Philippe Wanner

As an SRE, I’ve seen #Terraform codebases scale from a single file used by 10 developers... to complex, multi-cloud infrastructures powering thousands. At every step, the layout of the infrastructure-as-code repo had to evolve. And no matter how projects started, they always ended up adopting a familiar pattern—one that made collaboration safer, state management simpler, and automation easier to manage. In my latest post, I break down: - Why starting with a monolith is fine (for a while) - How to scale with application-based folders and modular state - Why a single Terraform state bucket is safer than it sounds - When it’s time to split infrastructure code across repos ... and the real trade-offs behind each decision. Whether you're still working on main.tf or dealing with dozens of repos applying infrastructure across clouds, this one's for you 👇 Link: https://lnkd.in/dmyZjTSY --- If you liked this post: 🔔 Follow Teodor Podobnik ♻ Repost to help others find it 💾 Save this post for future reference #DevOps #SRE #Infrastructure #Cloud #GCP #AWS #Azure #Learn

77

3 Kommentare

Is Kubernetes Losing Its Shine? Many companies are stepping back from Kubernetes—not because it isn't powerful, but because it might be too powerful for their current stage. From rising complexity to operational overhead, businesses are realizing that sometimes, simpler is smarter. Whether it's moving to PaaS solutions or leveraging serverless architecture, the goal remains the same: velocity without the weight. Check out this insightful article that breaks down the "why" behind this shift: https://lnkd.in/gHMVDxNd What’s your take? Are we overengineering too early in the product lifecycle? #Kubernetes #DevOps #CloudComputing #PlatformEngineering #Serverless #TechTrends #EngineeringExcellence #SaaS #StartupTech #ProductivityOverComplexity #InfrastructureSimplified

43

1 Kommentar

New release: GraalVM Reachability Metadata v0.3.19!🎉 This repository enables users of GraalVM Native Image to share and reuse metadata for libraries and frameworks in the Java ecosystem. This release extends the configuration for Kafka, specifically for Kafka Streams 3.9.0 and Kafka Clients. Access to the Reachability Metadata Repository is enabled by default if you are using GraalVM Native Build Tools — get started: https://lnkd.in/gA2HjEYt See more info at https://lnkd.in/g4_rqu4d. #GraalVM #Java #OpenSource

39

1 Kommentar

👉 Ever thought a free-forever Kubernetes cluster in the cloud was too good to be true? Well, it's not! It's actually perfect for your homelab projects. My colleague, Ulli Goschler, wrote a great handbook on how to set up a Kubernetes cluster using Oracle Cloud's Always Free services. This free tier gives you a lot to work with: * 4 ARM64 CPU cores * 24GB of RAM * 200GB of disk space I've been using this offering for years and it works really well. The handbook uses Terraform to make the setup fast and easy. In the end, you get a managed Kubernetes cluster (OKE) with two nodes. To set up ingress, you may need your own domain name. The handbook also shows you how to use Let’s Encrypt for free SSL certificates and Cloudflare for DNS to keep everything completely free. Ulli constantly updates the handbook, so it always has the latest information, including notes for OKE upgrades. Check out the project here: https://lnkd.in/dbZwattm If you find it useful, please leave a star on the project to show your support! Kubernetes is not yours? You can still benefit from the offerings by setting up one big Linux VM, just like I do.

263

18 Kommentare

NEW! Amazon S3 Tables - Now available in Frankfurt, Ireland and Stockholm AWS Regions. S3 Tables deliver the first cloud object store with built-in Apache Iceberg support, and the easiest way to store tabular data at scale. https://lnkd.in/d7jEDfHY

46

During my time at #AsyncAPIdays in Munich this week, I brought up an important topic that’s on everyone’s mind: the security of your OpenTelemetry Collector. This article from the official #OpenTelemetry page dives into a critical enhancement you can make—implementing mTLS (mutual TLS) in your Kubernetes Gateway API for better security. Check it out here: https://lnkd.in/dJmzV_wN Exposing OTel Collector in Kubernetes with Gateway API & mTLS | OpenTelemetry #asyncapi #eventdriven #observability #security #mtls #kubernetes

21

3 Kommentare

A significant advance in k8s reliability, https://lnkd.in/dJ6cqSkt. Too often users/applications retrieve very large lists from the API server which cause the cluster to become destabilized because of insufficient memory. This feature is designed to remedy that. Utilizing stream encoding, the API will now transmit each item individually, allowing memory to be freed progressively as frame or chunk is transmitted.

29

What infrastructure tools out there offer a model that doesn't lead to "Snowflakes as Code"? Most deployment tooling seems to assume that any infrastructure code project is only ever deployed to a single environment, like infra CD tools that only support "plan" and "apply" stages, or those which are based on pull requests to deploy to, again, a single environment. Most software teams' path to production involves multiple environments. Some teams maintain a separate copy of infrastructure code for each environment, in a branch or folder, and copy or merge changes back and forth, hopefully remembering to manually change the things in the code that are different between each environment. That's not what we do with software code, though, is it? I know there are some tools out there that support workflows for using a single infrastructure code project across environments, with parameters to manage necessary variations like resource sizing. Some of these tools use Terraform / OpenTofu modules as the code, with a separate project for each environment, which can work well but feels like it's working around the lack of support for a reusable, deployable component model. Other tools provide a way to package a deployable infrastructure component the same way you would with a software package. What do people use? (I've got a post on my site where I define "snowflakes as code": https://lnkd.in/eivxB4ya)

39

13 Kommentare

Reusability in IaC – The art of module development in Terraform 🚀 Just published!  (from our #Azure Meetup Konstanz) If you missed our live session on Terraform Module Best Practices, the full recording is now available on YouTube! 🎥 🔗 Watch here: https://lnkd.in/eBis7n_V In this session, Stefan Rapp walk you through how to build clean, reusable, and scalable Terraform modules — based on real-world experience from customer projects. Whether you're just starting with Infrastructure as Code or looking to improve your current setup, this talk is packed with practical tips, examples, and tools to help you do it right from the start. 💡 What you’ll learn: ✅ Why modularization matters   ✅ How to structure and version modules properly   ✅ Common pitfalls (and how to avoid them)   ✅ Tools to improve module quality (Terraform Lint, Checkov, etc.)   ✅ How to use Terraform Cloud & registries effectively 👥 With guest Julian Mooren If you're working with Terraform in Azure, AWS, or even on-prem — this one’s for you. Let me know what you think, and feel free to reach out with questions or feedback! #Terraform #DevOps #InfrastructureAsCode #IaC #CloudEngineering #Azure #AWS #TerraformModules #HashiCorp #Cloud #Automation #YouTube

17