Automated server management with open source tools
In this session I would like to show you how we automate all the operations-related work at Gawker Media. We only use open source software, which we glue together, and at the end all of all of our developers can do all the ops tasks by themselves without asking for help, whether it is a datacenter failover, replica creation, development database clone or even new server provisioning. We use ansible, puppet, jenkins and we glue them together with python and go. Our developers are responsible for the full life-cycle of their code, including the live environment. So we have to provide them the tools to monitor and control the behavior of the production environment. And we have to do all that without requiring them to ssh connect to the live servers -- if that is needed, it's a bug. During the talk I'll demo these tasks (or at least some of them) on our live system to show that the operations can run quietly under the hood without affecting the live service. This talk could be interesting for all operations/dba people who don't want to spread themselves thin struggling with daily tasks, and want to focus on the big picture of their systems.
DBA, Gawker Media LLC
Balazs worked mostly as a UNIX/Linux engineer in the past 15 years. He joined Gawker Media in 2013 and since he is working as a dedicated DBA there. He works mostly on database stack, but besides that, he is participating in the other parts of ops operations. He is the author of the ansible's MySQL replication module. Father of two.