Just read the website.
-
First Name: Tim
-
Last Name: Meusel
-
Twitter: BastelsBlog
-
Company: GoDaddy EMEA
-
Title: DevOps Engineer
-
Company: GoDaddy EMEA
-
City: Cologne
-
Country: Germany
-
Location: Amsterdam
Tim (bastelfreak) Meusel works as a DevOps Engineer for GoDaddy EMEA in Cologne, Germany where he develops and maintains a big public cloud platform. Tim is the driving force behind various open source solutions at work. He founded the VirtAPI-Stack and is a very active Vox Pupuli Maintainer. Tim has been doing work in the DevOps area since 2009 and is persuing Puppet solutions since 2012. Recently he was reelected to serve on the Vox Pupuli Project Management Committee. He enjoys good BBQ and ice hockey.
- Title: Vox Pupuli - The Community behind Puppet
- Level: Beginner
Vox Pupuli is a group of more than 120 puppet module/tooling/documentation authors, all working together to ensure a continuing development process. In 2016 we elected our first Project Management Committee with one of its goals being to include Vox Pupuli as an official project under the Software Conservancy and to guide the community. We do not only provide a home for every orphaned puppet module; we also offer proper tooling and automation for every interested developer. Some of the most frequently used puppet modules (zabbix, collectd, archive, yum, nginx) are managed by us. Vox Pupuli is in the Top 10 of the most frequent Puppet Forge release groups. Have a look at our funny journey of hunting broken gems and changing upstream software! We invite everybody to participate. We will get into the details of what Vox Pupuli is, how you can interact/contribute/benefit. We will tell you some stories, funny and sad, about all the bugs we've discovered in the Puppet and Ruby ecosystem!
- Title: Scaling and monitoring your puppetserver for thousands of clients - including all pitfalls!
- Level: Advanced
Throwing hardware on your Puppetserver to scale it vertically works up to a certain level. But how do you proceed once you reach the limit? How do you distribute your agents over several Puppetservers evenly, but without hardcoded assignments? How do you handle the massively increased load on your Foreman and PuppetDB intances, after the Puppetserver isn't the bottleneck anymore? And how the heck can this all be monitored?
I would like to show you a walkthrough from a small environment with a single physical server for Puppetserver/Foreman/PostgreSQL and a handful of agents that got scaled up to several thousand agents. This also includes probably all common pitfalls that you can find yourself in on this journey!