Duplicate of CVE-2024-45806.

Envoy is a cloud-native, open source edge and service proxy. A theoretical request smuggling vulnerability exists through Envoy if a server can be tricked into adding an upgrade header into a response. Per RFC https://www.rfc-editor.org/rfc/rfc7230#section-6.7 a server sends 101 when switching protocols. Envoy incorrectly accepts a 200 response from a server when requesting a protocol upgrade, but 200 does not indicate protocol switch. This opens up the possibility of request smuggling through Envoy if the server can be tricked into adding the upgrade header to the response.

Envoy is an open source edge and service proxy designed for cloud-native applications. Prior to versions 1.26.0, 1.25.3, 1.24.4, 1.23.6, and 1.22.9, the client may bypass JSON Web Token (JWT) checks and forge fake original paths. The header `x-envoy-original-path` should be an internal header, but Envoy does not remove this header from the request at the beginning of request processing when it is sent from an untrusted client. The faked header would then be used for trace logs and grpc logs, as well as used in the URL used for `jwt_authn` checks if the `jwt_authn` filter is used, and any other upstream use of the x-envoy-original-path header. Attackers may forge a trusted `x-envoy-original-path` header. Versions 1.26.0, 1.25.3, 1.24.4, 1.23.6, and 1.22.9 have patches for this issue.

Envoy is a cloud-native edge/middle/service proxy. Prior to versions 1.34.1, 1.33.3, 1.32.6, and 1.31.8, Envoy's URI template matcher incorrectly excludes the `*` character from a set of valid characters in the URI path. As a result URI path containing the `*` character will not match a URI template expressions. This can result in bypass of RBAC rules when configured using the `uri_template` permissions. This vulnerability is fixed in Envoy versions v1.34.1, v1.33.3, v1.32.6, v1.31.8. As a workaround, configure additional RBAC permissions using `url_path` with `safe_regex` expression.

Envoy is a high-performance edge/middle/service proxy. Prior to 1.37.1, 1.36.5, 1.35.8, and 1.34.13, the Envoy RBAC (Role-Based Access Control) filter contains a logic vulnerability in how it validates HTTP headers when multiple values are present for the same header name. Instead of validating each header value individually, Envoy concatenates all values into a single comma-separated string. This behavior allows attackers to bypass RBAC policies—specifically "Deny" rules—by sending duplicate headers, effectively obscuring the malicious value from exact-match mechanisms. This vulnerability is fixed in 1.37.1, 1.36.5, 1.35.8, and 1.34.13.

Envoy Gateway is an open source project for managing Envoy Proxy as a standalone or Kubernetes-based application gateway. In all Envoy Gateway versions prior to 1.2.7 and 1.3.1 a default Envoy Proxy access log configuration is used. This format is vulnerable to log injection attacks. If the attacker uses a specially crafted user-agent which performs json injection, then he could add and overwrite fields to the access log. This vulnerability is fixed in 1.3.1 and 1.2.7. One can overwrite the old text based default format with JSON formatter by modifying the "EnvoyProxy.spec.telemetry.accessLog" setting.

Envoy is a cloud-native high-performance edge/middle/service proxy. In Envoy version 1.17.0 an attacker can bypass authentication by presenting a JWT token with an issuer that is not in the provider list when Envoy's JWT Authentication filter is configured with the `allow_missing` requirement under `requires_any` due to a mistake in implementation. Envoy's JWT Authentication filter can be configured with the `allow_missing` requirement that will be satisfied if JWT is missing (JwtMissed error) and fail if JWT is presented or invalid. Due to a mistake in implementation, a JwtUnknownIssuer error was mistakenly converted to JwtMissed when `requires_any` was configured. So if `allow_missing` was configured under `requires_any`, an attacker can bypass authentication by presenting a JWT token with an issuer that is not in the provider list. Integrity may be impacted depending on configuration if the JWT token is used to protect against writes or modifications. This regression was introduced on 2020/11/12 in PR 13839 which fixed handling `allow_missing` under RequiresAny in a JwtRequirement (see issue 13458). The AnyVerifier aggregates the children verifiers' results into a final status where JwtMissing is the default error. However, a JwtUnknownIssuer was mistakenly treated the same as a JwtMissing error and the resulting final aggregation was the default JwtMissing. As a result, `allow_missing` would allow a JWT token with an unknown issuer status. This is fixed in version 1.17.1 by PR 15194. The fix works by preferring JwtUnknownIssuer over a JwtMissing error, fixing the accidental conversion and bypass with `allow_missing`. A user could detect whether a bypass occurred if they have Envoy logs enabled with debug verbosity. Users can enable component level debug logs for JWT. The JWT filter logs will indicate that there is a request with a JWT token and a failure that the JWT token is missing.

Envoy is a high-performance edge/middle/service proxy. Prior to 1.37.1, 1.36.5, 1.35.8, and 1.34.13, calling Utility::getAddressWithPort with a scoped IPv6 addresses causes a crash. This utility is called in the data plane from the original_src filter and the dns filter. This vulnerability is fixed in 1.37.1, 1.36.5, 1.35.8, and 1.34.13.

Envoy is an open source edge and service proxy designed for cloud-native applications. Prior to versions 1.26.0, 1.25.3, 1.24.4, 1.23.6, and 1.22.9, Envoy does not sanitize or escape request properties when generating request headers. This can lead to characters that are illegal in header values to be sent to the upstream service. In the worst case, it can cause upstream service to interpret the original request as two pipelined requests, possibly bypassing the intent of Envoy’s security policy. Versions 1.26.0, 1.25.3, 1.24.4, 1.23.6, and 1.22.9 contain a patch. As a workaround, disable adding request headers based on the downstream request properties, such as downstream certificate properties.

Envoy is an open source edge and service proxy designed for cloud-native applications. Prior to versions 1.26.0, 1.25.3, 1.24.4, 1.23.6, and 1.22.9, escalation of privileges is possible when `failure_mode_allow: true` is configured for `ext_authz` filter. For affected components that are used for logging and/or visibility, requests may not be logged by the receiving service. When Envoy was configured to use ext_authz, ext_proc, tap, ratelimit filters, and grpc access log service and an http header with non-UTF-8 data was received, Envoy would generate an invalid protobuf message and send it to the configured service. The receiving service would typically generate an error when decoding the protobuf message. For ext_authz that was configured with ``failure_mode_allow: true``, the request would have been allowed in this case. For the other services, this could have resulted in other unforeseen errors such as a lack of visibility into requests. As of versions 1.26.0, 1.25.3, 1.24.4, 1.23.6, and 1.22.9, Envoy by default sanitizes the values sent in gRPC service calls to be valid UTF-8, replacing data that is not valid UTF-8 with a `!` character. This behavioral change can be temporarily reverted by setting runtime guard `envoy.reloadable_features.service_sanitize_non_utf8_strings` to false. As a workaround, one may set `failure_mode_allow: false` for `ext_authz`.

Envoy is an open source edge and service proxy designed for cloud-native applications. Prior to versions 1.26.0, 1.25.3, 1.24.4, 1.23.6, and 1.22.9, the OAuth filter assumes that a `state` query param is present on any response that looks like an OAuth redirect response. Sending it a request with the URI path equivalent to the redirect path, without the `state` parameter, will lead to abnormal termination of Envoy process. Versions 1.26.0, 1.25.3, 1.24.4, 1.23.6, and 1.22.9 contain a patch. The issue can also be mitigated by locking down OAuth traffic, disabling the filter, or by filtering traffic before it reaches the OAuth filter (e.g. via a lua script).

Envoy is an open source edge and service proxy designed for cloud-native applications. Compliant HTTP/1 service should reject malformed request lines. Prior to versions 1.26.0, 1.25.3, 1.24.4, 1.23.6, and 1.22.9, There is a possibility that non compliant HTTP/1 service may allow malformed requests, potentially leading to a bypass of security policies. This issue is fixed in versions 1.26.0, 1.25.3, 1.24.4, 1.23.6, and 1.22.9.

Envoy is a high-performance edge/middle/service proxy. External authentication can be bypassed by downstream connections. Downstream clients can force invalid gRPC requests to be sent to ext_authz, circumventing ext_authz checks when failure_mode_allow is set to true. This issue has been addressed in released 1.29.1, 1.28.1, 1.27.3, and 1.26.7. Users are advised to upgrade. There are no known workarounds for this vulnerability.

Netty is a network application framework for development of protocol servers and clients. Prior to versions 4.1.135.Final and 4.2.15.Final, before reading the first request-line, `HttpObjectDecoder` skips every byte for which `Character.isISOControl(b)` is `true` (0x00–0x1F and 0x7F) as well as all whitespace. RFC 9112 §2.2 only asks servers to ignore empty CRLF lines preceding the request-line — a carefully scoped robustness allowance intended to handle HTTP/1.0 POST workarounds. Silently absorbing NUL bytes, SOH, STX, and other non-CRLF control characters goes significantly beyond this, and can be exploited for request-boundary confusion in pipelined or multiplexed transports where a front-end component treats those bytes differently. Versions 4.1.135.Final and 4.2.15.Final patch the issue.

In solidus before versions 2.8.6, 2.9.6, and 2.10.2, there is an bility to change order address without triggering address validations. This vulnerability allows a malicious customer to craft request data with parameters that allow changing the address of the current order without changing the shipment costs associated with the new shipment. All stores with at least two shipping zones and different costs of shipment per zone are impacted. This problem comes from how checkout permitted attributes are structured. We have a single list of attributes that are permitted across the whole checkout, no matter the step that is being submitted. See the linked reference for more information. As a workaround, if it is not possible to upgrade to a supported patched version, please use this gist in the references section.

A flaw was found in Wildfly's implementation of Xerces, specifically in the way the XMLSchemaValidator class in the JAXP component of Wildfly enforced the "use-grammar-pool-only" feature. This flaw allows a specially-crafted XML file to manipulate the validation process in certain cases. This issue is the same flaw as CVE-2020-14621, which affected OpenJDK, and uses a similar code. This flaw affects all Xerces JBoss versions before 2.12.0.SP3.

Typecho v1.3.0 was discovered to contain a Client IP Spoofing vulnerability, which allows attackers to falsify their IP addresses by specifying an arbitrary IP as value of X-Forwarded-For or Client-Ip headers while performing HTTP requests.

IBM DataPower Gateway V10CD, 10.0.1, and 2108.4.1 could allow a remote attacker to bypass security restrictions, caused by the improper validation of input. By sending a specially crafted JSON message, an attacker could exploit this vulnerability to modify structure and fields. IBM X-Force ID: 209824.

In wpa_supplicant and hostapd 2.9, forging attacks may occur because AlgorithmIdentifier parameters are mishandled in tls/pkcs1.c and tls/x509v3.c.

Frontier is Substrate's Ethereum compatibility layer. Prior to commit number 0b962f218f0cdd796dadfe26c3f09e68f7861b26, a bug in `pallet-ethereum` can cause invalid transactions to be included in the Ethereum block state in `pallet-ethereum` due to not validating the input data size. Any invalid transactions included this way have no possibility to alter the internal Ethereum or Substrate state. The transaction will appear to have be included, but is of no effect as it is rejected by the EVM engine. The impact is further limited by Substrate extrinsic size constraints. A patch is available in commit number 0b962f218f0cdd796dadfe26c3f09e68f7861b26. There are no workarounds aside from applying the patch.

Vyper is a pythonic Smart Contract Language for the Ethereum virtual machine. In versions 0.3.10 and prior, using the `slice` builtin can result in a double eval vulnerability when the buffer argument is either `msg.data`, `self.code` or `<address>.code` and either the `start` or `length` arguments have side-effects. It can be easily triggered only with the versions `<0.3.4` as `0.3.4` introduced the unique symbol fence. No vulnerable production contracts were found. Additionally, double evaluation of side-effects should be easily discoverable in client tests. As such, the impact is low. As of time of publication, no fixed versions are available.

Apache Batik 1.13 is vulnerable to server-side request forgery, caused by improper input validation by the NodePickerPanel. By using a specially-crafted argument, an attacker could exploit this vulnerability to cause the underlying server to make arbitrary GET requests.

Apache XmlGraphics Commons 2.4 and earlier is vulnerable to server-side request forgery, caused by improper input validation by the XMPParser. By using a specially-crafted argument, an attacker could exploit this vulnerability to cause the underlying server to make arbitrary GET requests. Users should upgrade to 2.6 or later.

Jenkins Git Parameter Plugin 439.vb_0e46ca_14534 and earlier does not validate that the Git parameter value submitted to the build matches one of the offered choices, allowing attackers with Item/Build permission to inject arbitrary values into Git parameters.

Translate is a package that allows users to convert text to different languages on Node.js and the browser. Prior to version 3.0.0, an attacker controlling the second variable of the `translate` function is able to perform a cache poisoning attack. They can change the outcome of translation requests made by subsequent users. The `opt.id` parameter allows the overwriting of the cache key. If an attacker sets the `id` variable to the cache key that would be generated by another user, they can choose the response that user gets served. Version 3.0.0 fixes this issue.

In JetBrains IntelliJ IDEA before 2023.3.3 a plugin for JetBrains Space was able to send an authentication token to an inappropriate URL

Versions of the package mysql2 before 3.9.3 are vulnerable to Improper Input Validation through the keyFromFields function, resulting in cache poisoning. An attacker can inject a colon (:) character within a value of the attacker-crafted key.

A flaw was found in Hibernate Validator version 6.1.2.Final. A bug in the message interpolation processor enables invalid EL expressions to be evaluated as if they were valid. This flaw allows attackers to bypass input sanitation (escaping, stripping) controls that developers may have put in place when handling user-controlled data in error messages.

A vulnerability in Vercel’s AI SDK has been fixed in versions 5.0.52, 5.1.0-beta.9, and 6.0.0-beta. This issue may have allowed users to bypass filetype whitelists when uploading files. All users are encouraged to upgrade. More details: https://vercel.com/changelog/cve-2025-48985-input-validation-bypass-on-ai-sdk

An issue was discovered in Joomla! before 3.9.16. Missing length checks in the user table can lead to the creation of users with duplicate usernames and/or email addresses.

An issue was discovered on Samsung mobile devices with M(6.0) software. An attacker can disable all Sound functionality by broadcasting an unprotected intent. The Samsung IDs are SVE-2016-7179 and SVE-2016-7182 (November 2016).

Multiple vulnerabilities in the Application Level Gateway (ALG) for the Network Address Translation (NAT) feature of Cisco Adaptive Security Appliance (ASA) Software and Firepower Threat Defense (FTD) Software could allow an unauthenticated, remote attacker to bypass the ALG and open unauthorized connections with a host located behind the ALG. For more information about these vulnerabilities, see the Details section of this advisory. Note: These vulnerabilities have been publicly discussed as NAT Slipstreaming.

In PEPPERL+FUCHS WirelessHART-Gateway <= 3.0.8 a vulnerability may allow remote attackers to rewrite links and URLs in cached pages to arbitrary strings.

guzzlehttp/psr7 is a PSR-7 HTTP message library implementation in PHP. Versions prior to 2.10.2 did not reject ASCII control characters, whitespace, or DEL in first-party URI host components. A vulnerable flow is: First, an application accepts a user-controlled URL. Second, the URL is used to construct a PSR-7 `Uri` or `Request`. Third, the host component contains CRLF or another header-unsafe character. Fourth, the host is copied into the PSR-7 `Host` header when no explicit `Host` header is provided. Finally, the request is serialized or sent by an HTTP client that does not independently reject the malformed host. In that flow, an attacker can cause the serialized request to contain additional attacker-controlled header lines. For example, a host containing `"\r\nX-Injected: yes"` can cause the generated `Host` header to span multiple HTTP header lines. Applications are affected when they use user-controlled URLs for outbound HTTP requests, URL forwarding, proxying, crawling, webhook delivery, or similar request-dispatch flows. In deployments involving HTTP/1.1 connection reuse, proxies, gateways, or load balancers, this malformed request may also contribute to request smuggling or cache poisoning, depending on how downstream components parse the request. The issue is patched in `2.10.2` and later. `1.x` is end-of-life and will not receive a patch. As a workaround, validate and reject all untrusted URI strings before constructing PSR-7 `Uri` or `Request` instances. Reject input containing ASCII control characters, whitespace, or DEL, including CRLF, tab, space, NUL, or DEL characters. Applications that forward requests should also ensure the final HTTP client or serializer rejects invalid URI and header data before writing requests to the network.

The Responsive Blocks – Page Builder for Blocks & Patterns plugin for WordPress is vulnerable to Unauthenticated Open Email Relay in all versions up to, and including, 2.2.0. This is due to insufficient authorization checks and missing server-side validation of the recipient email address supplied via a public REST API route. This makes it possible for unauthenticated attackers to send arbitrary emails to any recipient of their choosing through the affected WordPress site's mail server, effectively turning the site into an open mail relay.

The documentation specifies that the BT_GATT_PERM_READ_LESC and BT_GATT_PERM_WRITE_LESC defines for a Bluetooth characteristic: Attribute read/write permission with LE Secure Connection encryption. If set, requires that LE Secure Connections is used for read/write access, however this is only true when it is combined with other permissions, namely BT_GATT_PERM_READ_ENCRYPT/BT_GATT_PERM_READ_AUTHEN (for read) or BT_GATT_PERM_WRITE_ENCRYPT/BT_GATT_PERM_WRITE_AUTHEN (for write), if these additional permissions are not set (even in secure connections only mode) then the stack does not perform any permission checks on these characteristics and they can be freely written/read.

hyper is an HTTP library for rust. hyper's HTTP/1 server code had a flaw that incorrectly parses and accepts requests with a `Content-Length` header with a prefixed plus sign, when it should have been rejected as illegal. This combined with an upstream HTTP proxy that doesn't parse such `Content-Length` headers, but forwards them, can result in "request smuggling" or "desync attacks". The flaw exists in all prior versions of hyper prior to 0.14.10, if built with `rustc` v1.5.0 or newer. The vulnerability is patched in hyper version 0.14.10. Two workarounds exist: One may reject requests manually that contain a plus sign prefix in the `Content-Length` header or ensure any upstream proxy handles `Content-Length` headers with a plus sign prefix.

MISP is an open source threat intelligence and sharing platform. Prior to 2.5.37, MISP Collections did not enforce RFC 4122 UUID validation on the uuid field. As a result, a user able to create or modify Collection records could submit malformed UUID values, potentially causing integrity issues or unexpected behaviour in code paths that assume Collection UUIDs are valid identifiers. This vulnerability is fixed in 2.5.37.

Ping Identity PingAccess before 5.3.3 allows HTTP request smuggling via header manipulation.

Spring MVC and WebFlux applications are vulnerable to Multipart request smuggling attacks. Affected versions: Spring Framework 7.0.0 through 7.0.7; 6.2.0 through 6.2.18; 6.1.0 through 6.1.27; 5.3.0 through 5.3.48.

Starlet versions through 0.31 for Perl allows HTTP Request Smuggling via Improper Header Precedence. Starlet incorrectly prioritizes "Content-Length" over "Transfer-Encoding: chunked" when both headers are present in an HTTP request. Per RFC 7230 3.3.3, Transfer-Encoding must take precedence. An attacker could exploit this to smuggle malicious HTTP requests via a front-end reverse proxy.

AIOHTTP is an asynchronous HTTP client/server framework for asyncio and Python. Prior to version 3.13.4, multiple Host headers were allowed in aiohttp. This issue has been patched in version 3.13.4.

The REST API TO MiniProgram plugin for WordPress is vulnerable to Insecure Direct Object Reference in all versions up to, and including, 5.1.2. This is due to the permission callback (update_user_wechatshop_info_permissions_check) only validating that the supplied 'openid' parameter corresponds to an existing WordPress user, while the callback function (update_user_wechatshop_info) uses a separate, attacker-controlled 'userid' parameter to determine which user's metadata gets modified, with no verification that the 'openid' and 'userid' belong to the same user. This makes it possible for authenticated attackers, with Subscriber-level access and above, to modify arbitrary users' store-related metadata (storeinfo, storeappid, storename) via the 'userid' REST API parameter.

SEPPmail Secure Email Gateway before version 15.0.3 allows an attacker to hide security tags from users by crafting a long subject.

SEPPmail Secure Email Gateway before version 15.0.3 allows an attacker to bypass subject sanitization and forge tags such as [signed OK].

SEPPmail Secure Email Gateway before version 15.0.3 allows an attacker to bypass subject sanitization and forge security tags using Unicode lookalike characters.

An Improper Input Validation vulnerability in routing process daemon (RPD) of Juniper Networks Junos OS devices configured with BGP origin validation using Resource Public Key Infrastructure (RPKI), allows an attacker to send a specific BGP update which may cause RPKI policy-checks to be bypassed. This, in turn, may allow a spoofed advertisement to be accepted or propagated. This issue affects: Juniper Networks Junos OS 12.3 versions prior to 12.3R12-S18; 15.1 versions prior to 15.1R7-S9; 17.2 versions prior to 17.2R3-S3; 17.3 versions prior to 17.3R3-S7; 17.4 versions prior to 17.4R2-S9, 17.4R3; 18.1 versions prior to 18.1R3-S13; 18.2 versions prior to 18.2R3-S3; 18.3 versions prior to 18.3R3-S1; 18.4 versions prior to 18.4R3; 19.1 versions prior to 19.1R2; 19.2 versions prior to 19.2R2; 19.3 versions prior to 19.3R2.

A request smuggling vulnerability exists in libsoup's HTTP/1 header parsing logic. The soup_message_headers_append_common() function in libsoup/soup-message-headers.c unconditionally appends each header value without validating for duplicate or conflicting Content-Length fields. This allows an attacker to send HTTP requests containing multiple Content-Length headers with differing values.

Improper input validation in Admin UI of EZCast Pro II version 1.17478.146 allows attackers to manipulate files in the /tmp directory

Adobe Commerce versions 2.4.9-alpha3, 2.4.8-p3, 2.4.7-p8, 2.4.6-p13, 2.4.5-p15, 2.4.4-p16 and earlier are affected by an Improper Input Validation vulnerability that could result in a security feature bypass, with limited impact to integrity. Exploitation of this issue does not require user interaction.