The Practical Gamemaster: Design & Execution of IT Emergency Operations Drills
Practical IT drill design brings together emergency response and operations, business continuity, disaster recovery, and IT architecture. During this talk, you will learn key concepts in emergency operations center and incident headquarters design, methods of introducing such concepts to your organization, and a sequence of basic-to-advanced drill designs. Keeping IT folks engaged in a drill simulation can be very challenging. Become a practical gamemaster worthy of designing and executing drills on likely emergency scenarios and realistic function failures for your organization. d10s included.
Adele Shakal currently heads up project and knowledge management at Metacloud, Inc., a cloud solutions company providing on-premises private cloud based on OpenStack. In her prior work in technical project management and system administration at USC ITS, she designed IT emergency operations and the ITS Great Shakeout 2011. She has more than two decades of experience in information technology, with Bachelor of Science in geochemistry from California Institute of Technology. She has presented at CENIC, EDUCAUSE, APRU, USENIX LISA and CascadiaIT conferences.
slides (PDF 2.9MB). Adele's website also has a copy of the slides and some additional info http://adeleshakal.com/2014/04/09/bblisa.
May IT Project Management
20 years ago SAGE was defining IT job skills sets for HR departments. Today (no surprise) IT departments are the heart beat of the enterprise. With burgeoning budgets and even more complex technologies management is trying to get more of a handle on their IT processes and align them with standard management methodologies. Sometimes with good results.
I will touch on TQM, ITIL, PMP & Agile practices and why some practices work and some do not. Much of it depends on the culture of the enterprise. Managing projects at large enterprises (e.g. Microsoft), higher education (e.g. Harvard) and startups is vastly different and requires a large bag of tricks. A major issue of project management is who is "The Boss"? Is it the technical lead, a non-technical manager or the project manager? In IT, it is often the technical leader that is the real boss, regardless of titles.
Most Project Managers come from a business background. But IT Project Managers better be able to understand the technical aspects of IT. If an IT Project Manager asks naive questions like "Do you really need fail over for DNS?" They will be abused and ignored by the technical staff. Hence, many IT professionals see Project Managers as someone to avoid. Management hears "The IT department is not cooperating" from the Project Management staff.
I will share my horror stories, but actually explain how Good IT Project Management can help your IT department be more successful.
Peg is the Senior Technical Project Manager for Starfish Storage.
slides (PPT, 2.4MB)
Video of IT Project Management talk (from youtube)
- Lay waste to waste (Lean Six Sigma)
This talk will cover the basic ideas and some background behind Lean and 6 Sigma (LSS), discuss their origins and then walk through some examples while introducing the tools and techniques. Plus you get to test the speaker by bringing your own problems and (hopefully) leave with some ideas on how to tackle resolving the problem.
John Rouillard has been a system administrator trying to fix practices for more decades than he cares to remember. His interest in formal problem solving techniques developed by watching solutions that didn't really solve problems but merely added more problems while allowing the original issue to generate even more failures.
slides (ODP (open office), 1.8MB)
slides (PDF, 800KB)
Video of Lay waste to waste talk (from youtube)
- Trying to Outpace Log Collection with ELK
This talk will detail the adventures of centralizing log data with the ELK stack: Elasticsearch, Logstash, and Kibana. We wanted to centralize logs from many very separate networks because it's a really good idea to look at logs more than you probably are. The adventures in this project include building concurrent prototypes with Logstash, Graylog2, and Splunk. There's a bit of layer 7 routing and buffering with RabbitMQ. I might get a bit DevOps-y in the description of using Chef to deploy all the systems in this project. And finally, there are some lessons learned about running distributed magical databases like ElasticSearch in Amazon AWS. tldr; Lots of stuff only breaks when it gets big enough to break.
Neil Schelly has been a sysadmin, developer, or consultant for most of the last 20 years. Currently, he's a Principle Security Administrator at Dyn, Inc in Manchester, NH. Lately, there's been more of a security, auditing, and monitoring focus to the projects he is working on.
slides and notes(PDF, 3.2MB)
- Project Atomic: Server OS and app container delivery with Docker and OSTree
For many years, "traditional packaging" has been the default method for software delivery and management on Unix systems; exemplified by dpkg, RPM, and similar systems.
Project Atomic is a pattern that can be applied to a traditional distribution (for example, Fedora, CentOS, Red Hat Enterprise Linux), bringing together several upstream components: SELinux, Linux kernel containers, Docker, (RPM-)OSTree, and orchestration frameworks such as geard and Kubernetes.
This talk will explore the details of all of these technologies, with a particular focus on the RPM-OSTree side, which provides atomic upgrades and rollbacks for bare metal operating systems For example, we'll look at why /home is a symlink to /var/home on an Atomic system, and how configuration files in /etc are carried forward on upgrades in a fully atomic fashion.
Colin Walters is a Senior Software Engineer at Red Hat in the Server Experience group. He is the upstream author of OSTree, and has contributed to a variety of Free Software projects, such as Emacs, systemd, GNOME, OpenEmbedded, and both Debian and Fedora packaging.
slides and notes(PDF, 600KB)