Book Image

Demystifying Cryptography with OpenSSL 3.0

By : Alexei Khlebnikov
Book Image

Demystifying Cryptography with OpenSSL 3.0

By: Alexei Khlebnikov

Overview of this book

Security and networking are essential features of software today. The modern internet is full of worms, Trojan horses, men-in-the-middle, and other threats. This is why maintaining security is more important than ever. OpenSSL is one of the most widely used and essential open source projects on the internet for this purpose. If you are a software developer, system administrator, network security engineer, or DevOps specialist, you’ve probably stumbled upon this toolset in the past – but how do you make the most out of it? With the help of this book, you will learn the most important features of OpenSSL, and gain insight into its full potential. This book contains step-by-step explanations of essential cryptography and network security concepts, as well as practical examples illustrating the usage of those concepts. You’ll start by learning the basics, such as how to perform symmetric encryption and calculate message digests. Next, you will discover more about cryptography: MAC and HMAC, public and private keys, and digital signatures. As you progress, you will explore best practices for using X.509 certificates, public key infrastructure, and TLS connections. By the end of this book, you’ll be able to use the most popular features of OpenSSL, allowing you to implement cryptography and TLS in your applications and network infrastructure.
Table of Contents (20 chapters)
Part 1: Introduction
Part 2: Symmetric Cryptography
Part 3: Asymmetric Cryptography and Certificates
Part 4: TLS Connections and Secure Communication
Part 5: Running a Mini-CA

Comparing OpenSSL with BoringSSL

BoringSSL is another fork of OpenSSL that was made public in 2014. BoringSSL was made for the needs of the Google corporation. For years, Google maintained its own patches for OpenSSL for use in various Google products, such as Chrome, Android, and the server’s infrastructure. Finally, they decided to fork OpenSSL and maintain their fork as a separate library.

Like LibreSSL, in BoringSSL, Google removed a lot of the original OpenSSL code, which was responsible for supporting old and unpopular algorithms and features. Google also added some functionality that does not exist in OpenSSL. For example, the CRYPTO_BUFFER functionality allows you to deduplicate X.509 certificates in memory, thus reducing memory usage. It also allows you to remove OpenSSL’s X.509 and ASN.1 code from the application if OpenSSL is linked statically to the application. The X.509 code is a sizeable part of OpenSSL.

Unlike LibreSSL, BoringSSL does not aim for API compatibility with OpenSSL, or even with former versions of BoringSSL. Google wants to change the library API at will. It makes sense because Google controls both BoringSSL and the major software projects that use the library, which makes it possible to synchronize the API changes in BoringSSL and those projects. If the API is not kept stable, it is possible to free up development resources that would otherwise be spent on maintaining the old APIs.

But this also means that if someone outside Google wants to use BoringSSL, they should be ready for breaking changes in the library API, at the least suitable times. This is very inconvenient for developers who use the library. Google understands this and states that although BoringSSL is an open source project, it is not for general use.

My opinion is that BoringSSL was made open source mostly for third-party contributors to Google projects, such as Chrome and Android.

I do not recommend using BoringSSL in your applications due to its API instability. Furthermore, OpenSSL has more features, better documentation, and a much larger community.

With that, we have reviewed several competitors of OpenSSL. You should now understand the main differences between the popular TLS libraries and which library should be used in which case.

Let’s proceed to the summary.