KubeCon EU 2025

Table of Contents

In this article, I’ll document notes from kubecon talks, I’ve attended and some interesting ideas or tools I found useful.

Day01

The First day was day of Lighting talks, so the log is pretty long, and each topic pretty short. Many talks were just a couple slides, even then lots of interesting ideas, links.

A Hitchhiker’s Guide to the CNCF Landscape- Katherine Druckman, Lori Lorusso

  • History of CNCF
  • Landscape important
  • how to use landscape

Random facts:

  • 208 CNCF Projects
  • 756 CNCF members
  • Linux foundation managed CNCF
  • An Interesting tool for the overall health of your project is CLO monitor.

What is Copa?

CNCF Sandbox project and CLI tool for directly patching container images using reports from vulnerability scanners.

Empowering Data Protection for Stateful Applications of k8s withkanister.io

Datadog real world Container Use Report 2022 (link) The DAM book Digital Asses Management for Photograpers (link) Install via helm kanister controller, 3 new CRs: BluePrints, ActionSets and…

Kubernetes SIG Storage report

  • Volume Populators in k8s 1.33
  • Goal: challenges: adding support for mew data sources cannot break existing behaviour
  • Always honor P±V reclaim policy

Prometheus v3

7 years stability 9900 commits (v2 - v3) 1021 vommits 367 uniq contributors (Deep Dive talk Prome) Native histograms

LitmusChaos

Litmus is an open source Chaos Engineering platform that enables teams to identify weaknesses & potential outages in infrastructures by inducing chaos tests in a controlled way.

Perses project update v0.51 in beta

Observability tools like Grafana. Also nn open specification for dashboards. The open dashboard tool for Prometheus and other data sources. New pluging system, more details in docs.

Kepler Project Updates by Sunanay IBM research

Kepler (Kubernetes-based Efficient Power Level Exporter) uses eBPF to probe energy related system stats and exports as Prometheus metrics.

CI/CD observability using OpenTelemetry

New SIG, Semantic Conventions Attris for: *CI/CD pipelienes cicd.pipelinename

  • deployments cicd.pipelinename.result
  • VCS
  • testings
  • …

Extend Large Lang Models Training beyond single k8s cluster

LLM challenges: not enough “resources” the scalability of a single cluster is not enough the resource utilization is not good enough frameworks did not talk with infra

Vitness is

MySQL-compatible, horizontally scalable, cloud-native database solution.

  • created by youtube (store video metadata)
  • migrated in google (borg, ex borg)
  • contributed to CNCF
  • sacale by sharding
  • cong sharding key
  • zero downtime sharding operations
  • vitess.io/slack
  • vitessio/vitess

Dapr by Marc Duiker

APIs for building secure and relable microservices Dapr Workflow - long running jobs llm conversation API Dapr agents (pytho)

flux by tamao nakahara

Open and extensible continuous delivery solution for Kubernetes. Powered by GitOps Toolkit. what do you need from CI/CD? GitOps & Progressive deliver automation/relability/security/scale Usually we use it but we don’t even know. Security Slam

Buildpacks

WHY? heroku

Strimzi

Operator for Apache Kafka Doing mostly what we do in dynatrace-operator Strimzicon June 4, 2025 (remote) @strimzi

Essension gRPC resources for Development Jung-Yu

grpci youtuebe channel grpc-io-announce mailing list gRPCConf 2025 (Aug 26 2025) Call for Speakers still open https://events.linuxfoundation.org/kubecon-cloudnativecon-europe/program/schedule/

OpenFGA

OpenFGA is a high performance and flexible authorization/permission system built for developers and inspired by Google Zanzibar

Authorization system for developers ReBAC is an evolution of RBAC and ABAC. Used in Google drive, Google Cloud etc Maintened by Okta and Grafana, how is being used? search github by “OpenFGA”

What Porter?

Porter enables you to package your application artifact, client tools, configuration and deployment logic together as an installer that you can distribute, and install with a single command.

Porter yaml example: k8s operator \ with mongoDB as storage

What isKonveyor AI?

more details in README.md

Project Capsule by auth

multi-tenancy in k8s is hard we mad it simple Manging multi-tenancy envs in k8s is complex, costly and often inefficient. (based on Operator, CRDS) looking for maintainers

Day02

First day with big keynotes and huge audience: pic from phone

Keynote 1 by Chrish Aniszczyk

  • 12000 attendees largest kubecon
  • Cloud Native: 10-years old
  • History of KubeCon 2016 London (first kubecon) Dec 2016 first board meeting, march 2016 also #cNTEcf meetups this year
  • Golden kubeastrounat includes not only k8s certification but others like istio, argo
  • New certifications: CNAP/CNPE.
  • Certified Open source Developer Export(CODE).
  • kcp project (check)
  • Future kubecons (2026-2027):
  • KbeCon Ch, KubeCon India, KubeCon Japan, KubeCon North America
  • Next year → 2026 KubeCon EU in Amsterdam, LA kubeCon North America 2026, 2027 Barcelona, Spain 2027

Keynote 2: Christine honey comp

  • LLm != like APIs we know and love
  • Normal APIs against LLMs
Normal APILLMS
unit testcan cscope
reproducible

Keynote 3: Dynamic scheduling every layer

  • SIG Node
  • SIG Scheduling
  • SIG Multicluster

Keynote 5: Evolving the k8s UX

  • In-cluster web portal k8s dashboard++
  • unified management UI
  • k8s desktop experience (local) to bring more users
  • when talk is published, there was a slide with demostrating how windows UI brough 10mln users compare to windows CMD old school interface.

Headlamp project sanbox project, now in SIG-UI really nice one!

Keynote 6: Rust in the linux Kernel: a new era for cloud native performance and security by Greg kroah-hartman

  • C and Rust and Linux
  • what Linux is?
  • 4992 devs

Linux Developer Community

  • 76,496 changes accepted
  • 8.7 changes per hour
  • 40 changes per day in stable trees
  • 13 CVEs assigned per day
  • New release every 8-9 weeks

Rust can make reviewing code easier

  • Enforce validation of untrusted data
  • Enforce memory lifetime rules
  • Enforce locking rules
  • Enforce error handling
  • Enforce type safety

This talk was perfect, highly recommended to rematch!

Observability

OpenTelemetry

  • instrument once used everywhere
  • separate telemetry generation from analysis
  • make software bservable by default
  • improve how we use telemetry
  • OpenTelemetry Operator
  • Perses—Open dashboard cncf project (also operator-based)
  • node collector (based on Ebpf)

Standardizing Ci/CI observability wtih Otel: Insights from the ci/cd observability SIG

Great talk and demos!

TODO: will add link when it’s published

Cert manager Contrib fest

https://github.com/cert-manager/community/blob/main/GOVERNANCE.md https://docs.google.com/spreadsheets/d/1zThfUB22HHdHAiRvS3ctbj4Da7j30imnUleURjxTYE0/edit?gid=421492367#gid=421492367 I’ve created first PR, and will continue after kubecon. https://github.com/cert-manager/trust-manager/issues/457 https://github.com/cert-manager/cmctl?tab=readme-ov-file

Open Telemetry Contrib fest

https://github.com/open-telemetry/opentelemetry-go/issues/6524 https://github.com/cert-manager/csi-driver/issues/171 https://github.com/cert-manager/makefile-modules/issues/154

Kubernetes CRD Design for the Long Haul: Tips, Tricks, and Lessons Learned

by Christian Schlotter, Broadcom & Fabrizio Pandini, VMware by Broadcom:

  • avoiding making mistakes to at will force you
  • K8s API linter (KAL)
  • Read API types top down
  • Define a glossary

great talk. slides

Day03

Keynotes

It was hard to distingvish or find useful links/ideas from few leynotes, so I just put it all together. It is a big talent to talk 40 minutes about nothing and everything without clear ideas or thoughts.

  • Peptone
  • Backstage https://github.com/backstage/backstage
  • apple intelligence: gRPC
  • https://gitjobs.dev/
  • Podcast secret source (open source podcast )
  • Cloud native AI group
  • CNCF at Ten: “Next Chapter in Native Cloud Computing” first talk k8s" and “Introducing the foundation”
  • instead of trends find gaps
  • MCP (agent and tools)
  • Orelly book AI Gateways in enterprice
  • kgateway

Talk by Commonly used query language

CNCF Observability Tag is working on research around an open observability query semantic definition. Google is working on OSS SQL extensions with pipe syntax and timesirees operators to well support observability quieries and share with the community.

Beyond Limits: Scaling k8s controllers horizontally

Problem: Controllers must prevent conflicts

  • Perform leader election
  • Only a single active instance
  • Controllers are not horizontally scalable
  • Limits large-scale use cases
  • No standard solution exists

Design – Key Feature

  • Sharding mechanisms inspired by distributed databases
  • Dynamic membership and failure detection
  • Automatic failover and rebalancing
  • Transparent label-based coordination
  • Prevents concurrent reconciliations
  • Reusable implementation

Contribfest about Inspektor Gadget

https://github.com/inspektor-gadget/Contribfest-KubeCon-Europe2025/blob/main/labs/collecting-metrics-with-ig/README.md https://github.com/inspektor-gadget/inspektor-gadget/issues/1650

ContribFest KubeStellar

https://clotributor.dev/search

Day 4

ContribFest pipeCD

A GitOps style continuous delivery platform that provides consistent deployment and operations experience for any applications

Flight home ✈️