Older posts.
Kubernetes Cost Optimization in 2026: 8% CPU Use and What Fixes It
Kubernetes cost optimisation as of October 2026: Cast AI measures 8% CPU and 20% memory utilisation, Datadog puts 83% of container cost on idle, plus OpenCost vs IBM Kubecost 3.0, Karpenter 1.15 Balanced consolidation, in-place resize, spot and Graviton.
12 minOct 20, 2026Kubernetes Autoscaling Explained (2026): HPA, VPA, KEDA and Karpenter
Kubernetes autoscaling in 2026, explained: HPA v2 behavior and the new tolerance field, HPA scale to zero in 1.37, VPA 1.8 in-place resize, KEDA 2.21, Cluster Autoscaler vs Karpenter 1.15, GKE and AKS node auto-provisioning, and the mistakes that cost money.
13 minOct 20, 2026Is Kubernetes Still Worth It in 2026? v1.37, 82% in Production, Real Costs
Is Kubernetes worth it in 2026? What v1.34 to v1.37 shipped, what the CNCF survey's 82% production figure means, what EKS Auto Mode, GKE Autopilot and AKS Automatic cost, when ECS, Cloud Run, Nomad or Fly fit better, and a decision framework for small teams.
12 minOct 20, 2026Ingress NGINX Is Retired: What Replaces It and How to Choose in 2026
Ingress NGINX stopped receiving fixes in March 2026 after seven CVEs in its final seven weeks. What retirement means, why IngressNightmare mattered, where Gateway API stands at v1.6, and how to pick between Envoy Gateway, Cilium, Istio, NGINX Gateway Fabric, Traefik and kgateway.
13 minOct 20, 2026Incident Management in 2026: Opsgenie Ends April 2027, AI SREs Arrive
Incident management in 2026: Opsgenie shuts down on 5 April 2027, FireHydrant now belongs to Freshworks, Grafana OnCall OSS is archived, PagerDuty and incident.io sell AI SRE agents, and the on-call data says learning from incidents is still the weak link.
13 minOct 20, 2026How to Set Up Grafana Loki on Kubernetes with Alloy (2026)
Step-by-step Grafana Loki 3.7 setup on Kubernetes: the new grafana-community Helm chart, deployment modes, S3 storage, Grafana Alloy replacing end-of-life Promtail, labels vs structured metadata, LogQL, retention, and fixes for 429 and too many outstanding requests.
12 minOct 20, 2026Argo CD vs Flux in 2026: GitOps After Argo CD 3.5 and Flux 2.9
Argo CD vs Flux in 2026: what changed in Argo CD 3.0 to 3.5 and Flux 2.6 to 2.9, who maintains Flux after Weaveworks, what the CNCF surveys say about adoption, how each handles multi-cluster and secrets, and where GitOps costs more than it saves.
13 minOct 20, 2026How to Speed Up GitHub Actions CI: Caching, Sharding, Docker (2026)
A step-by-step guide to faster GitHub Actions in 2026: measure job timings, cancel stale runs, use setup-node, setup-python and setup-go caching, actions/cache v6, buildx type=gha layer caching, matrix test sharding and larger runners, plus a cache-miss troubleshooting table.
13 minOct 20, 2026GitHub Actions Security in 2026: 76 Hijacked Tags and the Fixes That Work
GitHub Actions security in 2026: what tj-actions, trivy-action, TanStack and issues-helper taught us about mutable tags and pull_request_target, plus SHA pinning policy, workflow lockfiles, immutable releases, attestations, zizmor and the new runner prices.
13 minOct 20, 2026How to Use GitHub Actions OIDC With AWS, GCP and Azure (2026)
Step-by-step GitHub Actions OIDC setup for AWS, Google Cloud and Azure with no stored secrets: trust policies pinned to sub claims, immutable subjects, environments, pinned actions, and fixes for sts:AssumeRoleWithWebIdentity, audience and AADSTS700213 errors.
14 minOct 20, 2026How to Fix Kubernetes CrashLoopBackOff: Causes and Commands (2026)
Fix Kubernetes CrashLoopBackOff step by step: what the state means, the 10s to 5 minute backoff and the new per-node max in 1.35+, kubectl describe and logs --previous, exit codes 1, 127, 137 and 143, probe and OOM causes, and kubectl debug for crashing containers.
13 minOct 20, 2026How to Fix 502 Bad Gateway and 504 Gateway Timeout Errors (2026)
Fix 502 Bad Gateway and 504 Gateway Timeout step by step: find which hop generated the error, read NGINX upstream errors, fix keepalive and idle timeout mismatches (ALB 60s, GCP 600s), Kubernetes readiness and rollout gaps, Gateway API timeouts and Cloudflare 520 to 526.
15 minOct 20, 2026Is Fine-Tuning LLMs Worth It in 2026? Prices, LoRA, RFT and OpenAI's Exit
When fine-tuning an LLM beats prompting and RAG in 2026: OpenAI is winding down its fine-tuning API, Google and AWS added RL tuning, LoRA matches full fine-tuning when configured right, and a 5,000-example run costs $5 to $300 depending on where you train it.
14 minOct 20, 2026How to Fine-Tune an LLM with LoRA Using Unsloth or TRL (2026)
Step-by-step LoRA and QLoRA fine-tuning of an open-weight LLM with Unsloth 2026.10 or Hugging Face TRL 1.16: dataset format, hyperparameters, GPU memory, training script, evaluation against the base model, GGUF and vLLM export, and fixes for OOM and loss problems.
14 minOct 20, 2026How to Detect Runtime Threats with Falco on Kubernetes (2026)
Step-by-step Falco 0.45 setup on Kubernetes: Helm chart 9.2.0 with the modern eBPF driver, default and custom rules, Falcosidekick alerts to Slack, tuning noise, Falco Talon response actions, drop metrics, and a troubleshooting table for driver errors.
15 min