| CVE |
Vendors |
Products |
Updated |
CVSS v3.1 |
| In the Linux kernel, the following vulnerability has been resolved:
ntfs: fix kmap_local leak in write_mft_record_nolock() error paths
write_mft_record_nolock() maps the MFT record folio with
kmap_local_folio(), but the pre_write_mst_fixup() and
bio_add_folio() failure paths jump to the error label without
unmapping it. kmap_local mappings are stack-ordered per task, so
leaking one corrupts the nesting for any outer mapping.
Unmap the folio on those error paths too. |
| vLLM is an inference and serving engine for large language models. Prior to 0.32.0, the Phi-4 multimodal audio encoder in phi4mm_audio.py and phi4mm_utils.py uses eager attention by default, allowing a remote client to submit one legal long audio clip that causes MultiHeadedAttention.forward_attention to materialize and retain full attention matrices in self.attn. The scaled dot-product attention branch does not materialize these matrices. The resulting quadratic GPU-memory use can cause an out-of-memory condition and terminate EngineCore for all users. This issue is fixed in version 0.32.0. |
| vLLM is an inference and serving engine for large language models. From 0.19.0 until 0.32.0, authenticated callers can supply a request chat template when no server-side --chat-template is configured, or a malicious model repository can provide a template that is compiled and rendered by tokenizer.apply_chat_template() without a wall-clock timeout, output-size limit, or iteration bound. Rendering runs in a ThreadPoolExecutor with one worker by default, so a nonterminating render can block /v1/responses and other chat-template users after the request ends, while an output-expanding render can exhaust server memory. Harmony models are not affected because they do not use this Jinja path. This issue is fixed in version 0.32.0. |
| The Bluetooth host pre-allocates a bt_conn object (a "reservation") from the fixed acl_conns[] pool for every connectable advertising set it starts, in le_adv_start_add_conn() in subsys/bluetooth/host/adv.c. When an incoming peripheral connection completes, find_pending_connect() in subsys/bluetooth/host/hci_core.c looked that reservation up using only the identity and the peer address, with a directed-then-undirected fallback, and had no knowledge of which advertising set actually terminated. With CONFIG_BT_EXT_ADV_MAX_ADV_SET > 1, where the LE Enhanced Connection Complete event is cached until the LE Advertising Set Terminated event identifies the set, this lookup can return the reservation belonging to a different, still-running advertising set.
The defect is reachable over the air by any adjacent device. If an application runs two connectable extended advertising sets on the same identity — one undirected and one directed at a peer P (the common "directed reconnect to a bonded central while still accepting new centrals" pattern) — a remote initiator whose address maps to P connecting on the undirected set causes the directed set's reservation to be consumed instead. The initiator address is chosen freely by the remote device and is not authenticated at link establishment, so no bonding, pairing or prior relationship is required to trigger the mis-attribution.
The undirected set's own reservation is then stranded: its BT_ADV_ENABLED flag has already been cleared, and le_adv_stop_free_conn() — the only path that returns a reservation to the pool — is reached only for sets that are still enabled, so bt_le_ext_adv_stop() returns early and the object is never released. Each occurrence permanently consumes one CONFIG_BT_MAX_CONN slot until bt_disable() or reboot. In application patterns that restart the directed advertising set after each disconnect, an attacker can repeat the sequence until the pool is exhausted, after which connectable advertising fails with -ENOMEM and the device can no longer accept connections. As a side effect the directed set keeps advertising without a reservation, so a genuine connection on it is dropped by bt_hci_disconnect(). The impact is availability only; no memory corruption or information disclosure results, and configurations with a single advertising set are unaffected. |
| HashiCorp Consul and Consul Enterprise are vulnerable to a denial of service in the HTTP API that may allow an unauthenticated attacker to exhaust memory on a Consul agent. This vulnerability (CVE-2026-105626) is fixed in Consul Community Edition 2.1.0 and Consul Enterprise 2.1.0 and 1.21.19. |
| vLLM is an inference and serving engine for large language models. From 0.16.0 until 0.32.0, the /v1/responses endpoint allows an authenticated caller to place an attacker-controlled template in chat_template_kwargs[chat_template] when no server template is configured. ChatParams.get_apply_chat_template_kwargs() preserves that value without enforcing trust_request_chat_template, so the request can bypass the intended trust policy and cause CPU or memory exhaustion through Jinja rendering. This issue is fixed in version 0.32.0. |
| vLLM from 0.18.0 before 0.31.0 contains a resource exhaustion vulnerability that allows API clients to bypass the video/jpeg frame-count cap by overriding num_frames through media_io_kwargs. Attackers can set num_frames to -1 with a large inline video so VideoMediaIO.load_base64 decodes every frame, exhausting memory and crashing the API frontend. |
| ImageMagick before 6.9.13-58 and 7.x before 7.1.2-33 contains a memory leak in the handling of the -cdl (color decision list) option, where allocated memory is not cleaned up after use. Attackers can supply input that causes the -cdl option to be processed, repeatedly leaking memory and degrading availability through resource exhaustion. |
| ImageMagick before 6.9.13-58 and 7.x before 7.1.2-33 contains a memory leak in the handling of XMP profiles by various decoders. Attackers can supply images with a crafted XMP profile that leak memory each time they are processed, leading to resource exhaustion and denial of service. |
| ImageMagick before 6.9.13-58 and 7.x before 7.1.2-33 contains a memory leak in the JNG encoder that fails to release allocated memory when processing a crafted image. Attackers can supply crafted images that are encoded to JNG to leak memory, potentially causing resource exhaustion and a limited denial of service. |
| ImageMagick before 6.9.13-58 and 7.x before 7.1.2-33 contains a policy bypass in the PS, PDF and XPS decoders that allows the security policy's list length limit on scene selection to be circumvented. Attackers supplying a specific argument can cause ImageMagick to process more scenes than permitted, potentially exhausting resources. |
| A vulnerability was determined in NocoDB up to 2026.07.0. This vulnerability affects the function UserRefreshToken.deleteAllUserToken of the file /auth/user/signup of the component PublicApiLimiterGuard. This manipulation causes allocation of resources. The attack is possible to be carried out remotely. The attack is considered to have high complexity. It is stated that the exploitability is difficult. Code-verified absence of in-app rate limiting on the unauthenticated signup and password-forgot endpoints and per-request refresh/OAuth token invalidation, but the vendor documents the pass-through limiter as intentional design. The vendor deleted the GitHub issue for this vulnerability without any explanation. Afterwards the vendor was contacted early about this disclosure via email but did not respond in any way. |
| A vulnerability was determined in QuantGeekDev docker-mcp 0.1.0/0.2.0. The impacted element is the function _build_windows_command of the file docker_executor.py. This manipulation of the argument project_name causes os command injection. It is possible to initiate the attack remotely. |
| A weakness has been identified in dbcli mycli up to 2.29.4/2.30.1. The impacted element is the function source_special_command_is_safe of the file mycli/packages/special/source.py of the component Whitelist. Executing a manipulation can lead to os command injection. It is possible to launch the attack on the local host. The exploit has been made available to the public and could be used for attacks. The vendor was contacted early about this disclosure but did not respond in any way. |
| Gitoxide gix-pack before 0.75.0 contains an uncontrolled memory allocation vulnerability that allows remote attackers to abort the process by supplying a huge pack-header object count. Malicious remotes can serve a 12-byte pack header claiming 0xFFFFFFFF objects, forcing Tree::with_capacity to reserve hundreds of gigabytes and crash clone or fetch operations. |
| In the Linux kernel, the following vulnerability has been resolved:
spi: mtk-snfi: unregister ECC engine on probe failure and remove() callback
mtk_snand_probe() registers the on-host NAND ECC engine, but teardown was
missing from both probe unwind and remove-time cleanup. Add a devm cleanup
action after successful registration so
nand_ecc_unregister_on_host_hw_engine() runs automatically on probe
failures and during device removal. |
| In the Linux kernel, the following vulnerability has been resolved:
PCI: mediatek-gen3: Prevent leaking IRQ domains when IRQ not found
In mtk_pcie_setup_irq(), the IRQ domains are allocated before the
controller's IRQ is fetched. If the latter fails, the function
directly returns an error, without cleaning up the allocated domains.
Hence, reverse the order so that the IRQ domains are allocated after the
controller's IRQ is found.
This was flagged by Sashiko during a review of "[PATCH v6 0/7] PCI:
mediatek-gen3: add power control support". |
| Allocation of Resources Without Limits or Throttling vulnerability in hackney_conn in benoitc hackney allows a malicious HTTP server to exhaust client memory via an oversized or highly compressed response body. hackney:request/5 and hackney:body/1 return the whole response body, and nothing bounds how much the connection process buffers. Any server the client requests can trigger it.
hackney_conn:read_full_body/2 and the HTTP/2 and HTTP/3 accumulators append every chunk until the body ends. With {auto_decompress, true} the buffered body is then inflated in one step with zlib:gunzip/1 or zlib:inflate/2, which return the whole output at once.
This issue affects hackney: from 2.0.0-beta.1 before 4.9.0. |
| MoviePilot through 3.1.5 contains a command injection vulnerability that allows attackers to execute arbitrary OS commands by embedding malicious instructions in untrusted content processed through tools like browse_webpage, search_web, and torrent metadata parsers, which then get passed to the unrestricted ExecuteCommandTool via asyncio.create_subprocess_shell. Attackers can craft content that instructs the model to invoke execute_command with confirm_dangerous set to true, bypassing the validate_command_safety denylist and exploiting the admin session's is_admin context to achieve remote code execution on the host. |
| Allocation of Resources Without Limits or Throttling vulnerability in the HTTP/2 request body reader of mtrudel bandit allows an unauthenticated remote attacker to hold stream processes open indefinitely via WINDOW_UPDATE or empty SETTINGS frames sent while a Plug reads the request body.
Bandit.HTTP2.Stream.do_read_data/4 in lib/bandit/http2/stream.ex waits for each frame with a fresh read_timeout. A stream WINDOW_UPDATE, or the zero-delta window update that every empty SETTINGS frame delivers to all streams, ends the wait without delivering any body data, and the loop starts a new full countdown. A client that sends one of these 9-byte frames before each countdown expires keeps read_body blocked, together with the stream process and whatever the application holds while reading, for as long as it likes.
This issue affects bandit: from 1.4.0 before 1.14.0. |
