The Client shall always prove possession of a UserIdentityToken when it passes it to the Server. Some tokens include a secret such as a password which the Server will accept as proof. In order to protect these secrets, the Token may be encrypted before it is passed to the Server. Other types of tokens allow the Client to create a signature with the secret associated with the Token. In these cases, the Client proves possession of a UserIdentityToken by creating a signature with the secret and passing it to the Server.

Each UserIdentityToken allowed by an Endpoint shall have a UserTokenPolicy specified in the EndpointDescription. The UserTokenPolicy specifies what SecurityPolicy to use when encrypting or signing. If this SecurityPolicy is null or empty then the Client uses the SecurityPolicy in the EndpointDescription. If the matching SecurityPolicy is set to None then no encryption or signature is required. The possible SecurityPolicies are defined in OPC 10000-7.

It is recommended that applications never set the SecurityPolicy to None for UserIdentityTokens that include a secret because these secrets could be used by an attacker to gain access to the system.

Clients shall validate the Server Certificate and ensure it is trusted before sending a UserIdentityToken encrypted with the Certificate.

The encrypted secret and Signature are embedded in a ByteString which is part of the UserIdentityToken. The format of this ByteString depends on the type of UserIdentityToken and the SecurityPolicy. Clients shall validate the Server Certificate and ensure it is trusted before sending a UserIdentityToken encrypted with the Certificate.

The legacy token secret format defined in 7.41.2.2 is not extensible and provides only encryption but the encrypted data is not signed. It is used together with the USERNAME UserIdentityToken. The password secret exchanged with this format shall not exceed 64 bytes. If the password exceeds 64 bytes, the EncryptedSecret format shall be used or the clear text password is sent over a SecureChannel that is encrypted.

The EncryptedSecret format defined in 7.41.2.3 provides an extensible secret format together with the definition how the secret is signed and encrypted. It allows for the layout to be updated as new token types are defined or new SecurityPolicies are added.

The EncryptedSecret format starts with a TypeId, EncodingMask and Length. These values allow a Server to determine how to handle the secret. If these fields are not valid and a USERNAME UserIdentityToken has been provided then the Server may attempt to handle the secret using the legacy token secret format. If these fields are valid but the TypeId is not recognized or not valid for the SecurityPolicy then the Server rejects the UserIdentityToken.

The UserIdentityToken types and the token formats supported by the Endpoint are identified by the UserTokenPolicy defined in 7.42.

To prevent the leakage of information useful to attackers, Servers shall ensure that the process of validating UserIdentityTokens completes in a fixed interval independent of whether an error occurs or not. The process of validation includes decrypting, check for padding and checking for a valid nonce. If any errors occur the return code is Bad_IdentityTokenInvalid.

Servers shall log details of any failure to validate a UserIdentityToken and shall lock out Client applications after five failures.