Chaos Computer Club - archive feed

cloud costs can feel like a nightmare, creeping up on your Kubernetes infrastructure. But with the right tools, you can be the hero your budget deserves! In this talk, we’ll dive into OpenCost, an open-source solution that can help you track and optimize your cloud spending in real time.You’ll discover how OpenCost works, why it matters, and how you can use it to become the cost-saving champion of your cloud environment. Get ready to conquer Kubernetes costs and take back control of your cloud! Licensed to the public under https://creativecommons.org/licenses/by-sa/4.0/ about this event: https://cfp.cfgmgmtcamp.org/ghent2025/talk/PVN3XX/

This talk will walk through and provide examples and war stories on how kubernetes can be used not only in large scale environments but also in small and small-ish scale environments. Kubernetes is often considered as the tool to tackle large scale traffic, which is supposed to be used by a big team of engineers. This talk presents an opposite approach which shows how Kubernetes can be used in a very small team with limited resources. It will explore the benefits of running k8s in a small scale and also what pitfalls come with it. It will walk through the steps of provisioning self hosted Kubernetes cluster - kOps - challenges of keeping clusters upgraded without downtime. It will discuss issues encountered in daily operations, applications taking too long to start up anyone, and then how it was tuned with tools like Goldilocks. It will delve into CI/CD on Kubernetes (using Jenkins and ArgoCD). Keeping an eye on operational costs is essential in a small environment and this talk will discuss how kOps can utlize spot instances everywhere and benefits/challenges with spot instances. The idea of downscaling on schedule with py-kube-downscaler project, mutating pods with kyverno will be discussed. Licensed to the public under https://creativecommons.org/licenses/by-sa/4.0/ about this event: https://cfp.cfgmgmtcamp.org/ghent2025/talk/A3QEP7/

Kubernetes is still quite a popular choice with wide community adoption to run containerised workloads in the Cloud, but it doesn’t come with batteries included. And some of that is intentional to allow freedom to make different choices or extend its functionality as needed. For example scaling compute nodes is one of the things which is not built-in. Making sure you’re doing it in most efficient and cost-efficient way is paramount. But it’s not just efficienty than separates Karpenter (an open-source node lifecycle management) from other options, but also how it can help you stay on top with compliance, patching and drift. The project has come a long way in the last couple of year and it was also adopted by CNCF/SIG Autoscaling making it alternative approach compared to de-facto Cluster Autoscaler project. I this talk I’ll show how to set it up, different use cases and demonstrate hands-on what to expect in the real world scenario. Licensed to the public under https://creativecommons.org/licenses/by-sa/4.0/ about this event: https://cfp.cfgmgmtcamp.org/ghent2025/talk/P7CTWQ/

Puppet is a mature tool, the company behind it has changed over the years and most of the people who developed it, are no more working there. For somebody Puppet is old, solving problems that are no more current. Yet, Puppet is still around , and as long as there'll be systems to manage over time, there'll be the need of such a tool. The question is if the tool of choice is going to be Puppet or not. What's its present and future? We will analyse the current Puppet situation, market demand and perception, and spend our two cents on what could be done to improve perception, usage and adoption. We will also try to raise the topic with the people in the audience, when the presentation will turn into a discussion, possibly stirring ideas and suggestions. Licensed to the public under https://creativecommons.org/licenses/by-sa/4.0/ about this event: https://cfp.cfgmgmtcamp.org/ghent2025/talk/VQUFXW/

Overview of possibilities to assign classes to nodes The Puppet tutorial assembles configuration aus snippets in manifests/site.pp node default { include apache } . There are more possibilities than advertised by Puppet: * External Node Classifier * Roles und Profiles * Hiera Chainloading as Array or Hash * Puppet Enterprise Console/Foreman Host Groups We will get a quick intro to each of them, an explanation on how to shoot yourself in the knee with these and a field report of ways that have proven to cause less pain. Licensed to the public under https://creativecommons.org/licenses/by-sa/4.0/ about this event: https://cfp.cfgmgmtcamp.org/ghent2025/talk/TCKR7P/

Managing compliance in infrastructure as code environments is essential but can be daunting. Enter `compliance_engine`, a new open-source Ruby gem designed to streamline the mapping of compliance standards to Puppet code. Building on the foundation of SIMP's `compliance_markup`, this reimagined backend prioritizes performance, flexibility, and maintainability. In this session, we’ll explore the evolution from `compliance_markup` to `compliance_engine`, highlighting the architectural improvements that make it faster and easier to use. We’ll dive into real-world examples, demonstrating how the gem simplifies the enforcement of compliance policies, reduces complexity, and supports emerging standards. Attendees will gain insights into the challenges of implementing compliance as code and learn how `compliance_engine` can transform their approach to regulatory compliance in Puppet environments. Licensed to the public under https://creativecommons.org/licenses/by-sa/4.0/ about this event: https://cfp.cfgmgmtcamp.org/ghent2025/talk/NXJTDG/

What is new in HDM Release 3? Hiera Data Manager (HDM) is a web UI, which provides insight into your Hiera Data. One can easily check, which values are set in which layer and recognize, why a node receives which configuration data. With the newest release we added some new features, which I would like to present. Licensed to the public under https://creativecommons.org/licenses/by-sa/4.0/ about this event: https://cfp.cfgmgmtcamp.org/ghent2025/talk/FCSFJP/

Kubernetes is the most popular container orchestration platform out there, and for anyone who wants to do GitOps on Kubernetes, ArgoCD is a leading open source project in this space. This presentation will walk you through the management of multi-architecture applications for Kubernetes with ArgoCD. In this presentation, we will run through the process of managing container applications on hybrid arm64 and x86 Kubernetes clusters using ArgoCD for GitOps, including: * Why add arm64 compute nodes to your Kubernetes clusters? * Tooling to build and manage multi-arch containers * Continuous integration and delivery patterns * Workload placement and orchestration in Kubernetes Licensed to the public under https://creativecommons.org/licenses/by-sa/4.0/ about this event: https://cfp.cfgmgmtcamp.org/ghent2025/talk/L7WJRN/

Managing secrets in Kubernetes can be a complex and overwhelming process, especially with the wide range of available options. This talk, designed for intermediate users, aims to demystify the process by providing a practical roadmap drawn from my own journey. I will explore common challenges and share insights from transitioning through various approaches, from Kubernetes' built-in secrets to external tools like Sealed Secrets, CSI Secrets Store, and External Secrets. Through real-world examples and lessons learned, attendees will leave with actionable strategies to manage secrets more securely and efficiently in their Kubernetes environments, while contributing to stronger community practices and more resilient applications. Licensed to the public under https://creativecommons.org/licenses/by-sa/4.0/ about this event: https://cfp.cfgmgmtcamp.org/ghent2025/talk/HS8ELE/

A real life view into how an enterprise company uses Choria for orchestration and what we had to build around it. This talk gives the basics of Choria along with infrastructure considerations such as running only from Jenkins and code considerations including control repo organization, org specific stdlib and interacting with other teams. Licensed to the public under https://creativecommons.org/licenses/by-sa/4.0/ about this event: https://cfp.cfgmgmtcamp.org/ghent2025/talk/W7WAZG/

How do you upgrade 3000 individual PE environments? 3000 environments that you don't manage and others own. 3000 environments without SSH access. Come with me on a "funny" journey and learn how we made this possible and how the PE upgrade process differs from Open Source. Lets do a deep dive into PE 2019->2021->2023 Upgrades and our open source tooling that made this possible. You can also watch the slides online at: https://bastelfreak.de/cfgmgmtcamp2025/pe.html#1 Licensed to the public under https://creativecommons.org/licenses/by-sa/4.0/ about this event: https://cfp.cfgmgmtcamp.org/ghent2025/talk/9NUL9E/

Monitoring Kubernetes doesn’t have to be complicated. In this talk, I’ll introduce a new module we’re developing for Icinga, currently in beta, that simplifies Kubernetes monitoring in the same way Icinga has for traditional infrastructure. We’ll explore how this module makes it easier to monitor your clusters’ health and performance, allowing you to identify issues early. Whether you’re new to Kubernetes or managing large-scale clusters, this session will provide a preview of what’s to come and how it can streamline your monitoring processes. Feedback and insights are welcome as we refine the tool. Kubernetes offers powerful orchestration capabilities, but monitoring its dynamic environment can be tricky. In this session, we’ll dive into the development of a new module for Icinga that simplifies Kubernetes monitoring, making it more accessible for users who are familiar with traditional IT infrastructure setups. While the module is still in the beta phase, I’ll walk through its current features, show how it integrates with your existing Icinga setup, and discuss future enhancements. We’ll look at practical examples of monitoring critical aspects like node health, pod status, and resource utilization, all through Icinga’s familiar interface. Licensed to the public under https://creativecommons.org/licenses/by-sa/4.0/ about this event: https://cfp.cfgmgmtcamp.org/ghent2025/talk/AH78JZ/

Description: Most of us remember how long it took for Puppet to get Debian 12 packages. The build pipeline was long and complex and used a lot of internal tooling that had to be updated manually. In current news though, the new OpenVox build pipeline has been totally revamped and simplified and adding support for RHEL 10 took about 10 minutes. Most of that was spent waiting for the build to complete. Nick would like to explain how it works and what we still have left to do. Licensed to the public under https://creativecommons.org/licenses/by-sa/4.0/ about this event: https://cfp.cfgmgmtcamp.org/ghent2025/talk/7HXT7V/

It is very common now for developers to code and test their applications on VMs, either locally hosted or on the cloud. As individuals have editor preferences (nvim, vscode, etc), so they have hypervisor. Once you create a bolt inventory file listing the server or servers, then bolt can easily configure those servers using custom puppet code. Instead of manually creating the bolt inventory, it is easy to create a dynamic inventory plugin--if it doesn't already exist--to suit your particular use case. This talk illustrates how we setup our own local dynamic inventory plugins to help with our automated development and testing. Licensed to the public under https://creativecommons.org/licenses/by-sa/4.0/ about this event: https://cfp.cfgmgmtcamp.org/ghent2025/talk/GYKK3P/

Testing Puppet code can be a hassle, but voxbox is here to save the day! Testing Puppet code can be a hassle, but voxbox is here to save the day! Voxbox is a complete testing environment in a container, with all the good gems from Vox Pupuli. Active maintained and ready to run locally or in your CI. It also has jq and yamllint on board. I will showcase how it is build, how it is used and how it can be integrated into gitlab-ci. Licensed to the public under https://creativecommons.org/licenses/by-sa/4.0/ about this event: https://cfp.cfgmgmtcamp.org/ghent2025/talk/WHTKEC/

To understand the inner workings of Kubernetes and to prepare for the K8s certification exams, I decided to create a K8s cluster from scratch, the hard way, on premises (“de meterkast”) on virtual machines all using Alpine Linux. This talk is how I tried to do it, how I succeeded, failed and added a CEPH cluster and ETCD cluster along the way. It includes a lot of technical details, but if there is one thing that you should learn during this talk, it’s not about K8s at all: Containers are not VMs! Licensed to the public under https://creativecommons.org/licenses/by-sa/4.0/ about this event: https://cfp.cfgmgmtcamp.org/ghent2025/talk/CR8UGL/

In this talk well discuss what's happened in the open source product releases from Puppet to developer tools recently and what direction we're thinking.. did anyone say Puppet 9? We will also look at an overview of the state of community and where we think we can focus working better together. Licensed to the public under https://creativecommons.org/licenses/by-sa/4.0/ about this event: https://cfp.cfgmgmtcamp.org/ghent2025/talk/3BC7QX/

about this event: https://c3voc.de

Let's join in a quiet moment to bid farewell to the chaotic wonderland that has been 38C3 and prepare ourselves for the harsh reality outside. Gather round and take a deep breath and enjoy the unique atmosphere before you will feel the spirit again at the next hacker event close to you. Licensed to the public under http://creativecommons.org/licenses/by/4.0 about this event: https://events.ccc.de/congress/2024/hub/event/38c3-return-to-legal-constructions/

This talks gives a behind the scenes on how the infrastructure side of the event is done. A lot of teams help to make this event happen. This talk gives them the opportunity to show you what they do and how they do it. Licensed to the public under http://creativecommons.org/licenses/by/4.0 about this event: https://events.ccc.de/congress/2024/hub/event/38c3-infrastructure-review/

Der IT-Sicherheitsalptraum-Rück­blick: Manchmal belustigend, zuweilen beunruhigend, aber mit Ausblick. Es ist wieder ein Jahr vergangen und niemand ist von einem Smartmeter erwürgt worden: Ist überhaupt etwas Berichtenswertes passiert? Und wenn nein, wird es denn nächstes Jahr wenigstens schlimmer? Wir betrachten das vergangene Jahr, versuchen Muster zu erkennen und zu ahnen, wie es weitergehen muss, denn vorgewarnt zu sein, heißt gewappnet zu sein. Und sei es nur mit Popcorn und „In Übereinstimmung mit der Prophezeihung!“-Schildern. Publikumseinwürfe willkommen. Licensed to the public under http://creativecommons.org/licenses/by/4.0 about this event: https://events.ccc.de/congress/2024/hub/event/security-nightmares/

Wetter- und Notfallwarnungen empfangen zu können kann Leben retten. Nutzende, die ihre Privatsphäre nicht Google oder Apple ausliefern möchten sollten dabei nicht im Nachteil sein. Wir berichten über den aktuellen Stand der FOSS Entwicklung und allerlei Beobachtungen rund um Notfallwarnungen. Die Flutkatastrophe vom Juli 2021 hat schmerzlich bewusst gemacht, wie wichtig die effektive Verteilung von Katastrophenwarnungen ist. Mit der Einführung von Cell-Broadcast in Deutschland gab es diesbezüglich eine deutliche Verbesserung, andere Verbreitungswege werden dadurch aber nicht weniger relevant. Apps wie NINA oder KATWARN stellen mehr Informationen zur Verfügung als in einer Cell Broadcast Nachricht übermittelt werden kann, und ermöglichen es auch, Regionen zu beobachten, in denen man sich nicht selbst aufhält. Diese Apps sind allerdings nur für die Plattformen von Google und Apple verfügbar, Nutzende freier Plattformen sind außen vor. Kein befriedigender Zustand. Was macht man in so einem Fall? Na, das, was man in so einem Fall immer macht: Wir bauen uns die Warn-Apps und die dazu nötige Infrastruktur halt selbst. Basis dafür bildet das Common Alerting Protocol (CAP) was seit vielen Jahren weltweit im Einsatz ist, und UnifiedPush als freie Alternative zu proprietären Push-Benachrichtigungen. Daraus ergibt sich ein Aggregations-Server der Warnmeldungen aus derzeit 100 Ländern einsammelt und Clients über Ereignisse in für sie relevanten Gebieten informiert. In diesem Talk erklären wir, wie CAP funktioniert, wie das in der Welt eingesetzt wird und welche merkwürdige Beobachtungen wir während der Entwicklung gemacht haben. Von den Entwicklern von FOSSWarn und dem FOSS Public Alert Server. Licensed to the public under http://creativecommons.org/licenses/by/4.0 about this event: https://events.ccc.de/congress/2024/hub/event/wie-man-auch-mit-foss-katastrophen-warnungen-bekommt/

Ihr wolltet schon immer wissen was der „Morgenthau-Plan“ mit Kreisverkehren und „Schönwetterfreizeitsportgeräten“ zu tun hat? Dann lasst mich euch mitnehmen in die wundersamen, obskuren und humoristisch wertvollen Untiefen eines lokalen Nachrichtenportals. Was kann die interessierte Beobachterin von außen über das System lernen? Welche Werkzeuge brauchen wir für diese Expedition? Welche Kreaturen der Nacht kriechen durch die Untiefen der anonymen Kommentarfunktion? Und kann man eigentlich auch etwas Schönes aus den Daten machen, die da täglich ins Netz gekippt werden? Wie viele Orte in Deutschland hat Lübeck eine von diesen etwas schrägen Lokalnews-Seiten, die wirken, als wären sie in der Zeit stecken geblieben. Aber dennoch sind sie irgendwie wichtig sind für das Leben in der Region. Der schnöde Wetterbericht, Beschwerden über Baustellen, Filz-Workshops und Veranstaltungsankündigungen für die LAN-Party des CDU-Ortsverbandes - alles kann einem hier begegnen. Natürlich garniert von Kommentaren aus dem ganzen Spektrum des Wahnsinns. Seit über einem Jahr sammle ich die Daten, die diese obskure Seite ins Internet bläst, werte sie aus und bastele daraus nützliche oder wenigstens lustige Dinge. Von all diesen Abenteuern meines Hobby-Projekts „hl-lol“ möchte ich euch berichten. Licensed to the public under http://creativecommons.org/licenses/by/4.0 about this event: https://events.ccc.de/congress/2024/hub/event/lokalnews-mining/

This talk examines philosophical, legal, and ethical questions of the merging of human minds with intelligent machines through Brain-Computer-Interfaces, provides an overview of current debates and international regulatory development - and what might be at stake when technologies increasingly access the human brain. Human minds and machines, or organic and artificial intelligence (AI), are increasingly merging through neurotechnologies such as Brain-Computer-Interfaces (BCIs) that may record or alter brain activity. While most current devices are developed and used for rehabilitative purposes, more and more consumer devices are about to come on the market, and some stakeholders such as Elon Musk and his company Neuralink pursue more transhumanist objectives. This merging of minds and machines raises multiple intriguing philosophical, ethical, and legal questions: Do these devices become part of the person, even more, might the AI operating these devices become part of her? (I argue that it does under certain conditions, creating the most intimate conceivable connection between AI and persons). Are there ethical boundaries, and what is the legal situation, especially with respect to human rights? (I call for a renaissance of the right to freedom of thought to provide at least some principled protection for privacy of thought). Moreover, the topic has received the attention of international organizations, which will negotiate the first international treaty on the ethics of neurotechnology under the auspices of UNESCO in the beginning of 2025 (expected to be concluded in late 2025). This will set the standards for the future trajectory of the technology, but whether agreement can be found is to be seen. The EU, US, and China have different regulatory approaches with different visions for the future. This talk addresses these political, philosophical, legal and ethical questions and presents results of an international research cooperation on the topic, HYBRID MIND, that is funded in Germany by the Federal Ministry of Education and Research and comes to its official conclusion during the days of the 38C3. Licensed to the public under http://creativecommons.org/licenses/by/4.0 about this event: https://events.ccc.de/congress/2024/hub/event/philosophical-ethical-and-legal-aspects-of-brain-computer-interfaces/

Ever wondered what data is stored inside DB print-at-home train tickets or those in your local transport association's app? Join me for the deep dive into digital railway ticketing you didn't know you needed. After getting my shiny new Deutschlandsemesterticket from University I was so annoyed with the quality of the SaarVV app that I set out to put my train tickets into Apple Wallet - whether the train companies wanted me to or not. What followed was several weeks of banging my head against the wall and googling various terms with "filetype:pdf" until I understood how they're encoded. This talk is a highly condensed executive summary of the most interesting parts of that journey - from the surprising to the downright weird. Finally, I'll cover how you can issue your own train tickets - for fun and absolutely no profit! Licensed to the public under http://creativecommons.org/licenses/by/4.0 about this event: https://events.ccc.de/congress/2024/hub/event/what-s-inside-my-train-ticket/