| CVE |
Vendors |
Products |
Updated |
CVSS v3.1 |
| vLLM before 0.29.0 contains a resource-limit bypass vulnerability in PyNvVideoCodec decoder allocation where sampler subclass shadowing allows independent counter increments. Unauthenticated attackers can select different sampler subclasses in video requests to exceed configured decoder limits and exhaust unaccounted GPU memory. |
| vLLM before 0.29.0 accepts user-controlled stop_token_ids on the OpenAI-compatible POST /v1/completions and POST /v1/chat/completions endpoints but validates only that the values are integers, not that each token id is within the model vocabulary/logits range. When min_tokens > 0, the stop token ids are used as logits indices to suppress stop tokens, so an out-of-range id reaches a CUDA indexing operation (index_put_) and triggers a device-side assertion. An authenticated API user can send a single malformed completion request that returns 500 Internal Server Error and puts EngineCore into a fatal state, causing subsequent requests to fail until the service is restarted (denial of service). |
| Netty's HTTP/3 codec (io.netty:netty-codec-http3) versions 4.2.0.Final through 4.2.17.Final contain an uncontrolled resource consumption vulnerability in the QPACK encoder-stream instruction decoder (QpackEncoderHandler, installed on the peer-initiated unidirectional QPACK encoder stream, type 0x02). The handler accepts an attacker-declared string-literal length of up to Integer.MAX_VALUE (~2 GiB) for the Name Length and Value Length fields of the "Insert With Literal Name" instruction (RFC 9204 §4.3.3), with no per-instruction or per-literal length cap and no cumulation-size limit; the existing HTTP/3 limits (maxHeaderListSize, maxUnknownFramePayloadLength, DEFAULT_MAX_FIELD_SECTION_SIZE) are not applied to this handler. A remote, unauthenticated peer with an established HTTP/3 connection to a default Netty HTTP/3 server can declare a very large literal length and then trickle fewer bytes than declared, causing the ByteToMessageDecoder MERGE cumulator to retain and grow the per-connection buffer, and ultimately triggering a large byte-array allocation. This leads to unbounded per-connection heap growth and OutOfMemoryError, resulting in denial of service. Fixed in 4.2.18.Final. |
| A stack-based buffer overflow vulnerability in the CGI program of Zyxel GS1900-48HPv2 firmware versions through 2.90(ABTQ.1)C0 could allow a LAN-based, unauthenticated attacker to exploit the flaw and potentially execute OS commands via a crafted HTTP request. |
| D-Link DAP-2610 up to 2.06B08r099 contains an authenticated command injection vulnerability within the web interface at the /index.xgi endpoint. An attacker with authenticated access can exploit some parameters to execute arbitrary system commands. |
| Piwigo before v16.4.0 is vulnerable to arbitrary file read and remote code execution in image upload handling when using the Imagick library due to insufficient validation and unsafe processing of user-supplied image files. By abusing format confusion (e.g., disguising SVG content as PNG), an attacker can trigger unintended interpretation of embedded SVG elements that reference local files. In more advanced scenarios, the Imagick support for Magick Scripting Language (MSL) may be abused to process attacker-controlled instructions, potentially leading to unauthorized server-side file writes and remote code execution, depending on configuration. This has been patched in 16.4.0. |
| Cross Site Scripting vulnerability in Netgate pfSense 26.03.1-RELEASE allows an attacker to execute arbitrary code via the pfBlockerNG package |
| A heap use-after-free flaw was found in Gnumeric. When a user opens a crafted Gnumeric workbook containing a malformed SheetObjectComponent element, the XML parser can dereference a freed sheet-object component, causing Gnumeric to crash. |
| OpenEye Apex Network Video Recorder (NVR) firmware 3.2.9.376 contains an OS command injection vulnerability in recbackup. An authenticated administrator can supply crafted backup-area configuration input that is passed to a shell command, allowing commands to execute with the privileges of the nvr user. The underlying design has been present since at least firmware 2.2.3.4.
This vulnerability is resolved in OpenEye Apex version 3.4.3. |
| The Onion module in AIL Framework contained a performance shortcut in its URL extraction logic that accepted URLs as valid .onion targets based solely on a length check (exactly 69 characters) and a suffix check (ending in ".onion"), without performing proper hostname parsing or onion-domain validation. An unauthenticated attacker who could publish or control web content crawled by the framework could embed a crafted URL containing an IP address or non-onion hostname with a path ending in ".onion" that satisfied the length and suffix conditions. Such a URL would be extracted, its domain naively sliced from the string, and queued as a legitimate onion crawler task. This allowed unauthenticated content publishers to inject arbitrary non-onion targets into the crawler's task queue, influencing crawler behavior and potentially directing it toward unintended network resources. The vulnerability required no authentication, no user interaction, and only the ability to place crafted content in a location the framework would crawl. The security impact is a loss of integrity in the crawler's target selection: the framework processes and acts upon URLs that do not correspond to legitimate .onion services. |
| Cotonti through 1.0.0 contains a reflected cross-site scripting vulnerability in the search plugin highlight parameter that performs no HTML or JavaScript escaping. Attackers can craft malicious links with injected JavaScript in the highlight parameter that executes in the browser of any visitor who opens the link, including administrators. |
| vm2 through 3.11.6 does not normalize `node:`-prefixed builtin specifiers when evaluating user-supplied negative (deny) entries in a NodeVM wildcard require policy. Although NodeVM strips the `node:` prefix during require() resolution, negative wildcard entries are matched by exact string comparison against the canonical builtin names, so a policy such as `new NodeVM({ require: { builtin: ['*', '-node:child_process'] } })` fails to deny the canonical `child_process` module. Sandboxed code can therefore obtain the host `child_process` builtin via `require('child_process')` or `require('node:child_process')`, gaining references to process-spawning APIs such as execSync and spawn, which is equivalent to host command-execution capability for untrusted sandbox code. Fixed in vm2 3.11.7. |
| vm2 is a sandbox library for running untrusted JavaScript in Node.js. In versions >= 3.10.0 and <= 3.11.7, Promises returned from the host realm into the sandbox are not marked as handled at the bridge boundary; only Promises created inside the sandbox are wrapped with a rejection-swallowing handler (lib/setup-sandbox.js), and the bridge only installs host-side rejection sanitizers when sandbox code calls .then/.catch/.finally. As a result, code running in the sandbox can invoke a host function that returns a rejected Promise (for example events.once() exposed via the NodeVM events builtin, or any embedder-provided Promise-returning API) and simply ignore the return value, leaving the host Promise unhandled so that Node.js's default unhandled-rejection behavior terminates the host process. This is an incomplete fix of GHSA-hw58-p9xv-2mjh. The issue is fixed in version 3.11.8. |
| PLANET IGS-5225-8P2T4S industrial managed switch V1 and V2 firmware versions before 1.2412b260707 and 2.2412b260519 contain a stack-based buffer overflow in the web server. Insufficient bounds checking on data copied into a stack buffer allows a remote administrator to cause a denial of service or potentially execute arbitrary code on the underlying operating system. |
| ClipBucket v5 before 5.5.3-#182 contains a file upload vulnerability that allows authenticated users to achieve remote code execution by uploading a PHP file with valid image magic bytes through the photo upload endpoint. The FileUpload::manageFile() function in fileupload.class.php fails to update the file extension after MIME validation, allowing an attacker-controlled .php extension to persist on disk and execute as PHP via PHP-FPM when the uploaded file is retrieved. |
| froxlor is a server administration panel. In versions 2.3.10 and earlier, Validate::validateUrl rejects carriage return and line feed characters only in the path, query and fragment components returned by parse_url, and never inspects the userinfo (user:pass@) components. This is an incomplete fix for GHSA-c3p2. An authenticated low-privilege customer with subdomain-create rights (no admin or change_serversettings privilege required) can supply a subdomain redirect URL that carries a CR/LF payload in the userinfo portion (e.g. http://user%0areturn 200 "pwned";%0a@evil.com/). The value passes validation, survives IDNA encoding, and is written verbatim into the generated nginx or Apache vhost configuration, allowing the attacker to break out of the emitted directive and inject arbitrary web-server configuration lines. froxlor regenerates and reloads the web-server configuration as root, so the injected directives take effect server-wide and can hijack responses or read local files. The issue is fixed in version 2.3.12. |
| Adminer 6.0.0 through 6.0.1, when the official ClickHouse driver plugin (plugins/drivers/clickhouse.php, rewritten in 6.0.0) is loaded, is vulnerable to pre-authentication server-side request forgery. An unauthenticated attacker can submit auth[driver]=clickhouse with auth[server] set to an arbitrary URL (for example http://127.0.0.1:18089), causing the Adminer server to issue an HTTP POST containing 'SELECT version()' to that host. In rootQuery(), if the target returns a status outside 200-299 (other than 401/403), the raw HTTP response body is assigned to the connection error and rendered on the login page, so the attacker receives the full response body of the internal service. This enables internal network/port reconnaissance and disclosure of sensitive information contained in internal error pages (stack traces, internal hostnames, file paths, configuration identifiers). Fixed in Adminer 6.0.2. |
| Hugo is a static site generator. In versions from v0.56.0 through v0.165.x, content files mapped to the text/org media type are rendered without escaping raw HTML: Org export blocks and @@html:...@@ snippets pass HTML through unescaped, resulting in cross-site scripting (XSS) in the generated site. An attacker who can supply or influence a content file under /content or the output of a content adapter can inject scripts that execute in the browsers of visitors to the affected pages. Only pages whose source file or content-adapter output declares the text/org media type are affected, and sites that fully trust all content sources are not impacted. Version v0.166.0 fixes the issue by introducing a security.allowContent allowlist that denies text/org by default; sites that intentionally author Org Mode content can opt back in with [security] allowContent = ['.*']. |
| Hugo versions from v0.162.0 before v0.166.0 contain a case-sensitive validation flaw in the security.http.urls IP-literal deny rule that allows attackers to bypass restrictions. Attackers can use mixed-case URL schemes in resources.GetRemote calls to fetch from restricted IP addresses like localhost. |
| Hugo versions from v0.161.0 through v0.165.0 run Node.js tools (css.PostCSS, css.TailwindCSS, js.Babel) under the Node.js permission model to restrict file system reads to the project directory and configured mounts. Because the Node.js permission model validates only the lexical path and follows symbolic links that point outside the allowed set, Hugo did not detect symlinks escaping the sandbox. An attacker who can contribute content to a Hugo project (for example via a pull request) can commit a symlink such as assets/css/x.css -> /etc/passwd together with a PostCSS plugin that reads it, allowing any file readable by the Hugo build process to be disclosed and potentially embedded in the published site. This affects builds using the default security configuration; projects that do not invoke Node.js tools are unaffected. Fixed in v0.166.0, which scans allowed paths and fails the build when a symbolic link resolves outside them. |
