Looking for Trouble: How to rapidly diagnose and solve crazy problems when time matters

Abstract:

This talk will be about how to troubleshoot and resolve problems in your infrastructure when you're under the gun. It has occurred to me that a lot of junior people lack any framework from which to troubleshoot problems that happen in their infrastructure, whether in a crisis or just things that are happening on a regular basis. Most people are content with doing one of the 3 Rs (Restart, Reload, Reboot) but this generally makes the 4th R (the most important one) Root Cause, impossible to perform. I will talk about the attitude that you have to have, provide you with a methodology and some actual tools to start you on your journey from Padawan to Jedi as far as troubleshooting issues are concerned.

Speaker:

My first job out of college was writing CPU and device diagnostics in PDP-11 Assembler. This probably coloured my outlook on Software Engineering and Systems in general... trust but verify. Over the years, I've been a developer in Assembler, C, C++ and now Python. Also did a stint as a Sr. Business Critical Support Engineer for Enterprise Products at VMware and Nicira. I currently have the title of Sr. DevOps Engineer at StubHub/EBay where I push virtual StubHub environments to Dev teams for a living.