Showing entries 1 to 10 of 102
10 Older Entries »
Displaying posts with tag: Blog (reset)
Taking Full Advantage of Galera Multi-Master Replication-Galera Cluster Resources Updated

Last year Codership produced a lot of valuable content to help Galera users to get started with Galera and manage Galera. We have gathered  the resources to our website.

Taking Full Advantage of Galera Multi-Master replication video can be watched here.

We have now uploaded Slideshare with many new presentations. Check them out!

The best source of multi-master Galera Cluster help,  Galera Cluster documentation, is being updated constantly.

In 2016 Galera Cluster surged past one million downloads – setting up for a great year in 2017

Helsinki, Finland & London, UK  – January 9th 2017 – Codership, whose Galera Cluster technology brings high availability (HA) and scalability to open source databases worldwide, has seen its technology break the one million download barrier. Codership’s customer roster has grown over 100 percent thanks to the adoption of OpenStack by enterprises and the recent strategic partnership with MariaDB.


Galera has been recognised as the most widely used OpenStack High Availability technology for the second year in a row. According to the OpenStack Survey 2016, one-third of users rely on Galera Clusters running on MariaDB and MySQL.


To cope with an increase of customer demand coming from various industries such as financial services, retail and telecoms, Codership has grown its engineering hires by 50 percent. The new team members will ensure Galera continues to deliver on its mission to …

[Read more]
To SST or Not To SST?


If a node leaves the cluster and subsequently rejoins, Galera will internally make sure that the node is brought up to speed with the rest of the cluster. It is important for DBAs that this process completes quickly, so ideally SST is avoided altogether and IST is used. However, when looking at the log, SST is still mentioned even if IST is ultimately chosen, which may cause confusion.

In this article, we will describe the entire process of getting a restarted node back to speed with the rest of the cluster and explain the logic behind the various log messages.

Basic Principles

State transfers in Galera are governed by the following basic principles:

  1. The cluster will pick a donor node using an algorithm that favors IST and attempts to avoid any transfers over a wide-area network. A specific donor can be explicitly chosen using the wsrep_sst_donor variable.
  2. The joiner …
[Read more]
Taking Full Advantage of Galera Multi-Master Webinar – Dec 13th



This webinar will be the second in our series on best practices to follow when using Galera Cluster.

In this part, we will discuss important topics related to multi-master setups:
* Practical considerations when using Galera in a multi-master setup
* Evaluating the characteristics of your database workload
* Preparing your application for multi-master
* Detecting and dealing with transaction conflicts

This is a webinar presented by Codership, the developers and experts of Galera Cluster.

DATE AND TIME FOR USA:  Tuesday, 9-10 AM PST (Pacific ), 13th of December




[Read more]
Faster Cluster Restarts in 3.19 with Gcache Recovery

Galera 3.19 comes with improvements to better handle the situation where the entire cluster needs to be restarted. In previous blog post, we already discussed one such feature: Safe-to-Bootstrap.

In this post, we will discuss gcache recovery, which allows for the cluster to be restarted quickly after it was shut down, thus reducing the overall duration of planned or unplanned downtime.

The Gcache and Whole-Cluster Restarts

In previous versions of Galera, When a node starts up, its Gcache is wiped clean. This means that this node can not serve as an IST donor to other nodes that are also starting up at the same time. This is especially important when the entire cluster is being restarted. When the second node starts, it hopes to quickly rejoin the cluster via IST from the first node:

[Read more]
Introducing the “Safe-To-Bootstrap” feature in Galera Cluster

Galera Clusters are generally meant to run non-stop, so shutting down the entire cluster is not required during normal operation. Yet, if there is a need to perform such a procedure, it is likely that it will be happening under pressure, so it is important for it to complete safely and as quickly as possible in order to avoid extended downtime and potential data loss.

Galera 3.19 includes two important improvements to whole-cluster restart: the “Safe-to-Bootstrap” protection and Gcache recovery. In this article, we will describe the first feature.

Whole-Cluster Restart

First, a few words on cluster restarts in general. Regardless of whether it was an orderly shutdown or a sudden crash of all nodes, restarting the entire cluster is governed by the following principles:

  • Since the old cluster no longer logically exists, a new logical cluster is being created
  • The first node that is being started …
[Read more]
Announcing Galera Cluster 5.5.53 and 5.6.34 with Galera 3.19 with improvements to cluster restart

Codership is pleased to announce a new release of Galera Cluster for MySQL consisting of MySQL-wsrep 5.6.34 and Galera 3.19, wsrep API version 25.

This release includes enhancements to make cluster restarts faster and safer, which we will be blogging about over the coming days. It also incorporates all changes up to MySQL 5.6.34, including security fixes.

This and future releases will be available from
The source repositories and bug tracking are now on

New features and notable fixes in the Galera replication library since last binary release by Codership (3.18):

  • A new wsrep provider option, gcache.recover, enables gcache to be recovered at restart. This allows the restarted node to serve IST to any other nodes that …
[Read more]
Meet Codership at MariaDB Developers Meet Up 6-8.10 Amsterdam and OpenStack Summit Barcelona 24-27.10

Come and meet Codership’s Galera Cluster experts at OpenStack Summit Barcelona 24-27.10. Learn more about our plans and latest Galera Cluster for MySQL 5.7 release.  Our booth is A 18.  If you want to set up a meeting with us, please email to .


Codership’s CEO Seppo Jaakola will give Galera Cluster for MySQL 5.7 presentation today, 5th of October, at the Percona Live at 2:40pm to 3:05pm. He will also be available for discussions at the MariaDB Developers Meet Up Amsterdam 6-8.10.

Announcing Galera Cluster 5.7 Beta

Codership is pleased to announce a new beta release of Galera Cluster for MySQL 5.7, consisting of MySQL-wsrep 5.7.15 and Galera 3.18.

Galera Cluster for MySQL 5.7 contains enhancements in multiple key areas, including security, performance and monitoring.

This is a beta-quality release and should not be used in production, however we encourage you to give it a try and we welcome all feedback. You can use the bug tracker at to submit issues or you can email us at

This release incorporates all changes up to MySQL 5.7.15 and is available as targeted packages and package repositories for Ubuntu 16.06 and CentOS 7 that can be downloaded from …

[Read more]
Codership is hiring Software Packaging Engineer!

Software Packaging EngineerJob Description & Main Responsibilities


Codership is looking for a part-time Build and Packaging Engineer. We use Jenkins to build various packages for multiple distros and the engineer will take ownership of the build pipeline and extend it to cover additional platforms and testing scenarios.

There are plenty of interesting problems to tackle on the job, from crafting .spec files with complex dependencies to automatically testing no-downtime rolling upgrades of a distributed system.

Main Responsibilities

  •  take the ownership of and maintain Galera Cluster packages on Linux (.RPM and .DEB) and BSD
  • maintain the automatic package building pipeline and extend it to cover additional platforms and testing scenarios


Desired Skills & Requirements


  • knowledge of …
[Read more]
Showing entries 1 to 10 of 102
10 Older Entries »