CVE Vulnerability Catalog

In NLnet Labs Unbound up to and including version 1.26.1, when configured with both a root trust anchor and an additional 'trust-anchor:' for an island-of-security zone whose parent is provably insecure under the root chain, a malicious actor who can deliver one forged NXDOMAIN response for a name inside the island, with the NSEC/SOA RRSIG 'SignerName' set to the insecure parent, tricks Unbound's validator to use the root trust anchor instead, and consecutively the insecure parent, thus causing that record to be treated as insecure and denying existence for records in the trust island.

In NLnet Labs Unbound up to and including version 1.26.1, an observable cache-level side-channel can be exposed with 'unwanted-reply-threshold > 0' configurations (not a default configuration). When a response-like UDP datagram reaches the upstream source port currently used by a recursive query but its TXID does not match any pending query, Unbound counts it as an unwanted reply and, upon reaching the threshold, clears the worker's RRset and message caches, allowing an external observer to distinguish whether a candidate port matches the current upstream source port.

A flaw was found in the lm-evaluation-harness component of Red Hat OpenShift AI. An authenticated user able to submit evaluation jobs can supply a malicious model URL in JobSpec, leading to disclosure of the tenant's model authentication secret and enabling SSRF attacks against internal network endpoints.

In OpenStack Cinder through 29.0.0, cross-project information disclosure can occur in the Messages API. A non-admin user with the project-scoped 'member' or 'reader' role making the Block Storage API v3.3 "List messages" call finds non-expired user messages for other projects included in the response, resulting in leakage of UUIDs of resources owned by other projects.

A flaw was found in libmodulemd. In the module stream v2 rpm-map emission path, an inner loop incorrectly uses the count of digest algorithms as its bound instead of the number of checksums in the current digest table. Crafted modulemd v2 metadata with multiple digest types can trigger an out-of-bounds read during YAML serialization, typically causing a crash and resulting in a denial of service.

A flaw was found in libmodulemd. In builds with rpmio support enabled, compressed modulemd YAML is decompressed through a custom read handler that does not enforce a cumulative decompressed-size limit before parsing. An attacker who can supply a highly compressible compressed YAML file can cause excessive CPU and memory consumption, resulting in a denial of service.

A flaw was found in libmodulemd where parsing the xmd field in module metadata lacks recursion-depth and node-count limits. A crafted modulemd YAML with deeply nested xmd content can cause stack exhaustion or excessive resource consumption.

A SQL injection weakness was identified in projectworlds Advanced Library Management System 1.0 in the mysqli_query function of report_search.php. Manipulating the dateto/datefrom arguments allows SQL injection.

A path traversal weakness was identified in Tenstorrent tt-metal up to 0.80.0 in the load_hf_state_dict_filtered function of models/tt_transformers/tt/load_checkpoints.py. Manipulating the weight_map argument allows path traversal.

In Vyper 0.3.4 through 0.5.0b1, create_from_blueprint computes the remaining blueprint size via uint256 subtraction without checking that code_offset is below the code size. A huge offset can wrap to a small positive value.

Vyper 0.3.8 through 0.5.0b1 contains an optimizer miscompilation where (msg.value | K) or (len(msg.data) | K) with nonzero K compiles to constant 1 in a ternary branch under -O gas or -O codesize. Affected contracts use the wrong value without warning.

Vyper 0.1.0b1 before 0.1.0b17 and 0.4.1 through 0.5.0b1 builds its source-line table with str.splitlines(), which treats characters like form feed and U+2028 as line breaks ignored by the Python parser, causing numeric literals to be classified from shifted source text.

Vyper 0.1.0b1 before 0.1.0b12, 0.1.0b14, and 0.4.1 through 0.5.0b1 mix character-based line offsets with Python's UTF-8 byte col_offset values in vyper/ast/parse.py, so multi-byte characters shift source slices used to reconstruct later numeric literals.

Astral uv before 0.11.6 contains a path traversal vulnerability that allows attackers to delete arbitrary files by supplying wheels with relative traversal paths in RECORD entries. A malicious wheel can reference files outside the installation prefix.

adm-zip before 0.6.1 contains an incorrect permission assignment vulnerability that allows attackers to set setuid, setgid, or sticky bits on extracted files when keepOriginalPermission is enabled. Crafted zip external file attributes pass unmasked into chmod.

adm-zip before 0.6.1 contains an inconsistent entry resolution vulnerability that allows attackers to bypass content validation by supplying ZIP archives with duplicate entry names. getEntry() returns the last duplicate while extractAllTo() writes the first.

adm-zip 0.6.0 contains a decompression bomb protection bypass that allows attackers to evade the maxOutputLength guard by targeting the asynchronous getDataAsync() path in methods/inflater.js. Attackers can craft ZIP entries with falsified small declared sizes and high-ratio DEFLATE payloads to exhaust memory in applications reading entries asynchronously.

Coraza before 3.8.1 extracts multipart upload filenames with Go's mime.ParseMediaType, which silently ignores RFC 5987 filename* parameters whose charset is not utf-8 or us-ascii. Unauthenticated attackers can send a decoy filename with filename*=iso-8859-1''shell.php, or filename* alone, so FILES-based rules never see the real name, letting dangerous uploads reach backends that honor filename*.

Coraza from 3.0.4 before 3.8.1 contains a request body inspection bypass because request-body processor selection compares the entire Content-Type value exactly, so URL-encoded bodies whose Content-Type carries a media-type parameter are never parsed. Unauthenticated attackers can send application/x-www-form-urlencoded; charset=UTF-8 to keep ARGS_POST and REQUEST_BODY empty, evading body rules while the backend still processes the payload.

Coraza from 3.0.0 before 3.8.1 has a JSON body processor that flattens nested properties into dot-separated ARGS_POST names without escaping literal dots, so distinct properties collide. Unauthenticated attackers can add a later dotted property such as account.role that overwrites a malicious nested value, hiding SQL injection payloads from OWASP CRS rules while backends still receive them.