I could have set MySQL between parenthesis in the title as this article is more about how to use OpenTofu to deploy on OCI.

I will explain how to install OpenTofu and how to use it to deploy on OCI. I will also mention what are the required changes be able to use my previous Terraform deployment files.

As an example, let’s use the modules to deploy WordPress with MySQL HeatWave Database Service: oci-wordpress-mds.

Installing OpenTofu

If like me you are using a RPM based Linux distro, you can find the necessary information to create the yum repository on OpenTofu’s website:

$ sudo su -
# cat >/etc/yum.repos.d/opentofu.repo <<EOF
[opentofu]
name=opentofu …

We recently saw that we can benefit from using JSON documents with MySQL HeatWave cluster in OCI (HeatWave accelerator for MySQL Database Service).

However sometimes the data can’t be loaded to HeatWave Cluster or the query cannot use it.

And this is not always easy to understand why. Let’s get familiar on how to find the info.

Data not loaded in HW Cluster

Let’s see an example with this table (collection):

SQL > show create table listingsAndReviews\G
*************************** 1. row ***************************
       Table: listingsAndReviews
Create Table: CREATE TABLE `listingsAndReviews` (
  `doc` json DEFAULT NULL,
  `_id` char(28) GENERATED ALWAYS AS
 (json_unquote(json_extract(`doc`,_utf8mb4'$._id'))) STORED NOT NULL,
  `_json_schema` json GENERATED ALWAYS AS 
  (_utf8mb4'{"type":"object"}') VIRTUAL,
  PRIMARY KEY (`_id`),
  CONSTRAINT `$val_strict_B70EB65BDDBAB20B0EE5BB7532C9060C422A06F8` …

Since the release of MySQL 8.0, the MySQL X Dev API has provided users with the convenient ability to utilize MySQL without the need to write a single line of SQL!

MySQL X Dev API brings the support for CRUD operations on JSON documents that are stored in MySQL. MySQL Document Store is ACID compliant and is compatible with everything MySQL like replication, InnoDB Cluster, …

The MySQL X Dev API is available using the MySQL X Protocol, listening by default on port 33060.

If you are interested in learning more about MySQL Document Store, please refer to the these presentations [1], [2], [ …

Some time ago, we saw how we could deploy WordPress on OCI using MySQL HeatWave Database Service with Read Replicas. We had to modify WordPress to use a specific plugin that configures the Read/Write Splitting on the application (WordPress): LudicrousDB.

Today, we will not modify WordPress to split the Read and Write operations, but we will use MySQL Router 8.2.0 (see [1], [2], [3]).

Architecture

The …

The ability to store data on Object Storage and retrieve it dynamically when necessary is a notable advantage of Lakehouse when managing MySQL historical data we would like to archive.

Let’s illustrate this with the following table:

CREATE TABLE `temperature_history` (
  `id` bigint unsigned NOT NULL AUTO_INCREMENT,
  `time_stamp` timestamp NULL DEFAULT CURRENT_TIMESTAMP,
  `device_id` varchar(30) DEFAULT NULL,
  `value` decimal(5,2) NOT NULL DEFAULT '0.00',
  `day_date` date GENERATED ALWAYS AS (cast(`time_stamp` as date)) STORED NOT NULL,
  PRIMARY KEY (`id`,`day_date`),
  KEY `device_id_idx` (`device_id`)
) ENGINE=InnoDB AUTO_INCREMENT=129428417 DEFAULT CHARSET=utf8mb4 COLLATE=utf8mb4_0900_ai_ci SECONDARY_ENGINE=RAPID
/*!50500 PARTITION BY RANGE  COLUMNS(day_date)
(PARTITION p0_before2023_11 VALUES LESS THAN ('2023-11-01') ENGINE = InnoDB,
 PARTITION p2023_12 VALUES LESS THAN ('2023-12-01') ENGINE = InnoDB,
 PARTITION p2024_01 VALUES LESS THAN …

You may have come across numerous presentations showcasing MySQL HeatWave as a Query Accelerator for MySQL. However, if you have not yet had the opportunity to test it yourself (if you have, you already know the answer), allow us to conduct a test using actual data to determine the potential benefits of utilizing a MySQL HeatWave Cluster.

Data & Queries

The data is simple, some arduinos with a DHT22 sensor sending temperature and humidity.

So first let’s have a look at the amount of collected data:

select format_bytes(sum(data_length)) DATA,
       format_bytes(sum(index_length)) INDEXES,
       format_bytes(sum(data_length + index_length)) 'TOTAL SIZE'
from information_schema.TABLES order by data_length + index_length;
+-----------+-----------+------------+
| DATA      | INDEXES   | TOTAL SIZE |
+-----------+-----------+------------+
| 21.89 GiB | 14.06 GiB | 35.95 GiB  |
+-----------+-----------+------------+
1 row …

Since MySQL Shell 8.1, it’s even easier to create a logical backup of your MySQL instance and store it directly in Object Storage, an internet-scale, high-performance storage platform in Oracle Cloud Infrastructure (OCI).

MySQL Shell now offers the option of dumping to Object Storage Bucket using PAR (Pre-Authenticated Request).

Bucket Creation

The first step is to create an Object Storage Bucket in the OCI Console:

Let’s call it lefred-mysql-backups:

When created, we can click on the three-dots and create a new PAR:

Let’s see how to deploy WordPress and MySQL on a Kubernetes Cluster. The Kubernets cluster we are using is OKE (Oracle Kubernetes Engine) in OCI (Oracle Cloud Infrastructure):

OKE Cluster

We start by creating a Kubernetes Cluster on OCI using the Console:

We select the Quick create mode:

We need to name our cluster and make some choices:

When created, we can find it in the OKE Clusters list:

And we can see the pool of workers nodes and the workers:

kubectl

I like to use kubectl directly on my latop to manage my K8s Cluster.

On my Linux Desktop, I need to install kubernetes-client package (rpm).

Then on the K8s Cluster details, you can click on Access Cluster to get all the commands to use:

We need to copy them on our terminal and then, I like to also enable the bash completion for …

To copy a MySQL server to another server or to the cloud, there are several ways.

We can distinguish between two different types of copy:

• physical copy
• logical copy

The physical copy is often the fastest. However, it requires some tools to ensure that you have a consistent online backup. For example, you can use MySQL Enterprise Backup (MEB).

Alternatively, it’s possible to use the CLONE plug-in to provision a new instance with existing data from a source server. This is my preferred approach.

Finally, the last physical solution is the use of a file system snapshot, but this requires the right infrastructure and even more care to have a consistent …

Sometimes it’s convenient to retrieve the user creation statement and to copy it to another server.

However, with the new authentication method used as default since MySQL 8.0, caching_sha2_password, this can become a nightmare as the output is binary and some bytes can be hidden or decoded differently depending of the terminal and font used.

Let’s have a look:

If we cut the create user statement and paste it into another server what will happen ?

We can see that we get the following error:

ERROR: 1827 (HY000): The password hash doesn't have the expected format.

How could we deal with that ?

The solution to be able to cut & paste the authentication string without having any issue, is to change it as a binary representation (hexadecimal) like this:

And then replace the value in the user create statement:

But there is an easier way. MySQL …