In Percona Server for MySQL 8.0.29-21, we added one more patch that helps us to build server code on macOS. To be precise here, we still could do this even before this patch but only partially. Now it is possible to build RocksDB Storage Engine as well.

A word of disclaimer here, at the moment, by macOS we still understand macOS for Intel x86_64 architecture (the most recent ARM versions with Apple M1 / M2 processors are out of the scope of this blog post). Moreover, Percona does not provide …

There are two Operators at Percona to deploy MySQL on Kubernetes:

• one is based on Percona XtraDB Cluster (PXC), production-ready and widely adopted
• another is based on Percona Server for MySQL, now in tech preview and not recommended for production yet

We wrote a blog post in the past explaining the thought process and reasoning behind creating the new Operator for MySQL. The goal for us is to provide production-grade solutions to run MySQL in Kubernetes and support various replication configurations:

• Synchronous replication
  …

Percona Server for MySQL 8.0.28-20 introduces a new keyring component, component_keyring_kms, which allows using the AWS Key Management Service as a keyring. The component is also included in Percona XtraBackup 8.0.28, allowing backing up databases encrypted with this component.

The feature is similar to the AWS keyring plugin included in MySQL Enterprise, but it is implemented as a component instead, and because of that, it requires a different configuration.

Configuring AWS KMS

The component uses a single Master Key on KMS, which has to be created manually before configuring the component. This key can also be changed later, and existing keys using the old key will remain usable, as long as the key still exists …

Imagine you need to design an online system for storing documents on a per-user basis where nobody, including database administrators, would be able to change the content of those documents without being noticed by document owners.

In Percona Server for MySQL 8.0.28-20, we added a new component called Encryption UDFs – an open-source alternative to MySQL Enterprise Encryption that allows users to access a number of low-level OpenSSL encryption primitives directly from MySQL. This includes calculating digests (with a great variety of hash functions), asymmetric key generation (RSA, …

Temporary Tables handling in MySQL has a somewhat checkered past. Some time ago, my colleague Przemek wrote a post about differences in the implementations between versions of MySQL. If you are looking for the differences in implementation, that’s a perfect place to check.

In this post, I’d like to focus on Session Temporary Tablespaces, InnoDB on-disk internal temporary tables that come to play when a query requires storing more data than tmp_table_size or TempTable engine allocated more than …

As you are probably aware, Red Hat Enterprise Linux 6 (RHEL 6 or EL 6 in short) officially reached “End of Life” (EOL) on 2020-11-30 and is now in the so-called Extended Life Phase, which basically means that Red Hat will no longer provide bug fixes or security fixes.

Even though EL 6 and its compatible derivatives like CentOS 6 had reached EOL some time ago already, we continued providing binary builds for selected MySQL-related products for this platform.

However, this became increasingly difficult, as the MySQL code base continued to evolve and now depends on tools and functionality that are no longer provided by the operating system out of the box. This meant we already had to perform several modifications in order to prepare binary builds for this platform, e.g. installing custom compiler versions or newer versions of various system …

KMIP (Key Management Interoperability Protocol) is an open standard developed by OASIS (Organization for Advancement of Structured Information Standards) for the encryption of stored data and cryptographic key management.

Percona Server for MySQL 8.0.27 and Percona XtraBackup 8.0.27 now include a KMIP keyring plugin to enable the exchange of cryptographic keys between a key management server and the database for encryption purposes. The procedure to use them with HashiCorp Vault Enterprise is described below.

Install Hashicorp Vault Enterprise

We will first install Hashicorp Vault Enterprise on Ubuntu Linux “Bionic” and then enable …

The most common issue when using row-based replication (RBR) is replication lag due to the lack of Primary keys.

The problem is that any replicated DML will do a full table scan for each modified row on the replica. This bug report explains it more in-depth: https://bugs.mysql.com/bug.php?id=53375

For example, if a delete is executed on the following table definition:

CREATE TABLE `joinit` (
  `i` int NOT NULL,
  `s` varchar(64) DEFAULT NULL,
  `t` time NOT NULL,
  `g` int NOT NULL
) ENGINE=InnoDB DEFAULT CHARSET=latin1

With this amount of rows:

mysql> select count(*) from joinit;
+----------+
| count(*) |
+----------+
|  1048576 |
+----------+

The delete being:

mysql> flush status ;

mysql> delete from joinit where i > 5 and i < 150;
Query OK, 88 rows affected (0.04 sec) …

As we’ve communicated in our blog post in May, the TokuDB Storage Engine has been marked as “deprecated” in Percona Server for MySQL 8.0. It will be removed in a future version (Percona Server for MySQL 8.0.28, expected to ship in Q1 2022).

With the release of Percona Server for MySQL 8.0.26, the storage engine will still be included in the binary builds and packages but will be disabled by default. If you are upgrading from a previous version, the TokuDB Storage Engine plugin will fail with an error message at server startup if it is installed.

You will still be able to re-enable it manually so that you can perform the necessary migration steps.

Re-enabling the TokuDB …

When doing migrations or failovers in MySQL, there is usually a need to do a topology change and repoint replica servers to obtain replication data from a different server.

For example, given servers {A, B, and C} and the following topology:

If you need to repoint C to be a replica of B, i.e:

You can follow the next steps:

Note: log_replica_updates should be enabled on the soon-to-be primary as it is a prerequisite for chain replication.

Note: It is assumed that both replicas only stream from Server A and there are no conflicting replication filters in place that might break replication later on.

If Using File/Position-Based Replication:

1) Stop B and C

STOP REPLICA;

2) If replicas are multi-threaded, correct MTS gaps and make them …