For the 2016 policy please review this page: http://casitconf.org/casitconf16/conference-anti-harassment-policy/
50 minute talks on a wide variety of topics taught by top people in the field. Lightning talks are given in the final time slot of the conference. Attendees can sign up to give lightning talks during the the conference and present Saturday afternoon. Lightning talks are short impromptu talks given on any topic relevant to IT.
How do you upgrade a service while it is running? This class covers
eight techniques from the new book by Limoncelli/Chalup/Hogan, The
Practice of Cloud System Administration. Learn best practices from
Google, Facebook, and other successful companies and apply them to
your environment. Techniques include: The Google Canary process,
Facebook Dark Launches, proportional shedding, feature toggles, Erlang
live-code upgrades, and live SQL/NoSQL schema changes.
Being able to upgrade live systems is important in the 24/7 world of
the web, but also enables upgrades in the middle of the day, without
having to schedule night-time rollouts, which is a huge savings on
human capital.
: “Once upon a time”: powerful words that begin many oral
narratives and indicate that the story to be told will be imbued
withmagic and myth. Organizational folklore can be a very powerful
force for instilling or perpetuating behavior, systems, and culture
within an organization.
Too often, fear and negativity are the driving forces in the folklore
behind many organizational traditions. A positive narrative that
embraces the customs and traditions of a healthy, balanced feedback
loop can help jumpstart your DevOps journey. This talk will help you
frame your narrative alongside metrics and use folklore as a catalyst
for positive change.
Troubleshooting requires understanding the problem at sufficient depth to apply corrective measures. How do we gain that understanding? As the scientific method teaches us, we match theory to observation.
This presentation gives an overview of process execution at the system call and kernel level (theory). We then review sysdig, a relatively new tool for system inspection (observation). Finally, we review several use cases where things are awry, matching our theory to our observation to find a basis for applying corrective action.
Whether you call it Prod Ops, System Engineering, or simply “keeping it all working,” ops managers face some particular challenges. How do you build new projects and services while solving all the production emergencies caused by the old, broken infrastructure? How do you juggle the demands of other teams in the company while keeping the site running? Above all, how do you give your team agency and keep them happy in a high-pressure, distraction-driven, 24×7 environment? Learn practical skills and techniques for the human side of Ops in this talk by Connie-Lynne Villani.
This presentation is about writing functional tests for
use with configuration management that tests the actual functionality of
an entire system to ensure that while your automation evolves you are
stilling building the necessary functionality. Gain confidence in your
automation tooling.
You will learn how to test an actual
system after it has been built, presumably with the help of config
management (CM) software. This talk is aimed to be more CM tool agnostic
and also help folks that have not yet transitioned to using CM and give
them a road map and the confidence to begin. There will be live demo’s
interspersed with the talk.
Don’t like your monitoring system? Want something else? In this discussion I’ll go over techniques I used in my workplace to change our system from a 1200+ alerts per weekend status-quo, to one where each alert is actionable and customized for the team that should respond to it. Knowing what technology you want is only half of the problem. In larger teams, changing the culture is just as hard.
Many are trying to get started with DevOps and Continuous Delivery, but how exactly do we get there? We’ve all heard stories from the Unicorns of DevOps, but how do we distill those into actionable tasks we can start focusing on? There’s a long list of skills we need in order to move quickly and safely. How do we train our teams to get there? More importantly, if Mr Miyagi wanted to build a DevOps culture, what tasks would he make us do? We’ll look at emergent patterns in the DevOps world and distill a set of practices to focus on to sharpen our skills.
At Michigan State University, we have created a vendor-agnostic storage platform. This was driven by a need for low-cost, persistent, moderate-performance storage. This allows us to focus our efforts and resources on obtaining the best hardware, software, or services without worrying about vendor lock-in. We settled on ZFS-based file servers, running FreeBSD, made available via NFS and SMB on commodity hardware. For high-speed storage, we have standardized on the Lustre high-speed parallel tile system with varying degrees of vendor involvement. This presentation will detail the process of engineering these storage platforms and lessons learned during the process. A technical overview of MSU’s storage platforms and possible future plans will also be presented.
Test Driven Development is a popular software development methodology. A “test first” approach can be extremely useful for ensuring expected, reliable, repeatable results. Modern IT/ops best practice includes plenty of configuration management, but if we’re honest we could probably afford to do more testing. Starting the process of a production environment change by thinking about tests first is guaranteed to produce better results than just typing sudo.
Testing does not need to be a burden and can be extremely automated. And often the process of creating a test produces much simpler and more effective commands or code.
The talk will cover different types of tests and will provide examples of git hooks and Jenkins with virtual machines and Docker images.
The OpenStack CI infrastructure has grown to require over 3,800 Jenkins projects to run all of the builds and tests across every OpenStack project. Supporting that many projects can be quite insane so we created the Jenkins Job Builder (JJB) to help us automate and manage that complexity. This talk is to discuss JJB and how it helped us scale out the OpenStack CI infrastructure.
Wireshark is arguably the best network troubleshooting tool available today. It will help you find 2 machines with the same IP address, firewall malfunctions, software malfunctions, ARP failures, DHCP failures, HTTP problems.
IT Service Management for Technical Staff. Convinced that the terms ITSM & ITIL are just clever ways to hide additional work and bureaucracy? ITSM provides an important business function that can be an invaluable tool in demonstrating you and your team’s value to non-technical managers. During this interactive session I will solicit your thoughts and use them to drive the discussions. I will share with you my experiences, success stories and common pitfalls that make ITSM/ITIL crash and burn.
A dirty little secret in IT is that we don’t always know
everything we have, what our systems are doing or fully
monitor them. The Assimilation Project integrates continuous
discovery and monitoring, creating a graph CMDB of your
infrastructure and services – scalably monitoring them with
near-zero configuration. Come learn how to easily put your
infrastructure knowledge in one place, monitor your systems
services and configurations, and automatically update it and
examine it against best practices.
Puppet has been widely adopted by Linux server admins, but rarely is this expertise leveraged on Windows systems. Why not? In this talk, both experts and beginners will learn how Puppet behaves on Windows systems. We’ll look at (and demo!) how Puppet’s built-in resource types already solve the most pressing problems, and how to super-charge Puppet on Windows by:
