An issue was discovered in the yh_create_session() function of yubihsm-shell through 2.0.2. The function does not explicitly check the returned session id from the device. An invalid session id would lead to out-of-bounds read and write operations in the session array. This could be used by an attacker to cause a denial of service attack.

An issue was discovered in the _send_secure_msg() function of yubihsm-shell through 2.0.2. The function does not validate the embedded length field of a message received from the device. This could lead to an oversized memcpy() call that will crash the running process. This could be used by an attacker to cause a denial of service.

An information leak was discovered on Yubico YubiKey 5 NFC devices 5.0.0 to 5.2.6 and 5.3.0 to 5.3.1. The OTP application allows a user to set optional access codes on OTP slots. This access code is intended to prevent unauthorized changes to OTP configurations. The access code is not checked when updating NFC specific components of the OTP configurations. This may allow an attacker to access configured OTPs and passwords stored in slots that were not configured by the user to be read over NFC, despite a user having set an access code. (Users who have not set an access code, or who have not configured the OTP slots, are not impacted by this issue.)

An issue was discovered in Yubico libykpiv before 2.1.0. lib/util.c in this library (which is included in yubico-piv-tool) does not properly check embedded length fields during device communication. A malicious PIV token can misreport the returned length fields during RSA key generation. This will cause stack memory to be copied into heap allocated memory that gets returned to the caller. The leaked memory could include PINs, passwords, key material, and other sensitive information depending on the integration. During further processing by the caller, this information could leak across trust boundaries. Note that RSA key generation is triggered by the host and cannot directly be triggered by the token.

A PIN management problem was discovered on Yubico YubiKey 5 devices 5.2.0 to 5.2.6. OpenPGP has three passwords: Admin PIN, Reset Code, and User PIN. The Reset Code is used to reset the User PIN, but it is disabled by default. A flaw in the implementation of OpenPGP sets the Reset Code to a known value upon initialization. If the retry counter for the Reset Code is set to non-zero without changing the Reset Code, this known value can be used to reset the User PIN. To set the retry counters, the Admin PIN is required.

An issue was discovered in Yubico libykpiv before 2.1.0. An attacker can trigger an incorrect free() in the ykpiv_util_generate_key() function in lib/util.c through incorrect error handling code. This could be used to cause a denial of service attack.

The sync endpoint in YubiKey Validation Server before 2.40 allows remote attackers to replay an OTP. NOTE: this issue is potentially relevant to persons outside Yubico who operate a self-hosted OTP validation service with a non-default configuration such as an open sync pool; the issue does NOT affect YubiCloud.

The verify endpoint in YubiKey Validation Server before 2.40 does not check the length of SQL queries, which allows remote attackers to cause a denial of service, aka SQL injection. NOTE: this issue is potentially relevant to persons outside Yubico who operate a self-hosted OTP validation service; the issue does NOT affect YubiCloud.

Yubico PAM Module before 2.10 performed user authentication when 'use_first_pass' PAM configuration option was not used and the module was configured as 'sufficient' in the PAM configuration. A remote attacker could use this flaw to circumvent common authentication process and obtain access to the account in question by providing a NULL value (pressing Ctrl-D keyboard sequence) as the password string.

In Yubico pam-u2f 1.0.7, when configured with debug and a custom debug log file is set using debug_file, that file descriptor is not closed when a new process is spawned. This leads to the file descriptor being inherited into the child process; the child process can then read from and write to it. This can leak sensitive information and also, if written to, be used to fill the disk or plant misinformation.