Confidentiality, Integrity, Authentication, and Nonrepudiation.
Carlos F Rentas
Strayer University
CIS 333
November 17, 2012
Prof. Jonathan C. Thrall

Working as an Information Security Officer, our firm was task for a client who is small software company currently using a Microsoft Server 2008 Active Directory domain and is administered by a limited number of over-tasked network administrators. The rest of the client’s staff is mostly software developers and a small number of administrative personnel. The client has decided that it would be in their best interest to use a public key infrastructure (PKI) to provide a framework that facilitates confidentiality, integrity, authentication, and nonrepudiation.

Fundamentals of PKI. Public Key Infrastructure is a security architecture created to provide a high level of confidence for exchanging information over the internet that has become more and more insecure lately. The term can be very confusing, because it is used to mean several different things; for instance PKI may mean the technologies, techniques, and methods that used together provide a secure infrastructure. Additionally PKI may mean the use of a public key and private key pair for authentication mainly as well as “proof of content”. This uses a mathematical technique known as public key cryptography which uses a pair of related cryptographic keys to verify the identity of the sender (signing), and ensuring privacy (encryption). PKI have been developed to support secure information exchange over insecure networks like the Internet, in which features like these cannot be readily provided, and can, however, be used easily for information exchanged over private networks, (including corporate internal networks). PKI can also be used to deliver cryptographic keys between servers and users in a secure manner, and also facilitate other cryptographically delivered security services. It uses a pair of mathematically related cryptographic keys. When one key is used to encrypt information, then only the related key can decrypt the same. If only one of the keys is known, then the other key becomes extremely difficult to calculate, resulting in the following: * A public key. This is something made public, freely distributed and it can be seen by all users. * A corresponding (and unique) private key. This is kept secret and not shared among users.
A private key enables a user to prove, without any doubt, who they claim to be.

Positive and negative characteristics of a Public and In-house CA. Some of the expected advantages of a PKI infrastructure comes as; * Ensuring the quality of information electronically sent and received * Ensuring the authenticity of the source and destination of the information, * Provided the source of time is known, the assurance of the time and timing of the information * Ensuring information’s privacy * Information’s authenticity that may be introduced as evidence in court In a network that’s used to issue and manage security credentials and public keys for the encryption of messages, a CA is the authority. In a public key infrastructure, a CA verifies with a RA checks with (registration authority) (RA) and checks the information provided by whoever request a digital certificate. The CA issues a certificate once the RA verifies the requestor’s information. Depending on the public key infrastructure implementation, the certificate includes the owner’s name, the owner's public key, expiration date of the certificate, and any other public key owner information such as; * Cryptography In a Public Key system, communication is asymmetric, the sender and receiver do not need a common key to send encrypted messages. The sender needs however to know the receiver's public key so the message can be sent. However for that communication to stay private, the receiver needs to keep the sender's private key confidential. * Trust Due to its asymmetric encryption, communication in a PKI infrastructure should be secure. When an individual or companies make false representations or assume false identity is when problems arise, allowing confidential data to fall into unauthorized hands. * Certificate Authority In a website holding public keys, these are trusted third parties who verify identities. These contain names, addresses, phone numbers and other information subject to be verified in order to determine if the entity is legitimate. Some of these certificates issued provide an additional encryption known as Secure Socket Layer (SSL) making communication secure.

Recommendation for either a Public CA or an In-house CA. Security has become a priority for companies over the last few years, and many administrators that have probably never thought twice about digital certificates are now finding themselves in situations that may require issuing different types of certificates. When selecting a CA (certificate authority), one of the first things that administrators will notice is cost. Depending on the type of certificate needed will determine its cost which may vary, however a single certificate could be very expensive. On the other hand, having their own certificate authority gives a company the capability of issuing as many certificates as needed for free making it a more attractive option at first, especially if a large number of certificates are to be used at any given time. Some companies have decided to run the Certificate Authority related services on an existing Windows 2008 server, and this “piggybacking” may save the client some money, as long as a large number of certificates are not being run. What’s the reason for this? Because although a Certificate Authority is a set of services under Windows Server 2008, it’s also a certificate like no other server in your entire network. To begin with, it has security requirements that are completely unique. When doing so a company is basing their organization’s data security on one or more digital certificates. If any of those certificates are compromised, their overall security is also compromised, as well as their resources. As long as saving some money do not hurt productivity, it may not be an issue in this case. If the client is considering running their own CA, then it should be run on a dedicated server. Why? Although a Certificate Authority run under Windows Server 2008 as part of their services, it’s one that requires unique security measures and requirements as mentioned above. For instance; if a hacker ever access the client’s certificate server and steal a copy of it, they could set up their own server and convincingly spoof (alter or duplicate) the client’s server’s identity. A hacker could also decipher the SSL (secure socket layers) encryption in an effort to steal customer’s credit card numbers. Additionally if certificate server ever has a hard disk failure and a current back up is not available, the client’s operations could experience some serious problems. Deploying of the clients own in house certificate authority would require extra care when it comes to protecting, securing, and back up. If their consideration is to pursue a certificate authority deployment then it will require to be treated the way that a super power would treat their nuclear warheads; protect it at all costs.

References

An Introduction to PKI (2011). Retrieved from http://www.articsoft.com/public_key_infrastructure.htm

Public Key Authority vs. Certificate Authority (2011). Retrieved from http://www.ehow.com/facts_75820_public-authority-vs-certificate-authority.html

Public & Private Key Encryption Explained (2012). Retrieved from http://www.ehow.com/about_5521788_public-private-key-encryption-explained.html