Speaking at the 2012 Percona Live MySQL Conference

This year, I have the pleasure of returning to the MySQL Conference & Expo as a speaker. Percona have picked up the torch that O’Reilly had held as the conference organizers, and they’re putting together a 3-day conference this year. I am co-presenting a tutorial with Yves Trudeau from Percona.

Our tutorial is called High Availability Deep Dive: Pacemaker, DRBD, MySQL Replication, and more! and it’s going to be the only full-day tutorial offered in this year’s conference. In it, Yves and I are going to cover

• An overview of the Pacemaker cluster stack (the classic “this is Pacemaker” introduction)

  -

  DRBD-backed MySQL replication (another classic and widely deployed scenario)

  -

  MySQL replication under Pacemaker management (a new option which Yves has vastly improved through a big patch set to the MySQL RA).

Do I expect this talk to be controversial? Definitely. The amount of “Pacemaker is terrible” and “Pacemaker is unsuitable for managing highly available databases” that has been around the blogosphere lately is pretty mind-boggling.

But strangely enough, most of the things brought forward against Pacemaker by its detractors seem like a time-warp back to about 2007.

• “We must use XML to manage Pacemaker!” Nonsense. In fact, that was never true – the release of Pacemaker as a separate project and the release of the crm shell coincided. Ever since, Pacemaker configuration has been as text-based as MySQL itself.

  -

  “All Pacemaker can do is react to node failure!” Nothing could be further from the truth. Pacemaker has some of the most sophisticated resource monitoring and auto-recovery capabilities under the sun.

  -

  “OK. But all it can do to react to resource failure is kill a daemon!” Bogus again. It will happily do whatever the resource agent specifies. Or the admin, through the configuration. 

In our tutorial, we’re going to dispel a few of these myths. We certainly make no claims as to Pacemaker being the one and only solution for MySQL HA, but it’s one that serves lots of use cases excellently.

Needless to say, I’ll also hang around for the conference proper, and I’m very much looking forward to seeing lots of familiar faces. I’ll also remain in the Bay Area for some time after the MySQL conference – more on that in a day or two.

---

This article originally appeared on my blog on the hastexo.com website (now defunct).