MySQL has InnoDB data encryption, and Galera Cluster has supported it since it appeared in the main server. The one thing that was not covered was the encryption of the Galera Cache (the galera.cache file).

Here is a simple extract from the binlog.000001 files.

strings binlog.* |grep Custom |wc -l
10000
strings galera.cache |grep Custom |wc -l
10000

strings galera.cache|tail -10
Customer9100
Customer9099
Customer9098
Customer9097
Customer9096
Customer9095
Customer9094
Customer9093
Customer9092
Customer9091

You need to edit your /etc/my.cnf to include:

early-plugin-load=keyring_file.so
keyring_file_data=/var/lib/mysql-keyring/keyring

Note that in this example we are using the keyring_file plugin, which stores keyring data in a file on the local server host. This is not intended for regulatory compliance. You need to use a key management server that protects encryption keys in key vaults or hardware …

We recently covered Upgrading your Galera Cluster from MySQL 5.7 to MySQL 8.0. We’ve also had questions on how the upgrade path from our usual Galera Cluster “community” edition to the Galera Cluster Enterprise Edition (EE), which includes many new features, like XA transaction support, arm64 support or GCache encryption.

First off, let’s start by installing a Galera Cluster on CentOS 7. The reason we have picked this distribution? Let’s edit the /etc/yum.repos.d/galera.repo and add:

[galera4]
name = Galera
baseurl = https://releases.galeracluster.com/galera-4/centos/7/x86_64
gpgkey = https://releases.galeracluster.com/GPG-KEY-galeracluster.com
gpgcheck = 1

[mysql-wsrep8]
name = MySQL-wsrep
baseurl = https://releases.galeracluster.com/mysql-wsrep-8.0/centos/7/x86_64
gpgkey = …

Continuing on from Setting up Galera Load Balancer (GLB), wouldn’t it be nice to ensure that Galera Load Balancer (GLB) handles failover, seamlessly?

Router:
------------------------------------------------------
        Address       :   weight   usage    map  conns
188.166.179.177:3306  :    1.000   0.500    N/A      1
  165.22.50.152:3306  :    1.000   0.500    N/A      1
   165.22.49.92:3306  :    1.000   0.000    N/A      0
------------------------------------------------------
Destinations: 3, total connections: 2 of 493 max

Pool: connections per thread:     1     1     0     0     0     0

Here is an example of a Galera Load Balancer (GLB) handling 3 backend MySQL servers, and it currently has 2 connections, evenly distributed between 2 hosts.

We can verify this by doing: select @@hostname; which shows:

Galera Load Balancer (GLB) is a scalable and performant, yet easy to use TCP/IP connection balancing proxy. It is the oldest, yet actively maintained load balancer in the MySQL ecosystem, with a wide array of customers using it in production.

Firstly, please request for binaries via contacting sales@galeracluster.com. Once you have access to the package repository, you’ll have access to RPMs. Installing the RPMs are straightforward, and you can also add it to your Yum repository. This blog presumes you already have access to the binaries.

You can start it up, really simply:

glbd --threads 6 127.0.0.1:3306 188.166.179.177:3306 165.22.50.152:3306 165.22.49.92:3306 …

A few weeks ago, we did a Galera Cluster webinar on Managed Galera Clusters on DigitalOcean.

Some of you may notice that the current version of Galera Manager is 1.6.4, and one of the major fixes we have in there is to support the new API token format of DigitalOcean. DigitalOcean has documented this here: Updated API Tokens new management features: in partnership with GitHub –Secret Scanning, Prefixes, and more! and as a consequence we needed to update to accept new API tokens, or it would fail in deployment. If you’re wondering how to create a …

The Galera Cluster Enterprise Edition

Codership is pleased to announce the Galera Cluster Enterprise Edition.

Galera Cluster Enterprise Edition consists of an enhanced Galera Cluster server, Galera Manager and Galera Load Balancer.

The enhanced Galera Cluster Server includes

1) A new Online Schema Upgrade method called Non Blocking Operations (NBO), adds more high availability to your cluster while performing schema changes.

• • Schema changes made on all cluster nodes in the same total order sequence, with much more efficient locking compared to the Total Order Isolation (TOI) method.
  • SET GLOBAL wsrep_OSU_method='NBO'

…