Some Choices Made in This Book

In the past few years, many forks of MySQL were born. The most important are Percona server and MariaDB. I skipped them completely in this book because I work mostly with MySQL and simply cannot describe servers I don’t work with daily. However, because they are forks, you can use most of the methods described here. Only if you are dealing with a particular feature added in the fork will you need information specific to that product.

To conserve space and avoid introducing a whole new domain of knowledge with a lot of its own concerns, I left out MySQL Cluster problems. If you use MySQL Cluster and run into an SQL or application issue, you can troubleshoot it in much the same way as any other storage engine issue. Therefore, this book is applicable to such issues on clusters. But issues that are specific to MySQL Cluster need separate MySQL Cluster knowledge that I don’t describe here.

But I do devote a lot of space to MyISAM- and InnoDB-specific problems. This was done because they are by far the most popular storage engines, and their installation base is huge. Both also were and are default storage engines: MyISAM before version 5.5 and InnoDB since version 5.5.

A few words about examples. They were all created either specially for this book or for conferences where I have spoken about troubleshooting. Although some of the examples are based on real support cases and bug reports, all the code is new and cannot be associated with any confidential data. In a few places, I describe customer “tickets.” These are not real either. At the same time, the problems described here are real and have been seen many times, just with different code, names, and circumstances.

I tried to keep all examples as simple, understandable, and universal as possible. Therefore, I use the MySQL command-line client in most of the examples. You always have this client in the MySQL installation.

This decision also explains why I don’t describe all complications specific to particular kinds of installations; it is just impossible to cover them all in single book. Instead, I tried to give starting points that you can extend.

I have decided to use the C API to illustrate the functions discussed in this book. The choice wasn’t easy, because there are a lot of programming APIs for MySQL in various languages. I couldn’t possibly cover them all, and didn’t want to guess which ones would be popular. I realized that many of them look like the C API (many are even wrappers around the C API), so I decided that would be the best choice. Even if you are using an API with a very different syntax, such as ODBC, this section still can be useful because you will know what to look for.

A few examples use PHP. I did so because I use PHP and therefore could show real examples based on my own code. Real examples are always good to show because they reflect real-life problems that readers are likely to encounter. In addition, the MySQL API in PHP is based on the C API and uses the very same names, so readers should be able to compare it easily to C functions discussed in this book.[1]

I omitted JDBC and ODBC examples because these APIs are very specific. At the same time, their debugging techniques are very similar, if not always the same. Mostly the syntax is different. I decided that adding details about these two connectors might confuse readers without offering any new information about troubleshooting.[2]



[1] mysqlnd uses its own client protocol implementation, but still names functions in the same style as the C API.

Get MySQL Troubleshooting now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.