Roee Avrahami

Kubernetes 1.31 Patch Cadence Raises End-of-Life Timing Concerns Trend: With 1.31.13 released and EOL approaching, operators must plan patch windows before Oct 28, 2025. Why it matters: Running EOL code risks security and compliance problems for production clusters. Question: Who owns your cluster EOL calendar, platform team, SRE, or a central governance group? 🔁 Repost if cluster EOL dates are on your Q4 checklist 🔔 Follow me for tooling and governance that reduce upgrade risk 🌟 Takeaway: EOL dates are operational deadlines; treat them like financial close dates

2

2 Comments

etcd in Kubernetes 💾 etcd is a strongly consistent, distributed key-value store that serves as the single source of truth for your entire cluster. Every pod creation, secret, and configuration change is recorded here. ➙ How does etcd work in Kubernetes? > Raft Consensus: It utilizes the Raft algorithm to replicate data across multiple nodes, ensuring immediate consistency . > Fault Tolerance: If a node fails, the system continues to function without data loss. > State Agreement: The remaining nodes communicate to agree on the current state, ensuring the cluster always has an accurate source of truth. ➙ Why is it essential in Kubernetes? > Exclusivity: It is the only stateful component in the control plane. > Resilience: Its distributed nature guarantees high availability. > Disaster Recovery: It contains the complete blueprint of your infrastructure, making it the critical component for restoration. Treat etcd with care. Isolate it from worker nodes to prevent resource contention and automate your snapshots. Losing etcd means losing the cluster.

12

🔐 Kubernetes Security Is Not a Single Layer, It’s a System of Systems This diagram might look overwhelming at first glance, but that’s the reality of securing a modern Kubernetes environment. From RBAC and admission control to service mesh policies, eBPF enforcement, logging, image scanning, and network egress control — every layer matters. Why? Because in Kubernetes: ✅ A single misconfigured RoleBinding can expose secrets. ✅ An overlooked network policy can open doors for lateral movement. ✅ Missing egress controls can let attackers slip data out unnoticed. This isn’t just complexity — it’s defense in depth. If you’re working on Kubernetes security, here are 3 key takeaways from this model: 1️⃣ Think holistically – Secure workloads, clusters, networks, and identities together. 2️⃣ Automate checks – Policy enforcement, scanning, and compliance should run continuously. 3️⃣ Shift left – Security starts with developers, not just ops or security teams. Kubernetes is powerful, but with great power comes the responsibility to design security as a first-class citizen — not an afterthought. Would love to hear: 👉 What’s the biggest Kubernetes security challenge you’ve faced? #Kubernetes #Security #CloudNative #DevSecOps #PlatformEngineering #ZeroTrust

1,650

32 Comments

Kubernetes NodeSwap is nearing stable status in v1.34, enabling swap usage on Linux nodes to improve resource utilization and reduce OOM kills, but careful tuning of Linux kernel parameters is crucial for performance and stability. This blog post dives into critical kernel parameters, test results, and optimal settings for stable, high-performing Kubernetes clusters. #kubernetes #nodeswap #linux #kernel #performance #tuning #oom #memory #resourceutilization #this_post_was_generated_with_ai_assistance #responsibleai https://lnkd.in/ead3RP2S

3