Lightnews — Scholar-powered news

Reposted by LearnKube

Kubesploit @kubesploit.io · 6h

This case study describes how the author’s EKS cluster autoscaler broke after migrating to Amazon’s AL2023 image and how they resolved it by switching to IRSA (IAM Roles for Service Accounts) and adjusting permissions

➜ https://ku.bz/PzHb6bP62

1 1

LearnKube @learnkube.com · 7h

Reflector is a Kubernetes addon designed to monitor changes to resources (secrets and configmaps) and reflect changes to mirror resources in the same or other namespaces

➜ https://ku.bz/ZTXWN7Qbm

2 3

Reposted by LearnKube

Kube Events @kube.events · 10h

https://kube.events/image-gen/digest?ids=12066d98-936a-480a-9be1-aefddc6b5fa1,dbb8c1e3-f77e-4ed0-a012-0b71bd33b5a7,08af35e5-f3d3-420b-968b-154dcdc98720,3bc4aec9-f7d6-4050-b11a-fdd092e8c345,f052bca1-11cd-43fc-a358-687045d466c3,e27faeab-8860-400f-a820-29ccb30d948b,cf484330-29a5-4877-a2e3-ff6e264f677e

1 1

Reposted by LearnKube

KubeFM @kube.fm · 13h

The Making of Flux Ep 4: The Platform Builders 🏗️

GitLab, Microsoft & Mirantis reveal why they're embedding Flux into their platforms, from Azure Arc's managed service to Cordant's multi-cluster vision

https://ku.bz/tVqKwNYQH

🌟 ControlPlane
🎙 🎙Bart

7 9

LearnKube @learnkube.com · 1d

This article explains how to compare the data plane resource consumption of Linkerd, Istio Legacy, and Istio Ambient using a benchmark test on GKE

➤ https://ku.bz/vJWcQchQn

https://linkerd.io/2025/05/21/behind-the-great-sidecar-debate/all-three-CPU.png

LearnKube @learnkube.com · 2d

Predictive Horizontal Pod Autoscalers (PHPAs) are Horizontal Pod Autoscalers (HPAs) with extra predictive capabilities baked in, allowing you to apply statistical models to the results of HPA calculations to make proactive scaling decisions

➤ https://ku.bz/hCqCLW8Yc

Reposted by LearnKube

Kube Builders @kube.builders · 2d

kubectl node-shell is a krew plugin that lets start a root shell in the node's host

➤ https://ku.bz/ZXkDtpn5g

1 1

Reposted by LearnKube

Kube Architect @kube.archi · 3d

This case study shows how to build architecture for scalable ML pipeline on Azure using Kubernetes and KEDA to generate inputs with 500 pods and deploy models via Azure ML

➤ https://ku.bz/0lYz58fTX

1 1

LearnKube @learnkube.com · 3d

This article explains how intra-node pod-to-pod traffic is handled when using Kindnet with the ptp CNI plugin

➤ https://ku.bz/HVprZ4nM2

LearnKube @learnkube.com · 3d

This article compares kube-proxy’s user-space model with Cilium’s eBPF-powered kernel routing, demonstrating how Cilium offers lower latency and better scalability

➜ https://ku.bz/-FxL6P4nW

Reposted by LearnKube

Kube Architect @kube.archi · 3d

kgateway is a Kubernetes-native, Envoy-based API gateway that routes, secures, and governs billions of API and AI requests across clouds

➤ https://ku.bz/WmWGQ11CZ

1 2

LearnKube @learnkube.com · 4d

k8s-csi-s3 is a CSI driver for S3 (or S3 compatible) storage

This tool can dynamically allocate buckets and mount them via a fuse mount into any container

➤ https://ku.bz/llM5w4dp9

1

LearnKube @learnkube.com · 4d

This tool enables automation-friendly, label-aware port forwarding for pods and services using the Kubernetes API, without relying on `kubectl port-forward`

➜ https://ku.bz/YtX6jdDB8

Reposted by LearnKube

KubeFM @kube.fm · 4d

Go's memory management can clash with Kubernetes resource limits

Emin Laletović explains this complex relationship and highlights potential OOM errors

Watch: https://ku.bz/7fnF-tJ8R

1 1

Reposted by LearnKube

Kube Architect @kube.archi · 4d

This article walks through real challenges the team faced while scaling EKS — from autoscaling, ingress, DNS tuning to graceful shutdowns — and shares how they tackled each layer

➜ https://ku.bz/G5hH_k9BP

1 1

Reposted by LearnKube

Kubesploit @kubesploit.io · 5d

kps-zeroexposure is a helm chart that fixes unhealthy or missing control-plane metrics targets in `kube-prometheus-stack` by deploying a secure Prometheus Agent as a DaemonSet

➤ https://ku.bz/jtT5DjB6h

https://github.com/adrghph/kps-zeroexposure/raw/master/images/schema.jpg

1 1

LearnKube @learnkube.com · 5d

In this case study, you will learn how the SRE team at Kapital Bank meticulously analyzed network kernel issues on a Kubernetes-based Memcached platform, uncovering and resolving intermittent performance problems through kernel parameter tuning

➜ https://ku.bz/spNnnpsM-

Reposted by LearnKube

Kube Builders @kube.builders · 5d

This case study shows how to implement a multi-cluster reconciler to manage Kubernetes resources across sharded clusters for fault tolerance

It covers sharding stateless workloads across 3 clusters to limit the impact of infrastructure failures

➜ https://ku.bz/1HTWb0GLC

1 1

LearnKube @learnkube.com · 5d

This week on the Learn Kubernetes Weekly:

🌍 Envs vs Clusters
🧩 Image Compatibility
🔁 Terraform → Crossplane
📊 Kube-State-Metrics
⚙️ Local Dev on K8s

⭐️ Testkube

Read it now: https://kube.today/issues/153

https://res.cloudinary.com/learnk8s/image/upload/v1760446374/linkedin-153_u2o697.png

7 8

LearnKube @learnkube.com · 6d

This operator automates provisioning and operating Redis in standalone, cluster, replication, or sentinel mode on Kubernetes with support for TLS, monitoring (via Redis Exporter), dynamic PVCs, and failover management

➜ https://ku.bz/Cx0XD_m0Q

https://github.com/OT-CONTAINER-KIT/redis-operator/raw/main/static/redis-operator-architecture.png

2

Reposted by LearnKube

KubeFM @kube.fm · 6d

🗣️ Niels Claeys shares how his team built a data platform processing up to 1.5 million core hours monthly and the specific optimizations they discovered: from scheduler changes to spot instance usage

https://ku.bz/hGRfkzDJW

🌟 Testkube
🎙 🎙Bart

6 9

LearnKube @learnkube.com · 7d

Kubernetes-MCP is a server that lets AI tools or clients connect to Kubernetes clusters in read-only mode

It can show resources, CRDs, logs, and events across multiple clusters but cannot make any changes

➜ https://ku.bz/Zj4fsrrnN

Reposted by LearnKube

Kube Events @kube.events · 7d

https://kube.events/image-gen/digest?ids=c339c1b2-c5c4-4e42-8d4f-12a02e703b4e,ed92baa6-b2a3-4a78-8f61-284f02b2b12a,6c5428e0-e833-428d-9e5d-d8060bd0235f,ccadcf86-3537-4f3d-b0d6-45d4e87d635f,996f6deb-039f-4b4c-81ad-d2959032d32e,023ceb58-df6b-4e1e-80b7-78b7483f83d3,0da294f0-fbf6-461d-95e7-7d6c410d50d7

1 1 1

Reposted by LearnKube

Kube Careers @kube.careers · 7d

🍽️ Platform Engineering eating DevOps' lunch?

We analyzed 509 Kubernetes jobs in Q3 2025:

💰 Avg salary: $177,983 (NA) / €92,113 (EU)
🏠 67% remote… but only 0.29% truly location-free

Full report: https://kube.careers/state-of-kubernetes-jobs-2025-q3

⭐️ LearnKube

https://res.cloudinary.com/learnk8s/image/upload/v1760333097/preview-2025q3_ugarmg.png

6 8

Reposted by LearnKube

KubeFM @kube.fm · 7d

🗣️ The Making of Flux Episode 3 is here!

From hitting the "scaling wall" to operational excellence—Philippe (Orange) and Arnab (Nomura) share how they transformed their Kubernetes operations with GitOps

https://ku.bz/tWcHlJm7M

🌟 ControlPlane
🎙 🎙Bart

6 8