Search

Search Results (374307 CVEs found)

CVE Vendors Products Updated CVSS v3.1
CVE-2026-17594 1 Sonatype 1 Nexus Repository Manager 2026-08-07 N/A
Nexus Repository 3 CE/Pro versions 3.0.0 through 3.94.x contain an incorrect authorization vulnerability (CWE-863) in the repository-creation user interface. An individual user account holding a delegated repository-admin privilege scoped to a specific repository format could create a repository of a different, unauthorized format, because authorization was checked against one request field while a separate, attacker-controlled field determined the repository format actually created. This does not affect the anonymous user, which cannot hold this privilege by default. Fixed in version 3.95.0.
CVE-2026-43632 1 Ggml-org 1 Llama.cpp 2026-08-07 8.1 High
llama.cpp builds b7492 through the latest b9060 contains a use-after-free vulnerability in llama-server affecting six tokenization endpoints (/tokenize, /detokenize, /infill, /apply-template, /rerank, and /anthropic/count_tokens) that bypass the task queue and access ctx_server.vocab directly on HTTP worker threads. Attackers can exploit a time-of-check-time-of-use race condition where the main thread destroys and frees vocab after the synchronization lock is released but before the handler finishes using it, causing a crash or potential code execution when --sleep-idle-seconds is configured.
CVE-2026-43631 1 Ggml-org 1 Llama.cpp 2026-08-07 8.1 High
llama.cpp builds b7492 through the latest b9060 contains a use-after-free vulnerability in the vocab pointer of llama-server when the --sleep-idle-seconds feature is enabled, allowing unauthenticated remote attackers to execute arbitrary code. Attackers can trigger the vulnerability by sending requests to affected endpoints while the server transitions to sleep mode, causing concurrent worker threads to dereference a freed vocab pointer that can be reclaimed with attacker-controlled data to achieve remote code execution.
CVE-2026-43630 1 Ggml-org 1 Llama.cpp 2026-08-07 6.5 Medium
llama.cpp builds b5702 through b7653 contain an out-of-bounds read vulnerability in the recurrent memory state restore path that allows attackers with write access to the slot save directory to read memory past the end of the allocated cells array. Attackers can craft a malicious slot file with an oversized seq_id value to trigger an out-of-bounds read that leaks heap data including pointer values into server logs, defeating ASLR protections and facilitating further exploitation.
CVE-2026-43629 1 Ggml-org 1 Llama.cpp 2026-08-07 8.1 High
llama.cpp builds b4882 through b9058 contain a heap buffer overflow vulnerability in the KV cache state restore path where the state_read_data() function computes write size without overflow checking, allowing attackers with write access to the slot_save_path directory to corrupt heap memory. Attackers can craft malicious state files where cell_count multiplication overflows or exceeds tensor buffer allocation to write attacker-controlled bytes past buffer boundaries, potentially resulting in heap metadata corruption, model weight corruption, or arbitrary code execution via function pointer overwrite.
CVE-2026-43628 1 Ggml-org 1 Llama.cpp 2026-08-07 7.8 High
llama.cpp builds b3978 through b9058 contain an integer underflow and out-of-bounds read vulnerability in the DRY sampler that allows unauthenticated attackers to trigger a heap buffer underflow by sending a crafted HTTP request with dry_allowed_length set to INT32_MIN to the /v1/completions or /v1/chat/completions endpoints. Attackers can exploit this vulnerability to crash the server with SIGSEGV causing denial of service for all connected users, or corrupt token sampling probabilities by reading garbage values from memory before the allocated buffer.
CVE-2026-43627 1 Ggml-org 1 Llama.cpp 2026-08-07 7.8 High
llama.cpp builds b1283 through b9058 contain an integer overflow vulnerability in the llama_batch_init() function where unchecked multiplications in malloc() calls can wrap past INT32_MAX when computing allocation sizes. Attackers can pass specially crafted parameters to trigger integer overflow, causing heap corruption and potentially achieving arbitrary code execution through subsequent batch operations that write past allocated buffer boundaries.
CVE-2026-17435 2026-08-07 N/A
File::Rotate::Simple versions before 0.4.0 for Perl create the target of dangling symlinks when rotating files. When the file to be rotated is a symbolic link to a missing file, and the touch option is enabled, then the rotate method assumes that the file is absent (since the existence check is against the target), and does not rotate it. But it touches the file, which creates the target. An attacker that has the ability to create the symlink can use this to create an arbitrary file with permissions of the process rotating the files (which may be different from the process that normally writes to the log file that is being rotated). Note that the touch option is disabled by default.
CVE-2026-70640 1 Ggml-org 1 Llama.cpp 2026-08-07 7 High
llama.cpp builds b1886 through b7445 contain a race condition use-after-free vulnerability in the LLaMA-Android JNI wrapper where bench_1model() and free_1context() lack synchronization, allowing Thread A to operate on freed memory while Thread B concurrently frees the llama_context. Attackers can exploit this by performing heap spray with attacker-controlled data containing a fake vtable to hijack the vtable pointer at offset +0x30, causing llama_batch_allocr::clear() to dereference arbitrary memory and achieve remote code execution.
CVE-2026-70639 1 Ggml-org 1 Llama.cpp 2026-08-07 5.5 Medium
llama.cpp builds b1886 through b7445 contain a null pointer dereference vulnerability in the LLaMA-Android JNI wrapper where the bench_1model() function fails to validate the model context pointer before dereferencing it. Attackers can supply a malicious, corrupt, or truncated model file to trigger a null context condition, causing a SIGSEGV crash that terminates the Android application process and results in denial of service.
CVE-2025-71409 2026-08-07 7.1 High
Lack of authentication for Very High Frequency Data Link messages allows rogue ground stations to inject CPDLC messages leading to unexpected or misleading clearances and potential pilot confusion. This type of attack can be carried out remotely over radio frequency.
CVE-2026-70638 1 Ggml-org 1 Llama.cpp 2026-08-07 7.8 High
llama.cpp builds b1886 through b7445 contain an integer overflow vulnerability in the LLaMA-Android JNI wrapper where the new_1batch() function multiplies sizeof(llama_seq_id) by an attacker-controlled n_seq_max parameter without overflow validation, causing heap buffer allocation to wrap and allocate insufficient memory. Attackers can exploit this by providing a crafted n_seq_max value through a malicious model file or JNI call to trigger heap corruption and achieve denial of service or arbitrary code execution on Android applications using the LLaMA-Android binding.
CVE-2026-43622 1 Ggml-org 1 Llama.cpp 2026-08-07 7.8 High
llama.cpp builds b1886 through b7445 contain a double free vulnerability in the LLaMA-Android JNI wrapper where new_1batch() allocates memory using malloc() while free_1batch() deallocates it using the C++ delete operator, causing heap metadata corruption. Attackers can trigger this memory management mismatch to cause denial of service through process crashes or potentially achieve arbitrary code execution depending on allocator state.
CVE-2025-71410 2026-08-07 5.3 Medium
Unnumbered Disconnect (U DISC) and malformed Aviation Very High Frequency Link Control frames can terminate sessions and lead to a loss of CPDLC functions requiring a reversion to voice communication and increased controller workload. This type of attack can be carried out remotely over radio frequency.
CVE-2025-71411 2026-08-07 5.3 Medium
Broadcast control frames can disconnect multiple aircraft simultaneously leading to delayed clearances and air traffic controller overload. This type of attack can be carried out remotely over radio frequency.
CVE-2025-71412 2026-08-07 7.1 High
Injection of false emergency or status messages over CPDLC may lead to misallocation of resources, operational confusion, and improper response actions by flight crews, traffic controllers, and ground operations. This type of attack can be carried out remotely over radio frequency.
CVE-2025-71413 2026-08-07 5.3 Medium
Malformed or out-of-sequence frames at the Aviation Very High Frequency Link Control X.25 layers cause repeated resets which may result in increased workload and reduced situational awareness. This type of attack can be carried out remotely over radio frequency.
CVE-2026-15149 2 Wordpress, Wp Hotel Booking 2 Wordpress, Wp Hotel Booking 2026-08-07 5.3 Medium
The WP Hotel Booking WordPress plugin before 2.3.3 does not ensure that room quantities and the resulting order total are non-negative when placing a booking, and relies on client-controlled cart data, allowing unauthenticated users to create confirmed reservations for free or at an arbitrarily reduced price.
CVE-2022-4995 2026-08-07 9.8 Critical
Weaver (Fanwei) E-cology 9.0 versions prior to 10.52 contain a file upload vulnerability that allows a remote, unauthenticated attacker to upload arbitrary files, including JSP webshells, by submitting a multipart/form-data POST request to /workrelate/plan/util/uploaderOperate.jsp with arbitrary secId and plandetailid field values. Successful exploitation results in remote code execution under the privileges of the application server process. Exploitation evidence was first observed by the Shadowserver Foundation on 2023-10-14 (UTC).
CVE-2026-71851 2026-08-07 9 Critical
crypto-js is a JavaScript library of crypto standards. Versions of crypto-js prior to 4.0.0 generate randomness in CryptoJS.lib.WordArray.random() using a custom variation of the Multiply-With-Carry pseudorandom number generator, seeded from Math.random(), instead of a cryptographically secure source. This generator was introduced in version 3.1.2-4 and remained present in nearly every 3.x release. Nominal requests for 128 or 256 bits of entropy through this function produce effective search spaces of approximately 2 to the 39th and 2 to the 47th possibilities, small enough to enumerate on commodity hardware. Downstream wallet applications that used CryptoJS.lib.WordArray.random() as the entropy source for BIP39 recovery phrases are affected, and an attacker who enumerates the reduced output space can recover the resulting private keys and control the associated funds. This issue is fixed in version 4.0.0.