The latest from the MySQL community

In this post, We will learn how to handle online schema change if the table has triggers.

In PXC, an alter can be made directly ( TOI ) on tables with less than a 1G ( by default) , but on a 20GB or 200GB table we need some downtime to do ( RSU ).

Pt-osc is a good choice for Percona Cluster/Galera. By default percona toolkit’s pt-online-schema-change will create After “insert / update / delete” triggers for maintaining the sync between the shadow and the original table.

pt-online-schema-change process flow:

Check out the complete slides for effective MySQL administration here

If the tables has triggers already then pt-osc wont work well in …

In this blog post, we’ll look at what is needed when migrating MySQL users to Amazon RDS. We’ll discuss how we can transform MySQL user grants and make them compatible with Amazon RDS.

In order to deliver a managed service experience, Amazon RDS does not provide shell access to the underlying operating system. It also restricts access to certain procedures that require advanced privileges.

Every MySQL instance has some users with ALL PRIVILEGES, and you can’t directly migrate these users to Amazon RDS because it does not support following privileges for regular users.

• SUPER – Enable use of other administrative operations such as CHANGE MASTER TO, KILL, PURGE BINARY …

As you know, foreign keys establish a sort of relationship between 2 tables. MySQL requires InnoDB storage engine to support foreign keys.

In our example, we have the following parent table in a MySQL 5.7.21 server:

mysqld3-(root@localhost) [sakila]> show create table actor\G
*************************** 1. row ***************************
       Table: actor
Create Table: CREATE TABLE `actor` (
  `actor_id` smallint(5) unsigned NOT NULL AUTO_INCREMENT,
  `first_name` varchar(45) NOT NULL,
  `last_name` varchar(45) NOT NULL,
  `last_update` timestamp NOT NULL DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP,
  PRIMARY KEY (`actor_id`),
  KEY `idx_actor_last_name` (`last_name`)
) ENGINE=InnoDB AUTO_INCREMENT=201 DEFAULT CHARSET=utf8
1 row in set (0.00 sec)

and a foreign key is defined on the child table by using the “FOREIGN KEY… REFERENCES” syntax:

mysqld3-(root@localhost) [sakila]> show create …

Every MySQL database programmer should learn and apply the newly added MariaDB and MySQL Window Functions and Common Table Expressions(CTEs) in their daily work. Both CTEs and window functions enable easy solutions to many query challenges that in prior releases have been difficult and sometimes impossible to surmount. Mastering these features opens the door to query solutions that are more robust, execute faster, and are easier to maintain over time than prior solutions using older techniques.

In our last blog we compared User Roles in MySQL 8 vs. MariaDB.  Today, we will here compare Window Functions and Common Table Expressions in both databases.

Window Functions

While all database administrators are familiar with aggregate functions like COUNT(), SUM(), and AVG(), far less people make use of window functions in their queries.  Unlike aggregate functions, which operate on an entire table, window …

In part 1 of this series we looked at implementing support for the caching_sha2_password authentication plugin using the libmyqlclient C library.  This is possible for C based connectors or connectors that can make use of an external C library.  For some, this is not possible.  For example, Java and C# both function better if they don’t have to thunk out to an external library.  For these, implementing the MySQL client/server protocol natively is required.  This part 2 is about implementing support for the caching_sha2_password plugin natively.  It is expected that you already have a working connector and understand the client/server protocol.  For reference please internal documentation …

With MySQL 8.0, engineers improved the visibility of data locking. Usually it was not that easy to deal with the output of SHOW ENGINE INNODB STATUS and mixing Information_Schema tables like INNODB_LOCKS and INNODB_LOCK_WAITS.

So in MySQL 8.0 we instrument data locks in Performance_Schema and we changed the SYS Schema view related to the InnoDB Locks Wait too.

This means that when a transaction A is locking row R, and transaction B is waiting on this very same row, B is effectively blocked by A. What we added is extra instrumentation to expose which data is locked (R), who owns the lock (A), and who is waiting for the data (B).

Testing

Let’s have a look at this simple scenario in MySQL 5.7.21 and 8.0.4 to compare:

2 clients starting a …

As you may already know, MySQL 8.0 is coming with SQL Roles . I’ve already quoted them before in some posts (here and here). Giuseppe Maxia is doing a great job testing and promoting them (thank you for that!). Of course he also made some remarks on …

Prior to MySQL 8.0, MySQL was missing a very well know feature in other RDBMS: optimizer histograms.

The Optimizer Team implemented this feature as more and more MySQL DBAs were requesting it.

Definition

But what are histograms ? For wikipedia, a histogram is an accurate representation of the distribution of numerical data. For RDBMS, a histogram is an approximation of the data distribution within a specific column.

So in MySQL, histograms help the optimizer to find the most efficient Query Plan.

Example

To illustrate how histograms  influence the optimizer, I will use data from dbt3.

My query is simple: …

Come and listen to Percona CEO Peter Zaitsev on March 6, 2018 at 12:00 pm PST (UTC -8) / 3:00 pm EST (UTC -5) as Peter discusses Percona software news and what’s new in Percona open source software, including Percona Server for MySQL and MongoDB, Percona XtraDB Cluster, Percona XtraBackup, …

We've been busy in the MySQL Curriculum team and are pleased to announce the recent update of the MySQL Performance Tuning course.

This course has been substantially revised to take advantage of the performance enhancements available in MySQL 5.7, including new Performance Schema instrumentation and improvements in InnoDB and the Query Optimizer.

Even more so than in previous versions of the course, we try to filter out what's "nice to know" and focus on the really important configuration settings that can make the most positive impact on your server workload. We provide more in-depth coverage and practice of using system databases like Performance Schema/sys and graphical tools like MySQL Enterprise Monitor and MySQL Workbench that can help you identify performance bottlenecks and be proactive about resolving them before they start to become a major issue for your users. We also cover the use of Oracle tools like Oracle …