wolfssl CVE Vulnerabilities & Metrics

Fault Injection vulnerability in wc_ed25519_sign_msg function in wolfssl/wolfcrypt/src/ed25519.c in WolfSSL wolfssl5.6.6 on Linux/Windows allows remote attacker co-resides in the same system with a victim process to disclose information and escalate privileges via Rowhammer fault injection to the ed25519_key structure.

Fault Injection vulnerability in RsaPrivateDecryption function in wolfssl/wolfcrypt/src/rsa.c in WolfSSL wolfssl5.6.6 on Linux/Windows allows remote attacker co-resides in the same system with a victim process to disclose information and escalate privileges via Rowhammer fault injection to the RsaKey structure.

The side-channel protected T-Table implementation in wolfSSL up to version 5.6.5 protects against a side-channel attacker with cache-line resolution. In a controlled environment such as Intel SGX, an attacker can gain a per instruction sub-cache-line resolution allowing them to break the cache-line-level protection. For details on the attack refer to: https://doi.org/10.46586/tches.v2024.i1.457-500

In function MatchDomainName(), input param str is treated as a NULL terminated string despite being user provided and unchecked. Specifically, the function X509_check_host() takes in a pointer and length to check against, with no requirements that it be NULL terminated. If a caller was attempting to do a name check on a non-NULL terminated buffer, the code would read beyond the bounds of the input array until it found a NULL terminator.This issue affects wolfSSL: through 5.7.0.

In wolfSSL prior to 5.6.6, if callback functions are enabled (via the WOLFSSL_CALLBACKS flag), then a malicious TLS client or network attacker can trigger a buffer over-read on the heap of 5 bytes (WOLFSSL_CALLBACKS is only intended for debugging).

wolfSSL prior to 5.6.6 did not check that messages in one (D)TLS record do not span key boundaries. As a result, it was possible to combine (D)TLS messages using different keys into one (D)TLS record. The most extreme edge case is that, in (D)TLS 1.3, it was possible that an unencrypted (D)TLS 1.3 record from the server containing first a ServerHello message and then the rest of the first server flight would be accepted by a wolfSSL client. In (D)TLS 1.3 the handshake is encrypted after the ServerHello but a wolfSSL client would accept an unencrypted flight from the server. This does not compromise key negotiation and authentication so it is assigned a low severity rating.

wolfSSL SP Math All RSA implementation is vulnerable to the Marvin Attack, new variation of a timing Bleichenbacher style attack, when built with the following options to configure: --enable-all CFLAGS="-DWOLFSSL_STATIC_RSA" The define “WOLFSSL_STATIC_RSA” enables static RSA cipher suites, which is not recommended, and has been disabled by default since wolfSSL 3.6.6.  Therefore the default build since 3.6.6, even with "--enable-all", is not vulnerable to the Marvin Attack. The vulnerability is specific to static RSA cipher suites, and expected to be padding-independent. The vulnerability allows an attacker to decrypt ciphertexts and forge signatures after probing with a large number of test observations. However the server’s private key is not exposed.

If a TLS 1.3 client gets neither a PSK (pre shared key) extension nor a KSE (key share extension) when connecting to a malicious server, a default predictable buffer gets used for the IKM (Input Keying Material) value when generating the session master secret. Using a potentially known IKM value when generating the session master secret key compromises the key generated, allowing an eavesdropper to reconstruct it and potentially allowing access to or meddling with message contents in the session. This issue does not affect client validation of connected servers, nor expose private key information, but could result in an insecure TLS 1.3 session when not controlling both sides of the connection. wolfSSL recommends that TLS 1.3 client side users update the version of wolfSSL used. 

In wolfSSL before 5.5.2, if callback functions are enabled (via the WOLFSSL_CALLBACKS flag), then a malicious TLS 1.3 client or network attacker can trigger a buffer over-read on the heap of 5 bytes. (WOLFSSL_CALLBACKS is only intended for debugging.)

An issue was discovered in wolfSSL before 5.5.0. A fault injection attack on RAM via Rowhammer leads to ECDSA key disclosure. Users performing signing operations with private ECC keys, such as in server-side TLS connections, might leak faulty ECC signatures. These signatures can be processed via an advanced technique for ECDSA key recovery. (In 5.5.0 and later, WOLFSSL_CHECK_SIG_FAULTS can be used to address the vulnerability.)

In wolfSSL before 5.5.1, malicious clients can cause a buffer overflow during a TLS 1.3 handshake. This occurs when an attacker supposedly resumes a previous TLS session. During the resumption Client Hello a Hello Retry Request must be triggered. Both Client Hellos are required to contain a list of duplicate cipher suites to trigger the buffer overflow. In total, two Client Hellos have to be sent: one in the resumed session, and a second one as a response to a Hello Retry Request message.

wolfSSL through 5.0.0 allows an attacker to cause a denial of service and infinite loop in the client component by sending crafted traffic from a Machine-in-the-Middle (MITM) position. The root cause is that the client module accepts TLS messages that normally are only sent to TLS servers.

An issue was discovered in wolfSSL before 5.5.0 (when --enable-session-ticket is used); however, only version 5.3.0 is exploitable. Man-in-the-middle attackers or a malicious server can crash TLS 1.2 clients during a handshake. If an attacker injects a large ticket (more than 256 bytes) into a NewSessionTicket message in a TLS 1.2 handshake, and the client has a non-empty session cache, the session cache frees a pointer that points to unallocated memory, causing the client to crash with a "free(): invalid pointer" message. NOTE: It is likely that this is also exploitable during TLS 1.3 handshakes between a client and a malicious server. With TLS 1.3, it is not possible to exploit this as a man-in-the-middle.

An issue was discovered in wolfSSL before 5.5.0. When a TLS 1.3 client connects to a wolfSSL server and SSL_clear is called on its session, the server crashes with a segmentation fault. This occurs in the second session, which is created through TLS session resumption and reuses the initial struct WOLFSSL. If the server reuses the previous session structure (struct WOLFSSL) by calling wolfSSL_clear(WOLFSSL* ssl) on it, the next received Client Hello (that resumes the previous session) crashes the server. Note that this bug is only triggered when resuming sessions using TLS session resumption. Only servers that use wolfSSL_clear instead of the recommended SSL_free; SSL_new sequence are affected. Furthermore, wolfSSL_clear is part of wolfSSL's compatibility layer and is not enabled by default. It is not part of wolfSSL's native API.

wolfSSL before 5.4.0 allows remote attackers to cause a denial of service via DTLS because a check for return-routability can be skipped.

In wolfSSL before 5.2.0, a TLS 1.3 server cannot properly enforce a requirement for mutual authentication. A client can simply omit the certificate_verify message from the handshake, and never present a certificate.

In wolfSSL before 5.2.0, certificate validation may be bypassed during attempted authentication by a TLS 1.3 client to a TLS 1.3 server. This occurs when the sig_algo field differs between the certificate_verify message and the certificate message.

wolfSSL 5.x before 5.1.1 uses non-random IV values in certain situations. This affects connections (without AEAD) using AES-CBC or DES3 with TLS 1.1 or 1.2 or DTLS 1.1 or 1.2. This occurs because of misplaced memory initialization in BuildMessage in internal.c.

wolfSSL wolfMQTT 1.9 has a heap-based buffer overflow in MqttClient_DecodePacket (called from MqttClient_WaitType and MqttClient_Subscribe).

wolfSSL wolfMQTT 1.9 has a heap-based buffer overflow in MqttClient_DecodePacket (called from MqttClient_WaitType and MqttClient_Unsubscribe).

wolfSSL wolfMQTT 1.9 has a heap-based buffer overflow in MqttClient_DecodePacket (called from MqttClient_WaitType and MqttClient_Connect).

wolfSSL wolfMQTT 1.9 has a heap-based buffer overflow in MqttDecode_Disconnect (called from MqttClient_DecodePacket and MqttClient_WaitType).

wolfSSL wolfMQTT 1.9 has a heap-based buffer overflow in MqttClient_DecodePacket (called from MqttClient_HandlePacket and MqttClient_WaitType).

wolfSSL wolfMQTT 1.9 has a heap-based buffer overflow (8 bytes) in MqttDecode_Publish (called from MqttClient_DecodePacket and MqttClient_HandlePacket).

wolfSSL wolfMQTT 1.9 has a heap-based buffer overflow (4 bytes) in MqttDecode_Publish (called from MqttClient_DecodePacket and MqttClient_HandlePacket).

wolfSSL before 4.8.1 incorrectly skips OCSP verification in certain situations of irrelevant response data that contains the NoCheck extension.

wolfSSL 4.6.x through 4.7.x before 4.8.0 does not produce a failure outcome when the serial number in an OCSP request differs from the serial number in the OCSP response.

In wolfSSL through 4.6.0, a side-channel vulnerability in base64 PEM file decoding allows system-level (administrator) attackers to obtain information about secret RSA keys via a controlled-channel and side-channel attack on software running in isolated environments that can be single stepped, especially Intel SGX.

DoTls13CertificateVerify in tls13.c in wolfSSL before 4.7.0 does not cease processing for certain anomalous peer behavior (sending an ED22519, ED448, ECC, or RSA signature without the corresponding certificate). The client side is affected because man-in-the-middle attackers can impersonate TLS 1.3 servers.

RsaPad_PSS in wolfcrypt/src/rsa.c in wolfSSL before 4.6.0 has an out-of-bounds write for certain relationships between key size and digest size.

wolfSSL before 4.5.0 mishandles TLS 1.3 server data in the WAIT_CERT_CR state, within SanityCheckTls13MsgReceived() in tls13.c. This is an incorrect implementation of the TLS 1.3 client state machine. This allows attackers in a privileged network position to completely impersonate any TLS 1.3 servers, and read or modify potentially sensitive information between clients using the wolfSSL library and these TLS servers.

An issue was discovered in the DTLS handshake implementation in wolfSSL before 4.5.0. Clear DTLS application_data messages in epoch 0 do not produce an out-of-order error. Instead, these messages are returned to the application.

An issue was discovered in wolfSSL before 4.5.0, when single precision is not employed. Local attackers can conduct a cache-timing attack against public key operations. These attackers may already have obtained sensitive information if the affected system has been used for private key operations (e.g., signing with a private key).

An issue was discovered in wolfSSL before 4.5.0. It mishandles the change_cipher_spec (CCS) message processing logic for TLS 1.3. If an attacker sends ChangeCipherSpec messages in a crafted way involving more than one in a row, the server becomes stuck in the ProcessReply() loop, i.e., a denial of service.

The private-key operations in ecc.c in wolfSSL before 4.4.0 do not use a constant-time modular inverse when mapping to affine coordinates, aka a "projective coordinates leak."

wolfSSL 4.3.0 has mulmod code in wc_ecc_mulmod_ex in ecc.c that does not properly resist timing side-channel attacks.

wolfSSL CyaSSL before 2.9.4 allows remote attackers to have unspecified impact via multiple calls to the CyaSSL_read function which triggers an out-of-bounds read when an error occurs, related to not checking the return code and MAC verification failure.

The SSL 3 HMAC functionality in wolfSSL CyaSSL 2.5.0 before 2.9.4 does not check the padding length when verification fails, which allows remote attackers to have unspecified impact via a crafted HMAC, which triggers an out-of-bounds read.

The DoAlert function in the (1) TLS and (2) DTLS implementations in wolfSSL CyaSSL before 2.9.4 allows remote attackers to have unspecified impact and vectors, which trigger memory corruption or an out-of-bounds read.

An issue was discovered in wolfSSL before 4.3.0 in a non-default configuration where DSA is enabled. DSA signing uses the BEEA algorithm during modular inversion of the nonce, leading to a side-channel attack against the nonce.

wolfSSL before 4.3.0 mishandles calls to wc_SignatureGenerateHash, leading to fault injection in RSA cryptography.

In wolfSSL before 4.3.0, wc_ecc_mulmod_ex does not properly resist side-channel attacks.

wolfSSL and wolfCrypt 4.1.0 and earlier (formerly known as CyaSSL) generate biased DSA nonces. This allows a remote attacker to compute the long term private key from several hundred DSA signatures via a lattice attack. The issue occurs because dsa.c fixes two bits of the generated nonces.

wolfssl before 3.2.0 has a server certificate that is not properly authorized for server authentication.

wolfssl before 3.2.0 does not properly authorize CA certificate for signing other certificates.

wolfssl before 3.2.0 does not properly issue certificates for a server's hostname.

In wolfSSL 4.1.0 through 4.2.0c, there are missing sanity checks of memory accesses in parsing ASN.1 certificate data while handshaking. Specifically, there is a one-byte heap-based buffer overflow inside the DecodedCert structure in GetName in wolfcrypt/src/asn.c because the domain name location index is mishandled. Because a pointer is overwritten, there is an invalid free.

wolfSSL and wolfCrypt 4.0.0 and earlier (when configured without --enable-fpecc, --enable-sp, or --enable-sp-math) contain a timing side channel in ECDSA signature generation. This allows a local attacker, able to precisely measure the duration of signature operations, to infer information about the nonces used and potentially mount a lattice attack to recover the private key used. The issue occurs because ecc.c scalar multiplication might leak the bit length.

In wolfSSL through 4.1.0, there is a missing sanity check of memory accesses in parsing ASN.1 certificate data while handshaking. Specifically, there is a one-byte heap-based buffer over-read in CheckCertSignature_ex in wolfcrypt/src/asn.c.

wolfSSL 4.1.0 has a one-byte heap-based buffer over-read in DecodeCertExtensions in wolfcrypt/src/asn.c because reading the ASN_BOOLEAN byte is mishandled for a crafted DER certificate in GetLength_ex.

wolfSSL 4.0.0 has a Buffer Overflow in DoPreSharedKeys in tls13.c when a current identity size is greater than a client identity size. An attacker sends a crafted hello client packet over the network to a TLSv1.3 wolfSSL server. The length fields of the packet: record length, client hello length, total extensions length, PSK extension length, total identity length, and identity length contain their maximum value which is 2^16. The identity data field of the PSK extension of the packet contains the attack data, to be stored in the undefined memory (RAM) of the server. The size of the data is about 65 kB. Possibly the attacker can perform a remote code execution attack.

examples/benchmark/tls_bench.c in a benchmark tool in wolfSSL through 3.15.7 has a heap-based buffer overflow.

It was found that wolfssl before 3.15.7 is vulnerable to a new variant of the Bleichenbacher attack to perform downgrade attacks against TLS. This may lead to leakage of sensible data.

wolfcrypt/src/ecc.c in wolfSSL before 3.15.1.patch allows a memory-cache side-channel attack on ECDSA signatures, aka the Return Of the Hidden Number Problem or ROHNP. To discover an ECDSA key, the attacker needs access to either the local machine or a different virtual machine on the same physical host.

wolfSSL prior to version 3.12.2 provides a weak Bleichenbacher oracle when any TLS cipher suite using RSA key exchange is negotiated. An attacker can recover the private key from a vulnerable wolfSSL application. This vulnerability is referred to as "ROBOT."

CyaSSL does not check the key usage extension in leaf certificates, which allows remote attackers to spoof servers via a crafted server certificate not authorized for use in an SSL/TLS handshake.

A specially crafted x509 certificate can cause a single out of bounds byte overwrite in wolfSSL through 3.10.2 resulting in potential certificate validation vulnerabilities, denial of service and possible remote code execution. In order to trigger this vulnerability, the attacker needs to supply a malicious x509 certificate to either a server or a client application using this library.

wolfSSL before 3.11.0 does not prevent wc_DhAgree from accepting a malformed DH key.

wolfSSL before 3.10.2 has an out-of-bounds memory access with loading crafted DH parameters, aka a buffer overflow triggered by a malformed temporary DH file.

In versions of wolfSSL before 3.10.2 the function fp_mul_comba makes it easier to extract RSA key information for a malicious user who has access to view cache on a machine.

The C software implementation of AES Encryption and Decryption in wolfSSL (formerly CyaSSL) before 3.9.10 makes it easier for local users to discover AES keys by leveraging cache-bank timing differences.

The C software implementation of RSA in wolfSSL (formerly CyaSSL) before 3.9.10 makes it easier for local users to discover RSA keys by leveraging cache-bank hit differences.

The C software implementation of ECC in wolfSSL (formerly CyaSSL) before 3.9.10 makes it easier for local users to discover RSA keys by leveraging cache-bank hit differences.

wolfSSL (formerly CyaSSL) before 3.6.8 does not properly handle faults associated with the Chinese Remainder Theorem (CRT) process when allowing ephemeral key exchange without low memory optimizations on a server, which makes it easier for remote attackers to obtain private RSA keys by capturing TLS handshakes, aka a Lenstra attack.

wolfSSL (formerly CyaSSL) before 3.6.8 allows remote attackers to cause a denial of service (resource consumption or traffic amplification) via a crafted DTLS cookie in a ClientHello message.

Multiple stack-based buffer overflows in the CertDecoder::GetName function in src/asn.cpp in TaoCrypt in yaSSL before 1.9.9, as used in mysqld in MySQL 5.0.x before 5.0.90, MySQL 5.1.x before 5.1.43, MySQL 5.5.x through 5.5.0-m2, and other products, allow remote attackers to execute arbitrary code or cause a denial of service (memory corruption and daemon crash) by establishing an SSL connection and sending an X.509 client certificate with a crafted name field, as demonstrated by mysql_overflow1.py and the vd_mysql5 module in VulnDisco Pack Professional 8.11. NOTE: this was originally reported for MySQL 5.0.51a.