Recently I’ve had to rename a node on a production cluster – however, I did not have an experience on this yet so I decided to try on a sandbox and establish a known procedure. Without further ado, assuming we have 3 nodes and we want to rename ha3’s hostname to ha03.

[root@ha01 ~]# crm status
Last updated: Wed Jun  4 21:55:01 2014
Last change: Wed Jun  4 21:53:35 2014 via crm_attribute on ha01.localdomain
Stack: classic openais (with plugin)
Current DC: ha01.localdomain - partition with quorum
Version: 1.1.10-14.el6_5.2-368c726
3 Nodes configured, 3 expected votes
6 Resources configured


Online: [ ha01.localdomain ha02.localdomain ha3.localdomain ]

 Master/Slave Set: ms_MySQL [p_mysql]
     Masters: [ ha01.localdomain ]
     Slaves: [ ha02.localdomain ha3.localdomain ]
 Clone Set: ping_gw_clone [ping_gw]
     Started: [ ha01.localdomain ha02.localdomain ha3.localdomain ]

Here we go:

1. Put ha3 on standby:

   [root@ha01 ~]# …

Over the last few days, I integrated the MySQL 5.6 GTID version of the Percona Replication Manager (PRM) work of Frédéric Descamps, a colleague at Percona. The agent supports the GTID replication mode of MySQL 5.6 and if the master suffers a hard crash, it picks the slave having applied the highest transaction ID from the dead master. Given the nature of GTID-based replication, that causes all the other slaves to resync appropriately to their new master which is pretty cool and must yet be matched by the regular PRM agent.

For now, it is part of a separate agent, mysql_prm56, which may be integrated with the regular agent in the future. To use it, download the agent with the link above, the pacemaker configuration is similar to the one of the regular PRM agent. If you start from scratch, have a look …

My December 4 webinar, “Geographical disaster recovery with  Percona Replication Manager (PRM),”  gave rise to a few questions. The recording of the webinar and the slides are available here, and I’ve answered the questions I didn’t have time to address below.

Q1: Hi, I was wondering if corosync will work in cloud environment. As far as I know it is hard to implement because of no support of unicast or multicast.

A1: Corosync supports the udpu transport since somewhere in the 1.3.0 branch. udpu stands for udp unicast and it works in AWS for instance. Most recent distribution are using 1.4.x so it is easy to find.

Q2: For token wouldn’t it …

The technologies allowing to build highly-available (HA) MySQL solutions are in constant evolution and they cover very different needs and use cases. In order to help people choose the best HA solution for their needs, we decided, Jay Janssen and I, to publish, on a regular basis (hopefully, this is the first), an update on the most common technologies and their state, with a focus on what type of workloads suite them best. We restricted ourselves to the open source solutions that provide automatic failover. Of course, don’t simply look at the number of Positives/Negatives items, they don’t have the same values. Should you pick any of these technologies, heavy testing is mandatory, HA is never beyond scenario that have been tested.

Percona XtraDB Cluster (PXC)

…

Two of the more popular solutions are  MySQL Master HA and Percona Replication Manager.

MySQL Master HA

MHA is based on a set of Perl scripts that monitors for replication and server health. When a failover scenario will happen, it can do automatic failover to a slave OR from a selection of slaves you have configured to make as the new master. The good thing about this is, after the initial failover, there will be no succeeding attempts to fail back, this is to protect your data and consistency of the cluster. You can also configure to only have manual failover for scheduled maintenance and the like. Between PRM and MHA, MHA is really easier to manage.

http://code.google.com/p/mysql-master-ha/

Percona Replication Manager

PRM is based on …

Selecting the most appropriate solution for a MySQL HA infrastructure is as much a business and philosophical decision as it is a technical one, but often the choice is made without adequately considering all three perspectives.  When too much attention is paid to one of these aspects at the cost of the others, the resulting system may be over-engineered, poorly-performing, and/or various other flavors of suboptimal.

On Wednesday, June 5, at 10 a.m. PDT (1700 UTC), I will be presenting a webinar entitled, Choosing a MySQL HA Solution, in which we’ll explore the topic of MySQL HA from each of these perspectives.  The goal will be to motivate your thinking about HA in a holistic fashion and help …