Swapping has always been something bad for MySQL performance but it is even more important for HA systems. It is so important to avoid swapping with HA that NDB cluster basically forbids calling malloc after the startup phase and hence its rather complex configuration. Probably most readers of this blog know (or should know) about [...]
Minimizing Downtime from Lengthy AWS Outages
October 24, 2012 By 6 Comments
Well, it happened again… Another lengthy EBS outage in the US-East region impacted several sites across the net. While failures like this are rare, they can be quite costly and translate into headaches for the operations team when impact production systems for any length of time. At Percona, we routinely help clients architect and deploy [...]
Automation: A case for synchronous replication
September 19, 2012 By 2 Comments
Just yesterday I wrote about math of automatic failover today I’ll share my thoughts about what makes MySQL failover different from many other components and why asynchronous nature of standard replication solution is causing problems with it. Lets first think about properties of simple components we fail over – web servers, application servers etc. We [...]
Why ALTER TABLE shows as two transactions in SHOW ENGINE INNODB STATUS
July 1, 2012 By 4 Comments
When executing an ALTER TABLE, InnoDB (and XtraDB) will create two InnoDB transactions: One transaction is created when the table being ALTERed is locked by the server. This will show up as something like “TABLE LOCK table `schema`.`table_name` trx id XXXX lock mode S” in SHOW ENGINE INNODB STATUS. Another is created when adding or [...]
Percona XtraDB Cluster reference architecture with HaProxy
June 20, 2012 By 63 Comments
This post is a step-by-step guide to set up Percona XtraDB Cluster (PXC) in a virtualized test sandbox. I used Amazon EC2 micro instances, but the content here is applicable for any kind of virtualization technology (for example VirtualBox). The goal is to give step by step instructions, so the setup process is understandable and [...]
How to recover deleted rows from an InnoDB Tablespace
February 20, 2012 By 4 Comments
In my previous post I explained how it could be possible to recover, on some specific cases, a single table from a full backup in order to save time and make the recovery process more straightforward. Now the scenario is worse because we don’t have a backup or the backup restore process doesn’t work. How [...]
Return of the Query Cache, win a Percona Live ticket
August 26, 2011 By 4 Comments
It’s Friday again, and time for another TGIF give-away of a Percona Live London ticket! But first, what’s new with the MySQL query cache? You may know that it still has the same fundamental architecture that it’s always had, and that this can cause scalability problems and locking, but there have been some important changes [...]
Tutorial Insights for Percona Live, London
August 24, 2011 By Leave a Comment
We have a great line up of Tutorials on Percona Live, London. I hand picked number of them after seeing outstanding speaker Performance in other Places. Let me tell in little bit more details about people we have invited and their talks. Yoshinori Matsunobu Talk on Linux Hardware and Optimizations for MySQL at Oreilly MySQL [...]
How to replace a NDB node on EC2
June 9, 2011 By 10 Comments
NDB cluster is a very interesting solution in term of high availability since there are no single point of failure. In an environment like EC2, where a node can disappear almost without notice, one would think that it is a good fit. It is indeed a good fit but reality is a bit trickier. The [...]
Percona welcomes Stewart Smith
May 16, 2011 By 8 Comments
Percona is pleased to welcome Stewart Smith to the team. Stewart does not need an extended introduction for MySQL Community, but just in case: Stewart has a long history with both the MySQL and Drizzle code bases. He’s been one of the core Drizzle developers since the start of the project (working on Drizzle for [...]
