immediate patching, though reliability varies with build, extensions, frontend behavior and allocator layout, and no complete PHP-FPM exploit has been demonstrated.
Remote code execution is possible in Node.js releases that bundle a vulnerable Brotli decoder: a malformed brotli-compressed stream can trigger an out-of-bounds write when decompressed, corrupting memory in the process handling the data. Exposure is limited to deployments that actually pass attacker-controlled brotli data into the decoder - for example services honouring Content-Encoding: br on untrusted requests or responses, or applications calling zlib.brotliDecompress/BrotliDecompress on untrusted input - and no authentication or user interaction is required in that scenario. The flaw is rated High with a CVSS 4.0 score of 9.2 (assessed CVSS 3.1: AV:N/AC:H/PR:N/UI:N/C:H/I:H/A:H), no CISA KEV listing confirms active exploitation, and a public PoC reference exists but is not verified as working exploit code; the vendor fix is Node.js v26.12.1, released 2026-10-21.
Stack-based buffer overflow in the TOTOLINK EX200 Wi-Fi range extender (firmware 5.2.3c.7200) lets remote, unauthenticated attackers crash or potentially execute code on the device by sending a crafted HTTP request to the setStaticDhcpConfig function in /cgi-bin/cstecgi.cgi with an oversized Comment parameter. Publicly available exploit code exists (a GitHub proof-of-concept write-up is linked from the VulDB record), and the CVSS 4.0 base score of 9.3 reflects full confidentiality, integrity and availability impact with PR:N/UI:N/AC:L - no credentials or user interaction required. The practical constraint is reachability: the flaw is only exploitable if the extender's web management interface is exposed to the attacker, so the real-world risk is highest on devices whose administrative UI is reachable from the WAN or from an untrusted network segment rather than the typical isolated LAN-side deployment.
Stack-based buffer overflow in TOTOLINK EX200 firmware 5.2.3c.7200 allows remote attackers to overflow a stack buffer by sending an oversized macAddress argument to the setWiFiAclAddConfig function in /cgi-bin/cstecgi.cgi (wireless.so), with high confidentiality, integrity, and availability impact per the supplied CVSS vectors. Publicly available exploit code exists, and while the CVSS vector lists PR:N (unauthenticated), authentication requirements are not confirmed from the available data; practical exploitation is normally limited to attackers who can reach the device's LAN/Wi-Fi management interface unless that UI is deliberately WAN-exposed.
A stack-based buffer overflow in the formSetFirewallCfg handler of the Tenda AC19 router's httpd web management service lets a remote attacker crash or potentially execute code on the device by sending an overlong firewallEn parameter to /goform/SetFirewallCfg. The issue is confirmed against firmware 16.03.10.07 and carries a 9.3 CVSS v4 score with a PR:N vector, meaning no authentication is asserted as a prerequisite, although Tenda goform endpoints commonly enforce a session on default builds and the independent assessment flags this as unverified. Publicly available exploit code exists (a GitHub PoC repository is linked from the CVE record), which raises the likelihood of opportunistic attempts against any AC19 whose management interface is reachable.
Laravel TokenGuard authentication bypass is a genuine high-priority issue, not an inflated CVSS score: unauthenticated attackers can send a crafted non-string credential, such as an array-valued api_token, to routes protected by TokenGuard and, under affected MySQL or MariaDB configurations, be matched to an unintended existing user, enabling account takeover. The flaw affects Laravel Framework before 12.68.0 and 13.x before 13.29.0 and requires the vulnerable guard plus database type-coercion behavior; no public exploit was identified at time of analysis, and vendor-released fixes are available in 12.68.0 and 13.29.0.
Unauthenticated remote attackers can read and tamper with critical data handled by Oracle Java SE through its JSSE (Java Secure Socket Extension) component, rated CVSS 9.1 (AV:N/AC:L/PR:N/UI:N, C:H/I:H/A:N). Affected builds are Java SE 8u511 and 8u511-perf, 11.0.33, 17.0.21, 21.0.13, 25.0.5 and 27.0.1, plus Oracle GraalVM for JDK 17.0.21 and 21.0.13 and GraalVM Enterprise Edition 21.3.20. Oracle's alert states the flaw is reachable via APIs of the affected component, such as a web service feeding untrusted data into JSSE, and adds a secondary sandbox-escape path for deployments running untrusted code in Java Web Start applications or applets; no public exploit code or active exploitation was identified at time of analysis, but the missing authentication prerequisite and network reachability make this a high-priority patch target rather than a paper finding.
Unauthenticated remote attackers can take full administrative control of Apache Pulsar Manager 0.3.0 and later by creating an administrator account through an endpoint that remains reachable after initialization, and every account has effective full privileges. In all versions, the manager proxies requests to the Pulsar admin API without enforcing its own authentication and can attach its configured Pulsar credential to a requester-chosen destination, allowing unauthenticated users to perform permitted admin operations and steal the credential. No public exploit code was identified at time of analysis, but the product is retired, unsupported, and will not receive fixes, so exposure persists until deployments are removed.
Apache Traffic Server versions 8.0.1 through 8.1.11, 9.0.0 through 9.2.15, 10.0.0 through 10.1.4, and 10.2.0 ship an experimental 'magick' plugin that, in its default unsigned configuration, passes client-controlled data into ImageMagick 'convert' arguments, letting remote attackers inject command-line options (CWE-88 argument injection) with no authentication or user interaction. Because the plugin is experimental and not enabled by default, only deployments that deliberately loaded it without enabling request signing are exposed; those that are exposed face a CVSS 3.1 score of 10.0 (AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:H) with scope change and full confidentiality, integrity, and availability impact, since injected arguments can drive ImageMagick into reading or writing arbitrary files or executing delegated commands. No public exploit code or CISA KEV entry is associated with this CVE at time of analysis, but the impact on a vulnerable, enabled instance is total, and upgrading to 9.2.16, 10.1.5, or 10.2.1 is recommended.
Arbitrary file upload in WP Mailster through 1.9.1.0 allows an attacker to place a dangerous executable file, such as a PHP web shell, into a web-accessible location and potentially gain remote code execution on the WordPress host. The provided CVSS vector rates the issue critical and asserts network exploitation with no privileges (PR:N), but the single Patchstack reference does not identify the upload endpoint or required WordPress capability, and the independent assessment notes that comparable plugin upload flaws often need at least a low-privileged account, so defenders should validate actual exposure. No public exploit code or CISA KEV entry was identified at time of analysis.
Splunk SOAR deployments running any version below 8.7.0 expose files and directories to external parties (CWE-552), and the vendor rates the issue at CVSS 9.9 (AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H), meaning a network-reachable attacker holding only a low-privileged authenticated account can reach the exposed resources with no user interaction, with high confidentiality, integrity and availability impact and a scope change beyond the vulnerable component. The flaw was found during Cisco's internal security review and shipped in a bundled software hardening release rather than as a single documented bug, so the precise endpoint or feature that exposes the files is not described in the available sources. SSVC records no known exploitation and marks the issue not automatable, and no public exploit code or proof-of-concept was identified at time of analysis; the vendor-prioritised decision is act now, with upgrading to 8.7.0 as the remediation.
Improper neutralization weaknesses in Splunk Enterprise allow a network-positioned attacker holding a valid low-privilege account to achieve high confidentiality, integrity, and availability impact, with a scope change (S:C) indicating the effect can extend beyond the initially compromised component. Affected branches are Splunk Enterprise 9.4 before 9.4.16, 10.0 before 10.0.11, 10.2 before 10.2.8, and 10.4 before 10.4.4; CVSS 3.1 is 9.9 (AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H), so authentication is required but no user interaction is, and only the listed versions are in scope. Vendor fixes are available and the issue is prioritized act_now, but no public exploit code and no CISA KEV entry were identified at time of analysis; the two supplied references are one Splunk advisory plus an NVD detail page, so the corroborating signal quality is thin relative to the score.
Improper access control in Splunk Enterprise 9.4, 10.0, 10.2, and 10.4 allows a low-privileged authenticated user to cross a security boundary and achieve high confidentiality, integrity, and availability impact, making the release a 9.9 severity fix. The flaw was found during Cisco's internal hardening review rather than through external reporting, and the affected ranges are 9.4 before 9.4.16, 10.0 before 10.0.11, 10.2 before 10.2.8, and 10.4 before 10.4.4. No public exploit code has been identified and there is no CISA KEV entry, but vendor patch availability combined with an act_now priority label places upgraded discipline on anyone running an unpatched instance; the trigger is not described in the available sources.
Pillow versions 11.0.0 through 12.x (fixed in 13.0.0) can reuse a stale native ImagingCore buffer when a multi-frame image is advanced to a later frame whose pixel mode widens from a narrow format to RGB or CMYK, or whose dimensions change, producing native out-of-bounds writes and crashes; the MPO path can additionally over-read memory and, on specific runtime/heap profiles, achieve command execution. Exploitation requires the application to open an attacker-supplied PSD, MPO, DCX, or MIC file and actually seek/load past the first frame, since reading only the initial frame does not trigger the bug; no credentials or user interaction are needed in an automated image-processing pipeline. No public exploit code has been identified at time of analysis and the issue is not in CISA KEV, although the 9.8 CVSS and the trivially craftable file format make weaponization plausible for any service that validates or thumbnails multi-frame uploads.
Heap-based buffer overflow in the GStreamer rtpL16depay element allows unauthenticated remote attackers to execute arbitrary code at the privilege level of the process hosting the pipeline. Only GStreamer-based software that actually instantiates rtpL16depay to depayload attacker-supplied RTP L16 (16-bit linear PCM) audio is exposed, so a passive RTP receiver or streaming server is effectively remotely exploitable while client-side applications may require the user to connect to a malicious stream. No public exploit code or confirmed active exploitation was identified at time of analysis, and the vendor has published an upstream fix commit but no tagged patched release has been independently confirmed.
Authenticated arbitrary file upload in the Mang Board WP WordPress plugin through 2.4.4 lets any low-privileged account plant a PHP web shell and execute code on the underlying web server, changing scope beyond WordPress (CVSS 9.9). Exploitation requires a valid PR:L account, so exposure is concentrated on sites that permit untrusted or self-service registration; no public exploit or KEV entry was identified at time of analysis, and no fixed version is specified in the supplied data.
Unauthenticated remote command execution as root affects Cisco Expressway Series (Expressway-C and Expressway-E) deployments that expose the database service. The flaw combines reusable persistent cookies with a failure to validate client certificates when clustering TLS Verify Mode is set to Enabled, so an attacker in possession of the cookies can connect to the database service and invoke RPC commands to run arbitrary OS commands as root. The CVSS 3.1 score is 9.8 (AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H); no public exploit code was identified at time of analysis, and exploitation requires network reachability to the database and clustering service.
Authentication bypass in Cisco Unified CM and related Cisco collaboration products - Emergency Responder, Unified CM IM&P, Unity Connection, Prime Collaboration Deployment, and Unified CM SME - allows remote, unauthenticated attackers with low complexity and no user interaction to defeat authentication and potentially achieve full confidentiality, integrity, and availability impact (CVSS 9.8 Critical, CWE-287). The flaws were found in an internal Cisco review and fixed through hardening releases; no public exploit code or confirmed active exploitation was identified at time of analysis. Practical risk is tempered by exposure: these telephony and collaboration servers typically sit on internal or management networks rather than the public internet.
Cisco's batched voice-portfolio hardening release addresses several internally discovered injection-class defects (CWE-74) that affect Cisco Unified Communications Manager, Unified CM IM & Presence, Unified CM Session Management Edition, Unity Connection, Emergency Responder, and Prime Collaboration Deployment, and the vendor scores the bundled issue set at CVSS 9.8 (AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H) - network reachable, low complexity, no privileges and no user interaction, with high confidentiality, integrity, and availability impact. Because the advisory is a consolidated hardening bulletin rather than a per-defect writeup, the specific injectable parameter, protocol, or service is not disclosed, so the exact trigger cannot be derived from available data and real-world exposure depends on whether the management and application interfaces of these appliances are reachable from an attacker's network segment. No public exploit code and no confirmed active exploitation were identified at time of analysis, and these are typically intranet-deployed enterprise voice appliances rather than internet-facing systems, which narrows practical attack surface to adjacent-network or intranet positions in most deployments.
Signals conflict and must be weighed carefully: Splunk Enterprise versions before 9.4.16, 10.0.11, 10.2.8, and 10.4.4 contain internally discovered protection-mechanism failures tracked as CVE-2026-76301 (CWE-693), rated CVSS 9.8 with a vector for unauthenticated network attackers and high confidentiality, integrity, and availability impact. The specific feature, protocol, configuration, trigger, and per-vulnerability impact are not disclosed in the available sources, and no public exploit was identified at time of analysis; EPSS is low at 0.17% (6th percentile). A vendor patch is available, and the vuln.today decision is act_now.
Code execution inside the target SageMaker execution environment, together with disclosure of the execution role's IAM credentials, is reachable through the Amazon SageMaker Python SDK's default_bucket() helper, which provisions and then consumes an S3 bucket whose name is derived predictably from the account ID and region but never verifies who actually owns that bucket. The defect affects V3 releases before 3.10.0 and V2 releases before 2.257.3; a remote actor who registers that predictable bucket name in their own AWS account first can influence the artifacts, scripts and data the victim's training, processing and hosting jobs read and write, which the vendor scores 9.4 (Critical) under CVSS 4.0 and which our independent assessment rates CVSS 3.1 AV:N/AC:H/PR:N/UI:N/S:C/C:H/I:H/A:H because the attacker must win a bucket-name pre-creation race. No public exploit code has been identified at time of analysis and the issue is not listed in CISA KEV; exploitation favors newly onboarded or not-yet-provisioned environments, since targets that already own their default bucket are not exposed.
Command injection in the `ssh` service of Lagoon core (uselagoon/lagoon) lets an authenticated Lagoon user execute arbitrary commands on the core container, which can be leveraged to escalate to elevated administrative permissions across the platform. Affected deployments are those running a Lagoon core release before the fix shipped in v2.33.2 (helm chart lagoon-core v1.62.2); note the advisory text says "prior to 2.33.1" while the actual patch landed in 2.33.2, so 2.33.2 should be treated as the first safe version. Exploitation requires a provisioned account (CVSS vector PR:L, authenticated) and network reachability to the Lagoon core ssh service specifically; the separate `ssh-portal` service is not affected. No public exploit code or confirmed active exploitation was identified at time of analysis, and the issue is detectable by watching the ssh service container logs for the string "too many arguments".
Authentication bypass and SYSTEM-level command execution in Best Practice Software Bp Premier before 1.14.1.1128 stem from a hardcoded, installation-wide Microsoft SQL Server sysadmin password, allowing unauthenticated remote attackers with TCP access to the database to bypass application authentication and execute operating system commands as SYSTEM under the documented default SQL service configuration. The credential is shared across all installations and cannot be changed without breaking the application, so no in-product hardening is possible pending the vendor patch. No public exploit code or CISA KEV entry was identified at time of analysis, but the main constraint is database reachability: deployments exposing SQL Server port 1433 or attackers already on the clinic LAN can exploit it.
Missing authentication in the HKUDS Vibe-Trading FastAPI server (pip package vibe-trading-ai 0.1.0 through 0.1.6) lets unauthenticated remote attackers reach every endpoint guarded by Depends(require_auth), because require_auth() returns immediately when API_AUTH_KEY is unset - which is the shipped default in agent/.env.example. Attackers can POST to /sessions/{id}/messages to drive the agent's BashTool toward shell command execution, upload .py/.sh/.yaml files via /upload, read other users' session histories, and recover partial API keys via _mask_secret(); because the shipped Dockerfile has no USER directive and docker-compose publishes 0.0.0.0:8899, any command executed lands as root inside the container. Exploitation matches the assessed vector AV:N/AC:L/PR:N/UI:N with high confidentiality, integrity and availability impact; the full chained RCE demonstration additionally requires the operator to have configured a working LLM API key so the tool-call loop can complete, while the unauthenticated API surface, file upload, history read and partial key disclosure work without it. No CISA KEV listing and no public exploit code identified at time of analysis, though the vendor advisory publishes detailed reproduction steps; fix in 0.1.7.
Cross-site scripting (XSS) in the web interfaces of Cisco Emergency Responder, Cisco Unified CM, Unified CM IM&P, Cisco Unity Connection, Prime Collaboration Deployment, and Unified CM SME lets unauthenticated, network-positioned attackers inject script that executes when a victim interacts with crafted content. The CVSS 9.3 vector reflects network reachability, low attack complexity, no attacker authentication, required victim interaction, scope change, and high confidentiality/integrity impact, so meaningful exploitation generally depends on luring an authenticated administrative user. No public exploit code or CISA KEV entry was identified at time of analysis, and Cisco's advisory describes hardening releases rather than a single emergency patch.
Blind SQL injection in WPdreams Ajax Search Lite through version 4.14.6 allows unauthenticated remote attackers to extract sensitive data from the WordPress database by inferring query results. The plugin's public-facing search endpoint is exposed to anonymous visitors by design, so no authentication or user interaction is required. No public exploit code was identified at the time of analysis, but the vulnerability is fully automatable and high severity.
PHPOffice PHPPresentation through version 1.2.0 will pass attacker-controlled serialized data to unserialize() when IOFactory::load() is pointed at a crafted presentation file, letting remote unauthenticated attackers instantiate arbitrary PHP objects inside the hosting application. Where a usable gadget chain exists in the application or its dependencies, that object instantiation escalates to code execution or arbitrary file writes; without gadgets the realistic outcome is object injection or denial of service, which is why the assessed vector marks a present attack requirement (AT:P) rather than an unconditional network exploit. No confirmed active exploitation (CISA KEV) and no public exploit identified at time of analysis, but the library is widely embedded in document-conversion and upload pipelines, so exposure depends entirely on whether the vulnerable Serialized reader path is reachable from untrusted input.
Remote code execution in vLLM 0.7.3 through 0.30.x lets remote, unauthenticated clients run arbitrary Python inside the API server process when a multimodal request to the /tokenize endpoint supplies a crafted mm_processor_kwargs.code_revision value, which vLLM forwards to Transformers AutoProcessor.from_pretrained and uses to select which revision of the model repository's dynamic processor code gets imported. Exploitation requires two non-default preconditions: the server must be running with trust_remote_code enabled, and the attacker must control (or be able to write) a revision in the model repository being served, so default hardened deployments with remote code disabled are not exposed. No public exploit code or confirmed active exploitation was identified at time of analysis, but the impact for exposed configurations is full confidentiality, integrity and availability loss (vendor CVSS 4.0 score 9.2; our independent CVSS 3.1 assessment of 8.1 reflects the high attack complexity created by the prerequisites).
Server-side request forgery in LocalAI 4.6.0 through 4.11.0 allows unauthenticated remote attackers to make the service fetch internal resources by submitting a gallery config URL to POST /models/apply that redirects to loopback, private, link-local, or cloud metadata addresses and returns the internal response as gallery content. The issue is a redirect-based bypass of the gallery URL guard; exploitation requires the model-management API to be network-reachable and the attacker to control or influence a redirecting server, and high impact depends on reachable internal services or a metadata endpoint. No public exploit identified at time of analysis.
Authentication bypass in Trigger.dev self-hosted v4 deployments (npm trigger.dev <=4.5.5) allows unauthenticated attackers to forge magic-link tokens and log in as any email address when the default MAGIC_LINK_SECRET from hosting/docker/.env.example is still in use. Publicly available exploit code exists; successful exploitation can expose or modify tenant data, and may escalate to full infrastructure compromise because default Postgres, Redis, and ClickHouse credentials are reachable from runner containers on the same Docker network. The issue is limited to Docker Compose deployments that never rotated the example secrets and left WHITELISTED_EMAILS unset; the hosted SaaS and instances with regenerated secrets or an email allowlist are not affected.
Unauthenticated role forgery in Apache Pulsar lets a client present a forged or unsigned JSON Web Token listing arbitrary roles and have those roles honored when the broker uses MultiRolesTokenAuthorizationProvider alongside a non-empty anonymousUserRole. The flaw is reachable over the broker's HTTP surfaces, including the admin REST API and the WebSocket API, where failed credentials can fall back to the anonymous role while the original Authorization header is retained and later trusted without signature verification; an attacker who names a role configured in superUserRoles can obtain super-user access to tenants, namespaces, topics, permissions and broker configuration. Only deployments combining MultiRolesTokenAuthorizationProvider with a non-empty anonymousUserRole and the fallback conditions are exposed, while installations using the default PulsarAuthorizationProvider, another authorization provider, or an unset anonymousUserRole are unaffected. Vendor-released patch: 4.0.14 and 4.2.5; no public exploit identified at time of analysis, and exploitation also requires the specific non-default configuration described.
Response-framing desync in Apache Traffic Server (8.0.0 through 8.1.11, 9.0.0 through 9.2.15, 10.0.0 through 10.1.4, and 10.2.0) allows a malformed 302 response that should have been rejected as a parse error to pass the rejection path through the Transfer-Encoding: chunked carve-out, so the proxy stores and serves a response that different peers interpret differently. Per the CVSS vector the issue is remote and unauthenticated (AV:N/AC:L/PR:N/UI:N, scope changed, I:H/C:L), enabling cache poisoning and response-framing desync against deployments acting as caching or forwarding proxies. No public exploit code or CISA KEV entry is associated at time of analysis; the practical precondition is that a suitable malformed upstream response actually traverses the proxy. Fixed versions 9.2.16, 10.1.5, and 10.2.1 are published by the vendor.
Authorization bypass in the Sentry Next.js SDK affects @sentry/nextjs from 9.36.0 before 9.47.2, 10.x before 10.75.2, and 11.0.0-rc.1 when the webpack integration is used with a configured tunnelRoute. An unauthenticated remote request can reach an application route that collides with the tunnel path, or resolves through a root dynamic segment, without the middleware authorization that would otherwise run; exploitation additionally requires a webpack build, a colliding route, and middleware being the sole authorization control. The flaw is fixed in 9.47.2, 10.75.2, and 11.0.0-rc.1, and no public exploit was identified at time of analysis.
Session token replay in the WebFig management interface of MikroTik RouterOS (through 7.24.5) lets an attacker who already possesses a captured administrator session token bypass authentication and gain persistent administrative access, because tokens stay valid across newly negotiated ECDH sessions until the device's login store is explicitly cleared. The 9.1 CVSS v4 score reflects high confidentiality and integrity impact, but its AC:H rating is meaningful: the attacker must first obtain a valid token out-of-band, for example through the unfixed transport-layer exposure described in CVE-2025-61481, and no user interaction is needed once a token is held. No public exploit identified at time of analysis and no CISA KEV entry exists; devices whose WebFig interface is not reachable from untrusted networks are not realistically exposed.
Request smuggling in Apache Traffic Server 8.0.0 through 8.1.11, 9.0.0 through 9.2.15, 10.0.0 through 10.1.4, and 10.2.0 allows remote unauthenticated attackers to have a later request read as leftover body data on a recycled origin connection, enabling access-control bypass, cross-user request disruption, and cache poisoning. The attacker must declare a body via Content-Length or Expect: 100-continue but never fully send it, then win a timing window while ATS returns the origin connection to its shared keep-alive pool. No public exploit has been identified at time of analysis, but the default configuration is affected and exploitation is probabilistic rather than guaranteed.
Acronis Cyber Infrastructure (ACI) does not properly enforce authorization in its management/API layer, letting remote attackers disclose sensitive information and manipulate protected resources (CVSS 9.1, AV:N/AC:L/PR:N/UI:N, C:H/I:H). The vendor vector specifies PR:N, so no authentication, user interaction, or special configuration is documented as required; exposure is mainly bounded by whether the ACI management interface is reachable from untrusted networks. Affected deployments are those running builds earlier than 6.0.1-112, 6.1.1-64, and 6.2.1-71. No public exploit code and no CISA KEV entry were identified at time of analysis, so active exploitation is not confirmed.
Out-of-bounds write flaws in the Cisco Snort 3 inspection engine allow remote, unauthenticated attackers to corrupt memory in any deployment where Snort 3 parses attacker-reachable network traffic, potentially leading to arbitrary code execution or a denial-of-service condition in the inspection process. The affected code ships not only as the standalone open-source Snort 3 project but also embedded in Cisco Secure Firewall Threat Defense (FTD), the Cisco UTD Snort IPS engine, and Cisco Cyber Vision, so exposure spans both Cisco appliances and third-party deployments of the open-source engine. Cisco addressed the issues through hardening releases produced during an internal security review, but the advisory groups the findings without publishing per-defect trigger details, and no public exploit code has been identified at time of analysis; CVSS AC:H reflects that an attacker must land traffic on the specific vulnerable parsing path rather than send a trivially repeatable packet.
Unauthenticated network access to the shared proxy cache is possible in Apache Traffic Server deployments that load and enable the experimental tsmemcache plugin, allowing remote attackers to read, modify, and delete cached objects without credentials. Affected releases span 8.0.0 through 8.1.11, 9.0.0 through 9.2.15, 10.0.0 through 10.1.4, and 10.2.0; the exposure is gated by the plugin being explicitly enabled and its memcache-style interface being network-reachable, since it is not active in a default installation. No public exploit code or confirmed active exploitation was identified at time of analysis, though the CVSS 4.0 base score of 9.4 and the unauthenticated nature of the interface once exposed make this a high-priority fix for anyone running the plugin.
Missing authentication in Apache bRPC through 1.17.0 allows remote attackers to bypass the configured Authenticator on deployments that expose the streaming RPC path to untrusted networks. The assessed vector is CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:L/A:N, so the impact is limited to low confidentiality and integrity effects with no availability impact. The issue is fixed in bRPC 1.18.0; no public exploit code or CISA KEV entry was identified in the available data, and the supplied CVE description is truncated, so exact triggering conditions should be confirmed against the full Apache advisory.
Unauthenticated remote attackers can abort an Apache bRPC process (versions through 1.17.0) by sending a single small HTTP/2 HEADERS block that over-fills the HPACK dynamic table's fixed-size ring buffer, tripping a CHECK(!_header_queue.full()) assertion that terminates the server. Because the h2 parser is enabled by default on every bRPC server port and decodes header blocks before any Authenticator runs, no credentials, user interaction, or non-default configuration are required - the default 4096-byte table is enough to reach the abort. The issue is fixed in bRPC 1.18.0 (upstream patch PR #3462); no public exploit code or confirmed active exploitation was identified at time of analysis, and no CVSS, EPSS, or KEV data was supplied alongside the advisory.
Remote denial of service in Apache bRPC through 1.17.0 allows attackers to crash servers or exhaust memory by sending crafted HTTP/2 (h2) frames to the default-enabled h2 parser, which processes peer-controlled stream and header metadata before authentication. The vendor advisory describes remote unauthenticated exploitation; the assessed impact is availability-only (CVSS 3.1 AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H). No public exploit code or CISA KEV entry was identified at time of analysis.
Remote pre-authentication denial of service in Apache bRPC through 1.17.0 lets any network peer that can reach an HTTP/2 or gRPC (h2) listener abort the server process. The HPACK decoder sizes its default 4096-byte dynamic table for 33-byte minimum entries (roughly 120 slots) but evicts only on byte usage, so 121 minimal 33-byte insertions (3993 bytes) fit under the byte cap while exhausting the fixed slot ring buffer; the 121st AddHeader trips CHECK(!_header_queue.full()) and terminates the process. No public exploit identified at time of analysis, and because the h2 parser decodes headers before any Authenticator check, no credentials or user interaction are required.
Remote unauthenticated denial of service in Apache bRPC's HPACK decoder can abort the server process when a peer sends a crafted HTTP/2 header block. bRPC through 1.17.0 is affected because the default 4096-byte HPACK dynamic table allocates about 120 ring-buffer slots while eviction is tracked by bytes, allowing 121 minimal 33-byte entries to fill the queue and trip a CHECK assertion on the next AddHeader. The h2 parser is enabled by default and runs before authentication, so any exposed HTTP/2 or gRPC listener is at risk; no public exploit code was identified at time of analysis, and 1.18.0 fixes the issue.
Unauthenticated remote attackers can abort Apache bRPC 1.17.0 and earlier servers that expose HTTP/2 or gRPC (h2) listeners by sending a header block that fills the HPACK dynamic table's ring buffer without triggering byte-based eviction. The flaw is a slot-accounting mismatch in the HPACK decoder: the default 4096-byte table is provisioned for 34-byte entries, but 33-byte minimal entries permit 121 insertions before the fixed-size ring buffer overflows, tripping CHECK(!_header_queue.full()) and terminating the process. No public exploit code was identified at time of analysis; the independent assessment rates this CVSS 3.1 7.5 (AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H), availability-only, with exploitation requiring only network reachability to a default-configured h2/gRPC port because HPACK decoding occurs before any Authenticator check.
Remote unauthenticated denial of service in Apache bRPC through 1.17.0 lets a network peer crash an HTTP/2 or gRPC (h2) server by sending a crafted sequence of minimal HPACK header entries. The default 4096-byte HPACK dynamic table is allocated as 120 slots based on a 34-byte entry size, while the true minimum entry is 33 bytes, so 121 minimal entries fit within the byte cap without triggering eviction and then hit a full-header-queue CHECK that aborts the process. No authentication or user interaction is required; bRPC 1.18.0 and upstream patch PR #3462 correct the sizing, and no public exploit code or CISA KEV entry was identified in the provided data.
Remote denial of service in Apache bRPC 1.17.0 and earlier allows unauthenticated network peers to drive disproportionate memory consumption or abort the process by sending input whose declared or decompressed size is never validated. Five parser paths are affected: gzip/zlib/snappy decompression (only the compressed body is counted against -max_body_size, so a 64 MiB cap-compliant body can expand to tens of GiB), AMF string readers that allocate the wire-declared length before the bytes arrive, RTMP chunk headers that re-declare a message length mid-stream into an ever-growing reassembly buffer, the unbounded meta_size field of the sofa-pbrpc parser, and mcpack2pb, which reserves on wire-declared item counts and underflows its buffer size when a string value_size is 0, triggering an abort. The flaw is fixed in 1.18.0 and no public exploit code was identified at time of analysis; reachability is limited to deployments that actually accept compressed bodies or rtmp/sofa-pbrpc/mcpack2pb traffic from untrusted clients, since these parsers run before any Authenticator check.
Unauthenticated remote attackers can crash an Apache bRPC server running version 1.17.0 or earlier by sending a crafted HPACK header block that fills the decoder's dynamic-table ring buffer past its capacity, tripping an internal CHECK() assertion and aborting the process. The flaw affects any bRPC deployment that exposes an HTTP/2 or gRPC (h2) listener to an untrusted network, since HPACK decoding happens before any Authenticator check and the h2 parser is enabled by default on bRPC ports. Impact is limited to availability (process termination); no code execution or data exposure is described, and no public exploit code was identified at time of analysis. This is one of 35 fixes bundled into the bRPC 1.18.0 security release, which also addresses related h2 resource-consumption and decompression issues.