CVE Database

In Eclipse ThreadX NetX Duo versions 6.0 through 6.5.2.202603a, the PPP IPCP code parses a peer's Configure-Nak without validating each option's length. A malicious PPP peer can stall the PPP thread indefinitely with a zero-length option, taking the link down, and can read back up to four bytes of packet-buffer memory per exchange with a short address option.

In Eclipse ThreadX NetX Duo versions 6.4.0 through 6.5.2.202603a, built with VLAN and SRP support, the MSRP parser uses the attribute list length from the received frame without comparing it with the packet length. An unauthenticated station on the same Layer 2 segment can send a short MSRP frame declaring a long list and make the MRP worker read far past the received packet, faulting it on targets with memory protection.

In Eclipse ThreadX NetX Duo versions 6.4.0 through 6.5.2.202603a, built with VLAN and SRP support, evicting the last attribute in an MRP participant's list writes through a null successor pointer. An unauthenticated station on the same Layer 2 segment can fault the MRP worker thread by declaring one MSRP attribute and withdrawing it, and ordinary withdrawals by legitimate stations reach the same path.

In Eclipse ThreadX NetX Duo versions 6.4.0 through 6.5.2.202603a, built with VLAN and SRP support, the MSRP attribute lookup writes through the null pointer returned by its allocator when the ten-entry attribute table is full. An unauthenticated station on the same Layer 2 segment can fault the MRP worker thread by declaring enough new attributes, and Talker Failed declarations hold table entries that are never released.

In Eclipse ThreadX NetX Duo versions 6.0 through 6.5.2.202603a, the FTP client in IPv4 passive mode opens its data connection to whatever address the server names in its 227 reply instead of the server it is talking to. A malicious FTP server can make the client connect to any host and port the device can reach, using it to scan the internal network and to deliver uploads to internal services.

In Eclipse ThreadX NetX Duo versions 6.0 through 6.5.2.202603a, the TFTP client copies the message of a received ERROR packet without bounding the copy by the end of the datagram, and the TFTP server reads one byte past a minimal ERROR packet. A malicious TFTP server can make the client copy up to 64 bytes of adjacent packet-pool memory into the error string the application retrieves, or fault on targets with memory protection.

In Eclipse ThreadX NetX Duo versions 6.0 through 6.5.2.202603a, the DHCP client's option walk advances its offset one byte less than its read pointer for every option it skips. A DHCP server, or any host on the link that answers first, can make the client read up to about a kilobyte past the received message, and the same walk copies out-of-bounds bytes into the application's buffer when it retrieves a user option.

In Eclipse ThreadX NetX Duo versions 6.0 through 6.5.2.202603a, the NetX Secure DTLS server fills HelloVerifyRequest cookies from NX_RAND(), by default the C library rand(), with no key and no binding to the client. An attacker who can predict the generator's output can pass the cookie exchange from a spoofed address and make the server send its handshake flight to a victim, using it as a traffic reflector.

In Eclipse ThreadX NetX Duo versions up to and including 6.5.2.202603, the SNMP agent's OID parser uses a multi-byte BER length from the request without checking it against the bytes present, and reads past the end of the packet buffer. A remote, unauthenticated attacker can send a crafted SNMP request that makes the agent read adjacent memory into its OID state, faulting the agent on targets with memory protection.

In Eclipse ThreadX NetX Duo versions up to and including 6.5.2.202603, the RTSP server returns a raw parser error for a malformed Session header and never releases the response packet it allocated. An unauthenticated client can drain the server's packet pool with a few dozen requests, stopping the RTSP server and, where the pool is shared, the rest of the network stack until the device restarts.

In Eclipse ThreadX NetX Duo versions up to and including 6.5.2.202603, the mDNS add-on's string cache accepts a shorter name as a match for a longer stored one. A host on the local link can send two mDNS records that make a device performing continuous service discovery write up to three bytes past its transmit packet, corrupting a neighbouring packet or the pool's free list, and misattribute discovered records.

In Eclipse ThreadX NetX Duo versions up to and including 6.5.2.202603, built with IPv6, the ICMPv6 option validator accepts an option area that ends in a one- or two-byte residue, and the Router Advertisement, Neighbor Solicitation, Neighbor Advertisement and Redirect handlers then walk it without length checks. A host on the local link can send one such message to make the IP thread read past the packet or loop forever, and can place off-packet memory in the neighbor cache.

In Eclipse ThreadX NetX Duo versions up to and including 6.5.2.202603, built with packet chaining, the TFTP server does not bound the length of a received DATA datagram. An unauthenticated peer can upload an oversized datagram that writes memory beyond the received packet into the file and read it back by downloading the file, and the same datagram can suspend the server thread permanently.

In Eclipse ThreadX NetX Duo versions up to and including 6.5.2.202603, built with DTLS server support, the NetX Secure DTLS server's ClientHello parser reads the session ID, cookie and ciphersuite list without bounding them by the bytes remaining in the message. An unauthenticated peer can send a ClientHello declaring fields longer than the message and make the server read past it; nothing read is sent back to the peer, and the read can fault the server only on targets that trap the address beyond the buffer.

In Eclipse ThreadX versions 6.0.1 through 6.5.2.202603, the Module Manager loaders take a module image without its length and trust the sizes and offsets in its preamble. A crafted module image can declare a code size larger than the image to read, or gain access to, the memory that follows it, and can place its entry points outside its code.

In Eclipse ThreadX versions 6.0.1 through 6.5.2.202603, built with TX_ENABLE_EVENT_TRACE, the Module Manager lets a user-mode, memory-protected module register any function pointer as the kernel's trace-buffer-full callback, which the kernel calls from privileged code when the trace buffer wraps. A malicious module can thereby run its own code with kernel privilege, defeating module isolation, or fault the kernel with an invalid pointer.

In Eclipse ThreadX versions 6.0.1 through 6.5.2.202603, the four non-secure-callable secure stack functions of the TrustZone-M ports dereference a thread pointer supplied by the non-secure world without checking that it points to non-secure memory. Non-secure code can make the secure firmware reveal which secure bytes are zero, write chosen secure words, and fault the secure world, breaking TrustZone-M isolation and enabling recovery of secure-world secrets such as cryptographic keys.

Electron is a framework for writing cross-platform desktop applications using JavaScript, HTML and CSS. From 42.3.3 until 42.10.0, 43.5.0, and 44.0.0-beta.6, Electron's sandboxed preload code cache did not verify that a cached entry matched the preload it was served for. A compromised renderer could write attacker-controlled cache data and cause Electron to reuse it for a later load, executing the renderer's code in the more privileged preload context. The issue affects applications that load untrusted content. This issue is fixed in versions 42.10.0, 43.5.0, and 44.0.0-beta.6.

MobilityDB version 1.3.0 and earlier contains an out-of-bounds read vulnerability in the MEOS binary and library WKB deserialization logic that allows unprivileged database users to crash the PostgreSQL backend process by supplying a crafted WKB payload with a negative length field. The negative length value wraps to a large unsigned size_t due to missing signed validation, bypasses an overflow-unsafe pointer arithmetic bounds check in wkb_parse_state_check(), and causes memcpy() in text_from_wkb_state() to operate with a corrupted unbounded length, resulting in a remote denial-of-service condition affecting all sessions on the PostgreSQL instance.

A flaw was found in libsoup. When the permessage-deflate WebSocket extension compresses a very large outgoing message, truncated size calculations used for GByteArray growth could wrap, causing zlib to write past the allocated buffer and resulting in a heap buffer overflow.

A flaw was found in libsoup. When constructing a masked WebSocket client frame for a very large outgoing payload, size values passed to GByteArray allocation APIs could be truncated while the masking routine still used the full length, causing a heap buffer overflow.

A flaw was found in libsoup. When max-incoming-payload-size is unlimited (0), SoupWebsocketConnection could grow its incoming GByteArray based on an attacker-controlled frame length until the length wrapped, causing a heap buffer overflow while reading frame data.

A flaw was found in libsoup. The soup_uri_decode_data_uri() function incorrectly treated base64 data-URI payloads as NUL-terminated strings when calling g_base64_decode_inplace(). If the percent-decoded payload contained embedded NUL bytes, the decoded length could remain uninitialized and be used as the size of the returned GBytes. This can lead to an out-of-bounds read or application crash when processing a crafted data URI.

Improper link resolution (CWE-59 / CWE-22) in the allowedLocalRoots path validation in Google MCP Toolbox for Databases versions 1.2.0 through 1.9.0 allows a remote authenticated attacker with tool execution permissions to bypass directory boundary restrictions via symbolic links. Because path validation checks directories lexically without resolving symbolic links first, an attacker can access or overwrite arbitrary local files located outside the permitted root directories.

Joomla! Core - [20260916] - Core - XSS filter bypass in InputFilter via whitespace characters in HTML data URIs in Joomla 1.5.0-5.4.8, 6.0.0-6.1.3 - The cleanAttribute method removes HTML data URIs, however injected whitespaces characters could circumvent that cleanup, causing an XSS vector.