# kubernetes.recipes

> Kubernetes recipes and guides — 1111 articles covering deployments, networking, security, AI/GPU, troubleshooting, and more.

## Recipes

- [How to Implement A/B Testing with Kubernetes](https://kubernetes.recipes/recipes/deployments/ab-testing-kubernetes/): Route traffic between application versions for A/B testing. Use service mesh, ingress, and custom routing rules to validate features with real users.
- [Access Zones on Scale-Out NAS for Kubernetes](https://kubernetes.recipes/recipes/storage/access-zones-scaleout-nas-kubernetes/): Configure access zones on scale-out NAS (Dell PowerScale/Isilon) for Kubernetes persistent storage. Multi-tenant isolation, CSI driver setup.
- [Configure SR-IOV agent-config.yaml Device by Path](https://kubernetes.recipes/recipes/networking/agent-config-device-by-path/): Use agent-config.yaml to select network devices by PCI path for SR-IOV VF creation, ensuring deterministic NIC targeting across OpenShift nodes.
- [Agentic AI and Multi-Agent Systems](https://kubernetes.recipes/recipes/ai/agentic-ai-multiagent-kubernetes/): Deploy autonomous AI agents and multi-agent orchestration on Kubernetes. LangGraph, CrewAI, AutoGen, tool-calling agents, agent-to-agent communication.
- [Volcano Job minAvailable Gang Scheduling](https://kubernetes.recipes/recipes/ai/ai-batch-processing-volcano/): Configure Volcano job minAvailable for gang scheduling on Kubernetes. Batch AI training, fair-share queues, job plugins, and GPU preemption guide.
- [AI Content Watermarking on Kubernetes](https://kubernetes.recipes/recipes/ai/ai-content-watermarking-kubernetes/): Deploy AI-generated content watermarking on Kubernetes. Invisible watermarks, SynthID integration, detection APIs, image and text watermarking pipelines.
- [AI Infrastructure Cost Optimization](https://kubernetes.recipes/recipes/ai/ai-infrastructure-cost-optimization-kubernetes/): Optimize AI infrastructure costs on Kubernetes. GPU sharing, spot instances, inference batching, model quantization, token economics.
- [AI ML Security and Compliance Kubernetes](https://kubernetes.recipes/recipes/security/ai-ml-security-compliance-kubernetes/): Secure AI and ML workloads on Kubernetes with model encryption, data governance, audit logging, network isolation for training jobs.
- [AI-Native Development Platforms on Kubernetes](https://kubernetes.recipes/recipes/ai/ai-native-development-kubernetes/): Run AI coding assistants, AI review bots, automated test generation, and AI-driven CI/CD as private Kubernetes workloads.
- [AI Resource Allocation Optimization](https://kubernetes.recipes/recipes/ai/ai-resource-allocation-optimization-kubernetes/): Optimize GPU and memory allocation for AI workloads on Kubernetes. Right-size GPU requests, bin-packing strategies, gang scheduling.
- [AI Security Platforms on Kubernetes](https://kubernetes.recipes/recipes/security/ai-security-platform-kubernetes/): Secure AI workloads on Kubernetes. Model supply chain security, prompt injection defense, LLM output filtering, AI RBAC, GPU isolation.
- [AI Supercomputing on Kubernetes GPU Clusters](https://kubernetes.recipes/recipes/ai/ai-supercomputing-kubernetes-gpu-clusters/): Build AI supercomputing platforms on Kubernetes. Multi-node GPU training, NVIDIA DGX SuperPOD, InfiniBand RDMA, NCCL tuning, Blackwell clusters.
- [AIPerf: Benchmark LLM Inference on Kubernetes](https://kubernetes.recipes/recipes/ai/aiperf-benchmark-llm-kubernetes/): Benchmark vLLM, NIM, Triton and SGLang endpoints on Kubernetes with NVIDIA AIPerf: TTFT, ITL, throughput, request rates, datasets, goodput and GPU telemetry.
- [AIPerf Concurrency Sweep on K8s](https://kubernetes.recipes/recipes/ai/aiperf-concurrency-sweep-kubernetes/): Run AIPerf concurrency sweeps on Kubernetes to find optimal LLM serving capacity. Automate 1-128 concurrent user benchmarks with batch Jobs.
- [AIPerf Goodput and SLO Benchmarks](https://kubernetes.recipes/recipes/ai/aiperf-goodput-slo-benchmark/): Measure LLM goodput with AIPerf on Kubernetes. Define SLOs for TTFT and ITL, calculate effective throughput, and benchmark with timeslice analysis.
- [AIPerf Multi-Model Benchmark on K8s](https://kubernetes.recipes/recipes/ai/aiperf-multi-model-benchmark/): Compare multiple LLM models and backends with AIPerf on Kubernetes. Benchmark vLLM vs TGI vs Triton with automated multi-run confidence intervals.
- [AIPerf Trace Replay Benchmarks on K8s](https://kubernetes.recipes/recipes/ai/aiperf-trace-replay-kubernetes/): Replay production traffic traces with AIPerf on Kubernetes. Use moon_cake format, ShareGPT datasets, and fixed schedules for realistic LLM benchmarks.
- [AIPerf Offline vLLM Benchmarking](https://kubernetes.recipes/recipes/ai/aiperf-vllm-benchmark-offline/): Benchmark vLLM inference with AIPerf in air-gapped Kubernetes clusters. Use dummy tokenizers, offline mode, custom endpoints.
- [Air-Gapped Image Import for OpenShift Clusters](https://kubernetes.recipes/recipes/deployments/air-gapped-image-import-openshift/): Import container images into disconnected OpenShift clusters. Use podman save/load and internal registries when DNS and TLS block external pulls.
- [Air-Gap OpenShift Upgrade oc-mirror OSUS](https://kubernetes.recipes/recipes/configuration/airgap-openshift-upgrade-oc-mirror-osus/): Upgrade air-gapped OpenShift with oc-mirror and OSUS. Mirror release payloads and Cincinnati graph, configure IDMS, and drive CVO upgrades.
- [Air-Gapped OpenShift with Quay Mirror](https://kubernetes.recipes/recipes/deployments/airgapped-openshift-quay-mirror/): Deploy OpenShift in air-gapped environments with local Quay registry mirror, ImageDigestMirrorSet, and custom CatalogSources.
- [Alertmanager on Kubernetes: Helm Setup and Config](https://kubernetes.recipes/recipes/observability/alertmanager-configuration/): Set up Alertmanager on Kubernetes with the kube-prometheus-stack Helm chart: routes, matchers, Slack/PagerDuty/email receivers, inhibition, silences, testing.
- [Automate Kubernetes Day-2 Operations with Ansible](https://kubernetes.recipes/recipes/deployments/ansible-kubernetes-day2-automation/): Use Ansible to automate Kubernetes day-2 operations — apply manifests, roll out upgrades, and reconcile cluster state with the kubernetes.core collection.
- [How to Configure Kubernetes API Access Control](https://kubernetes.recipes/recipes/security/api-access-control/): Set up secure API server access with authentication and authorization. Configure RBAC, API groups, and audit logging for cluster security.
- [Fix API Server Timeout and Overload](https://kubernetes.recipes/recipes/troubleshooting/api-server-timeout-troubleshooting/): Debug kubectl timeouts, API server overload, and connection refused errors. Covers etcd latency, webhook timeouts, and rate limiting.
- [ArgoCD App of Apps with Helm Values](https://kubernetes.recipes/recipes/helm/argocd-app-of-apps-helm-values/): Use the ArgoCD App of Apps pattern with Helm value overrides per environment, enabling templated Application manifests and DRY multi-environment configurations.
- [ArgoCD App of Apps Pattern Explained](https://kubernetes.recipes/recipes/deployments/argocd-app-of-apps-pattern/): Implement the ArgoCD App of Apps pattern to manage multiple applications from a parent Application for cluster bootstrapping.
- [ArgoCD App of Apps with Sync Waves](https://kubernetes.recipes/recipes/deployments/argocd-app-of-apps-sync-waves/): Combine the ArgoCD App of Apps pattern with sync waves to bootstrap entire clusters in dependency order, from CRDs and operators to application workloads.
- [ArgoCD ApplicationSets for Multi-Tenant GPUs](https://kubernetes.recipes/recipes/deployments/argocd-applicationsets-multi-tenant/): Use ArgoCD ApplicationSets to auto-discover and provision GPU tenant overlays from Git directories with per-tenant sync policies.
- [ArgoCD Declarative Application Setup](https://kubernetes.recipes/recipes/deployments/argocd-declarative-application-setup/): Define ArgoCD Applications, Projects, and repository credentials declaratively using Kubernetes manifests for reproducible GitOps configuration.
- [Argo CD GitOps: Install, Apps and ApplicationSets](https://kubernetes.recipes/recipes/deployments/argocd-gitops/): Argo CD GitOps on Kubernetes: install, Application and ApplicationSet YAML, Helm/Kustomize sources, sync waves, App of Apps, multi-cluster, RBAC, alerts.
- [ArgoCD Multi-Cluster App of Apps](https://kubernetes.recipes/recipes/deployments/argocd-multi-cluster-app-of-apps/): Manage multiple Kubernetes clusters with ArgoCD App of Apps, deploying shared infrastructure and cluster-specific workloads from a single GitOps repository.
- [Manage OperatorGroups with ArgoCD](https://kubernetes.recipes/recipes/deployments/argocd-operatorgroup-management/): Deploy and manage OLM OperatorGroup resources via ArgoCD for GitOps-driven operator lifecycle management in OpenShift namespaces.
- [ArgoCD PreSync and PostSync Hooks](https://kubernetes.recipes/recipes/deployments/argocd-presync-postsync-hooks/): Use ArgoCD PreSync hooks for database migrations and PostSync hooks for smoke tests, with SyncFail hooks for automated rollback and cleanup.
- [Detect ArgoCD Shadow Updates Out-of-Band](https://kubernetes.recipes/recipes/deployments/argocd-shadow-update-detection/): Detect and prevent ArgoCD shadow updates where manual kubectl changes bypass GitOps. Configure self-heal, sync, and drift detection.
- [ArgoCD Sync Waves for Canary Deployments](https://kubernetes.recipes/recipes/deployments/argocd-sync-waves-canary/): Use ArgoCD sync waves for canary deployments with Istio traffic splitting, automated validation, and progressive rollout strategies.
- [ArgoCD Sync Waves for CRD & Operator Ordering](https://kubernetes.recipes/recipes/deployments/argocd-sync-waves-crd-operators/): Use ArgoCD sync waves to deploy Custom Resource Definitions before operators and custom resources, preventing CRD race conditions in GitOps pipelines.
- [ArgoCD Sync Waves for Database Migrations](https://kubernetes.recipes/recipes/deployments/argocd-sync-waves-database-migration/): Use ArgoCD sync waves and PreSync hooks to run database migrations before deploying application code, with rollback strategies.
- [ArgoCD Sync Waves for Ordered Deployments](https://kubernetes.recipes/recipes/deployments/argocd-sync-waves-ordering/): Use ArgoCD sync waves to control the order of Kubernetes resource deployment, ensuring dependencies like namespaces and CRDs are created before workloads.
- [Automate NCCL Preflight Checks in CI/CD Pipelines](https://kubernetes.recipes/recipes/deployments/automate-nccl-preflight-ci/): Run NCCL smoke benchmarks automatically in CI/CD pipelines before promoting GPU cluster changes to production, catching regressions early.
- [Autonomous Industrial Systems on Kubernetes](https://kubernetes.recipes/recipes/ai/autonomous-industrial-systems-kubernetes/): Orchestrate autonomous factories and logistics with Kubernetes. Digital twins, robot fleet coordination, industrial IoT pipelines, predictive maintenance.
- [Backstage on Kubernetes: Deploy the Developer Portal](https://kubernetes.recipes/recipes/deployments/backstage-kubernetes-developer-portal/): Deploy Backstage on Kubernetes with Helm or your own image, PostgreSQL, catalog-info.yaml, software templates and the Kubernetes plugin with RBAC.
- [Build a K8s Operator with Docker Testing](https://kubernetes.recipes/recipes/deployments/build-kubernetes-operator-docker-testing/): Build a Kubernetes operator with Operator SDK and Kubebuilder. Test with Docker, Kind, and envtest. Full TDD workflow to OLM bundle.
- [Canary Deployment with Gateway API Traffic Splitting](https://kubernetes.recipes/recipes/deployments/canary-deployment-gateway-api-traffic-splitting/): Implement canary deployments using Kubernetes Gateway API HTTPRoute traffic splitting. Gradually shift traffic from stable to canary version with weight-based
- [How to Implement Canary Deployments](https://kubernetes.recipes/recipes/deployments/canary-deployments/): Learn to implement canary deployments in Kubernetes for gradual rollouts. Use native features and Ingress-based traffic splitting for safe releases.
- [cert-manager Cloudflare DNS01 K8s](https://kubernetes.recipes/recipes/networking/cert-manager-cloudflare-dns01-k8s/): Configure cert-manager with Cloudflare DNS01 challenge for wildcard TLS certificates on Kubernetes. API token secret, ClusterIssuer, and auto-renewal.
- [Fix Kubernetes Certificate Expiry Issues](https://kubernetes.recipes/recipes/security/certificate-expiry-troubleshooting/): Debug and renew expired Kubernetes certificates for API server, kubelet, and etcd. Covers kubeadm cert renewal, OpenShift auto-rotation, and monitoring expiry.
- [cert-manager OVH DNS-01 Wildcard TLS](https://kubernetes.recipes/recipes/security/certmanager-ovh-dns01-wildcard-tls/): Configure cert-manager with OVH DNS-01 challenge for automated wildcard TLS certificates on k3s. Let's Encrypt production certificates with zero downtime r.
- [Chaos Mesh Fault Injection on Kubernetes](https://kubernetes.recipes/recipes/troubleshooting/chaos-mesh-fault-injection-kubernetes/): Deploy Chaos Mesh for chaos engineering on Kubernetes. Covers PodChaos, NetworkChaos, IOChaos, StressChaos experiments, scheduling, RBAC
- [Check Bonding and Interface Status for SR-IOV](https://kubernetes.recipes/recipes/troubleshooting/check-bonding-and-interface-status/): Inspect bond membership, interface state, and link aggregation to confirm which NICs can be correctly targeted by SR-IOV network policies on Kubernetes.
- [Cilium ClusterMesh Multi-Cluster](https://kubernetes.recipes/recipes/networking/cilium-clustermesh-multicluster/): Connect multiple K8s clusters with Cilium ClusterMesh. Shared services, global service discovery, and cross-cluster network policies.
- [Cilium Debug Pod Troubleshooting](https://kubernetes.recipes/recipes/troubleshooting/cilium-debug-pod-troubleshooting/): Debug Kubernetes networking with Cilium debug pods and containers. cilium-dbg, netshoot, hubble observe, and endpoint connectivity troubleshooting.
- [Cilium eBPF Gateway API Hubble k3s](https://kubernetes.recipes/recipes/networking/cilium-ebpf-gateway-api-hubble-k3s/): Install Cilium with eBPF dataplane, Gateway API support, and Hubble observability on k3s. Replace kube-proxy with eBPF, configure GatewayClass.
- [Cilium Hubble Observability Guide](https://kubernetes.recipes/recipes/observability/cilium-hubble-observability-guide/): Monitor Kubernetes network flows with Cilium Hubble. CLI usage, Hubble UI, flow filtering, DNS visibility, and L7 HTTP observability.
- [Cilium Service Mesh: eBPF-Powered Kubernetes](https://kubernetes.recipes/recipes/networking/cilium-service-mesh-kubernetes/): Deploy Cilium service mesh on Kubernetes with eBPF. Sidecar-free mTLS, L7 traffic management, network policies, Hubble observability, and Gateway API support.
- [Cincinnati Graph OpenShift Upgrades](https://kubernetes.recipes/recipes/configuration/cincinnati-graph-openshift-upgrades/): Understand Cincinnati upgrade graph for OpenShift. Query graph endpoints, decode channels, blocked edges, conditional updates, and debug upgrade paths.
- [Cluster API on AWS: Provision EKS Clusters](https://kubernetes.recipes/recipes/deployments/cluster-api-aws-eks-management/): Use Cluster API (CAPI) to provision and manage EKS clusters declaratively. Install clusterctl, configure CAPA provider, and automate cluster lifecycle on AWS.
- [ClusterClass: Reusable Cluster API Templates](https://kubernetes.recipes/recipes/deployments/cluster-api-clusterclass-topology/): Define reusable ClusterClass templates in Cluster API for consistent multi-cluster provisioning. Variables, patches, and topology-based cluster creation.
- [Cluster API for K8s Lifecycle Management](https://kubernetes.recipes/recipes/deployments/cluster-api-infrastructure-as-code/): Manage Kubernetes cluster lifecycle with Cluster API. Declarative cluster creation, upgrades, scaling, and multi-cloud infrastructure provisioning as code.
- [Cluster API on vSphere: On-Prem K8s Clusters](https://kubernetes.recipes/recipes/deployments/cluster-api-vsphere-on-prem/): Provision on-premises Kubernetes clusters on vSphere using Cluster API (CAPV). VM templates, control plane HA, node scaling, and day-2 operations.
- [ClusterPolicy MOFED Upgrade Strategy](https://kubernetes.recipes/recipes/configuration/clusterpolicy-mofed-upgrade/): Configure safe MOFED driver upgrade policies in the NVIDIA GPU Operator ClusterPolicy with rolling updates, node draining, and rollback procedures.
- [CNCF AI Projects Landscape Kubernetes](https://kubernetes.recipes/recipes/ai/cnai-cncf-ai-projects-landscape/): A project-by-project tour of the CNCF AI ecosystem for Kubernetes, including Kubeflow, KServe, Katib, Feast, Volcano, Kueue, and KAITO.
- [CNPG Disaster Recovery and Replication](https://kubernetes.recipes/recipes/storage/cnpg-disaster-recovery/): Set up cross-region PostgreSQL disaster recovery with CloudNativePG using replica clusters, WAL shipping, and automated failover.
- [CloudNativePG (CNPG): PostgreSQL Operator on Kubernetes](https://kubernetes.recipes/recipes/deployments/cnpg-postgresql-operator/): Run HA PostgreSQL on Kubernetes with CloudNativePG: operator install, Cluster CR, anti-affinity, S3 WAL backups, PITR, PgBouncer Pooler and monitoring.
- [CNPG Cluster Scaling and Upgrades](https://kubernetes.recipes/recipes/deployments/cnpg-scaling-upgrades/): Scale CloudNativePG clusters, perform rolling PostgreSQL major upgrades, and manage storage expansion without downtime in Kubernetes.
- [Compare NCCL Intra-Node vs Inter-Node Speed](https://kubernetes.recipes/recipes/ai/compare-nccl-intra-inter-node/): Build a repeatable comparison between local and cross-node NCCL throughput to validate GPU cluster interconnect scaling and identify bottlenecks early.
- [Hardware Attestation for Kubernetes Workloads](https://kubernetes.recipes/recipes/security/confidential-computing-attestation-kubernetes/): Implement remote attestation for Kubernetes workloads. Verify TEE integrity with attestation services, release secrets to verified enclaves.
- [Confidential Computing on Kubernetes](https://kubernetes.recipes/recipes/security/confidential-computing-kubernetes/): Deploy confidential containers with encrypted memory using Intel SGX, AMD SEV-SNP, and Kata Containers. Protect data in use from even the cluster admin.
- [Confidential Computing: SGX and SEV-SNP](https://kubernetes.recipes/recipes/security/confidential-computing-sgx-sev-kubernetes/): Deploy confidential containers on Kubernetes with Intel SGX and AMD SEV-SNP. Encrypted memory, attestation, confidential VMs, Kata Containers.
- [Confidential Containers with Kata](https://kubernetes.recipes/recipes/security/confidential-containers-kata-kubernetes/): Deploy confidential containers using Kata Containers and TEEs on Kubernetes. Hardware attestation, encrypted container images.
- [Kubernetes ConfigMaps and Secrets Management](https://kubernetes.recipes/recipes/configuration/configmap-secrets-management/): Create, mount and update Kubernetes ConfigMaps and Secrets: env vs volume, immutable config, auto-reload with Reloader, RBAC, and encryption at rest.
- [Fix the Kubernetes ConfigMap Too Large Error](https://kubernetes.recipes/recipes/troubleshooting/configmap-too-large-error/): Resolve the 1MB ConfigMap size limit error. Split configs, use Secrets for binary data, mount volumes, or use external stores.
- [Configure ClusterPolicy kernelModuleType for GPU](https://kubernetes.recipes/recipes/configuration/configure-clusterpolicy-kernel-module-type/): Understand and configure the driver.kernelModuleType field in the NVIDIA GPU Operator ClusterPolicy to choose between auto, open, and proprietary kernel.
- [Configure GPUDirect RDMA with NVIDIA GPU Operator](https://kubernetes.recipes/recipes/networking/configure-gpudirect-rdma-gpu-operator/): Set up GPUDirect RDMA on Kubernetes using the NVIDIA GPU Operator with either DMA-BUF or legacy nvidia-peermem, including Network Operator integration.
- [Configure SriovNetwork with NVIDIA nv-ipam](https://kubernetes.recipes/recipes/networking/configure-sriovnetwork-nv-ipam/): Create a SriovNetwork resource that auto-generates a Multus NetworkAttachmentDefinition using nv-ipam for high-performance SR-IOV secondary interfaces.
- [Container Image Security Scanning on Kubernetes](https://kubernetes.recipes/recipes/security/container-image-security-scanning-kubernetes/): Implement container image security scanning in Kubernetes CI/CD pipelines. Trivy, Grype, and admission controllers to prevent vulnerable images from running.
- [Container Image Signing and Verification on Kubernetes](https://kubernetes.recipes/recipes/security/container-image-signing-verification-kubernetes/): Sign container images with Sigstore cosign and verify signatures at admission time with Kyverno or Connaisseur. Supply chain security for Kubernetes
- [How to Implement Container Logging Patterns](https://kubernetes.recipes/recipes/observability/container-logging-patterns/): Configure logging for Kubernetes applications. Implement sidecar logging, log aggregation, and structured logging best practices.
- [containerd certs.d Registry CA Trust](https://kubernetes.recipes/recipes/configuration/containerd-certs-d-registry-ca-trust/): Configure containerd to trust private registry CAs using /etc/containerd/certs.d. Set up hosts.toml for custom CA certificates and mirror registries.
- [Continuous Batching LLM Inference K8s](https://kubernetes.recipes/recipes/ai/continuous-batching-llm-kubernetes/): Configure continuous batching for LLM inference on Kubernetes. vLLM and TRT-LLM batch scheduling, max-num-seqs tuning, and throughput optimization.
- [Convert Connected to Disconnected OCP](https://kubernetes.recipes/recipes/deployments/convert-connected-disconnected-openshift/): Convert a connected OpenShift cluster to disconnected. Mirror images, configure IDMS, update pull secrets, fix Insights Operator, and verify applications.
- [Copy NVIDIA NIM Images to Internal Quay Registry](https://kubernetes.recipes/recipes/ai/copy-nim-image-internal-quay-registry/): Pull NIM container images from nvcr.io and push to an internal Quay registry. Covers authentication, tagging, air-gapped workflows, and curl token issues.
- [How to Configure Kubernetes Cluster DNS](https://kubernetes.recipes/recipes/networking/coredns-configuration/): Customize CoreDNS configuration for your cluster. Add custom DNS entries, configure forwarding, and optimize DNS resolution.
- [CoreDNS Troubleshooting: Fix Kubernetes DNS](https://kubernetes.recipes/recipes/troubleshooting/coredns-troubleshooting/): Fix CoreDNS resolution failures in Kubernetes: NXDOMAIN, SERVFAIL, i/o timeouts, loop detected, ndots latency, and NetworkPolicy blocking port 53.
- [How to Fix CrashLoopBackOff in Kubernetes](https://kubernetes.recipes/recipes/troubleshooting/crashloopbackoff-troubleshooting/): Fix CrashLoopBackOff in Kubernetes with step-by-step troubleshooting. Debug OOMKilled, failed probes, missing configs, and image errors causing pod crash loops.
- [Create an NVIDIA nv-ipam IPPool for SR-IOV](https://kubernetes.recipes/recipes/networking/create-nv-ipam-ippool/): Define a valid nv-ipam IPPool and node-aware sizing strategy so SR-IOV workloads can reliably obtain secondary interface IP addresses on Kubernetes.
- [Debug CRI-O Container Runtime Errors](https://kubernetes.recipes/recipes/troubleshooting/crio-container-runtime-errors/): Troubleshoot CRI-O issues on OpenShift nodes. Fix image pull failures, container start errors, storage driver problems, and CNI networking plugin failures.
- [Kubernetes CronJob concurrencyPolicy: Forbid, Replace](https://kubernetes.recipes/recipes/deployments/cronjob-concurrency-policy/): Kubernetes CronJob concurrencyPolicy explained: Allow vs Forbid vs Replace, how Forbid interacts with startingDeadlineSeconds and activeDeadlineSeconds.
- [crun vs runc: Memory and Performance Benchmarks 2026](https://kubernetes.recipes/recipes/configuration/crun-vs-runc-container-runtime/): Compare crun vs runc: 50% faster container startup and 10x lower memory footprint. Includes benchmarks, cgroup v2 support, and migration guide.
- [Validate CSI Storage Performance with FIO Kubernetes Job](https://kubernetes.recipes/recipes/storage/csi-storage-performance-fio-kubernetes-job/): Benchmark CSI storage performance using FIO inside a Kubernetes Job. Create a PVC backed by a CSI StorageClass, run sequential/random read/write
- [CUDA Version Compatibility K8s Guide](https://kubernetes.recipes/recipes/ai/cuda-compatibility-kubernetes-guide/): Match CUDA versions with GPU drivers and container images on Kubernetes. Forward compatibility, driver requirements, and container toolkit matrix.
- [Fix CUDA Out of Memory K8s Pods](https://kubernetes.recipes/recipes/troubleshooting/cuda-out-of-memory-kubernetes/): Troubleshoot CUDA out of memory errors in Kubernetes GPU pods. Memory fragmentation, batch size tuning, gradient checkpointing, and resource limits.
- [Add Custom CA Certificates in Kubernetes](https://kubernetes.recipes/recipes/security/custom-ca-kubernetes/): Configure custom Certificate Authority trust in vanilla Kubernetes using ConfigMap mounts, node-level trust stores, and containerd registry configuration.
- [Add Custom CA Certificates in OpenShift](https://kubernetes.recipes/recipes/security/custom-ca-openshift/): Configure custom Certificate Authority trust across an OpenShift cluster using proxy config, image config, and automatic CA bundle injection into pods.
- [Create Custom Resource Definitions (CRDs)](https://kubernetes.recipes/recipes/configuration/custom-resource-definitions/): Extend Kubernetes API with Custom Resource Definitions. Define custom objects, configure validation schemas, and manage CRD lifecycle.
- [CVE-2026-31431 Linux Kernel Crypto Fix](https://kubernetes.recipes/recipes/security/cve-2026-31431-linux-kernel-crypto-algif-aead/): Security advisory for CVE-2026-31431: Linux kernel crypto algif_aead vulnerability. Impact on Kubernetes nodes and how to patch container host kernels.
- [CVE-2026-3865: CSI SMB Driver Path Traversal](https://kubernetes.recipes/recipes/security/cve-2026-3865-csi-smb-path-traversal/): Fix CVE-2026-3865 Kubernetes CSI SMB driver path traversal vulnerability. Upgrade to v1.20.1, detect malicious PersistentVolumes.
- [CVE-2026-43284 "Dirty Frag": OpenShift Privilege Escalation](https://kubernetes.recipes/recipes/security/cve-2026-43284-dirtyfrag-openshift/): Mitigate CVE-2026-43284 'Dirty Frag' kernel privilege escalation on OpenShift 4 by disabling esp4, esp6, and rxrpc kernel modules via a GitOps DaemonSet.
- [CVE-2026-4342: ingress-nginx Code Execution Fix](https://kubernetes.recipes/recipes/security/cve-2026-4342-ingress-nginx-fix/): Patch CVE-2026-4342 in ingress-nginx — a CVSS 8.8 configuration injection vulnerability enabling arbitrary code execution. Upgrade to v1.13.9, v1.14.
- [Data Monetization Platforms on Kubernetes](https://kubernetes.recipes/recipes/deployments/data-monetization-platform-kubernetes/): Build data monetization platforms on Kubernetes. Data marketplace APIs, usage-based billing, data mesh architecture, secure data sharing, and catalog services.
- [Data Sovereignty and Geopatriation](https://kubernetes.recipes/recipes/security/data-sovereignty-geopatriation-kubernetes/): Implement data sovereignty and geopatriation on Kubernetes. Multi-region clusters, data residency policies, sovereign cloud, GDPR compliance.
- [Databases on K8s: Memory Overcommit](https://kubernetes.recipes/recipes/configuration/databases-kubernetes-memory-overcommit/): Why vm.overcommit_memory must be disabled for production databases on Kubernetes. Configure guaranteed QoS, disable swap.
- [DCB on Mellanox ConnectX for Lossless Ethernet](https://kubernetes.recipes/recipes/networking/dcb-mellanox-connectx-lossless-ethernet/): Configure Data Center Bridging (DCB) on Mellanox ConnectX NICs. DCBX negotiation, PFC, ETS, and CN for lossless RoCE Ethernet in Kubernetes AI clusters.
- [Debug Degraded MachineConfigPool Nodes](https://kubernetes.recipes/recipes/troubleshooting/debug-mcp-degraded-nodes/): Fix nodes stuck Degraded after MachineConfig updates. Check MCD logs, on-disk validation, and recovery for degraded workers.
- [Debug NCCL Timeouts and Hangs in Kubernetes](https://kubernetes.recipes/recipes/troubleshooting/debug-nccl-timeouts-kubernetes/): Systematically troubleshoot NCCL runs that stall or timeout across multi-GPU and multi-node Kubernetes jobs with step-by-step diagnostic commands.
- [How to Debug Kubernetes Node Issues](https://kubernetes.recipes/recipes/troubleshooting/debug-node-issues/): Diagnose and troubleshoot node problems in Kubernetes clusters. Identify resource pressure, connectivity issues, and component failures.
- [Kubernetes Pod Evicted: Causes, Debugging, Prevention](https://kubernetes.recipes/recipes/troubleshooting/debug-pod-eviction-reasons/): Why Kubernetes evicts pods: MemoryPressure, DiskPressure, ephemeral storage, preemption, NoExecute taints and drains. Diagnose, clean up, prevent.
- [How to Debug Pod Networking Issues](https://kubernetes.recipes/recipes/troubleshooting/debug-pod-networking/): Diagnose and fix Kubernetes networking problems. Troubleshoot connectivity, DNS resolution, service discovery, and network policies with practical tools.
- [Debug Pod Scheduling Failures in K8s](https://kubernetes.recipes/recipes/troubleshooting/debug-scheduling-failures/): Fix pods stuck in Pending from scheduling failures. Diagnose resource constraints, node affinity, taints, tolerations, and topology spread conflicts.
- [Decode and Inspect Kubernetes Docker Secrets](https://kubernetes.recipes/recipes/troubleshooting/decode-docker-registry-secrets/): Decode base64-encoded dockerconfigjson secrets to verify registry credentials, troubleshoot ImagePullBackOff errors, and audit pull secret configurations.
- [Deep Learning with Large Datasets on K8s](https://kubernetes.recipes/recipes/ai/deep-learning-large-dataset-kubernetes/): Optimize deep learning training with large datasets on Kubernetes. Covers data loading, caching strategies, parallel prefetch, and storage architecture
- [DeepSpeed ZeRO Training Kubernetes](https://kubernetes.recipes/recipes/ai/deepspeed-kubernetes-distributed/): Deploy DeepSpeed ZeRO-1/2/3 for large model training on Kubernetes. Multi-node config, NCCL tuning, memory optimization, and 70B+ model training.
- [Dell PowerEdge XE7740 GPU Node Setup](https://kubernetes.recipes/recipes/ai/dell-poweredge-xe7740-gpu-setup/): Configure Dell PowerEdge XE7740 GPU nodes with H200 GPUs for OpenShift and Kubernetes including BIOS, power, cooling, and network setup.
- [Dell Switch RoCEv2 PFC ECN DSCP](https://kubernetes.recipes/recipes/networking/dell-switch-rocev2-pfc-ecn-dscp/): Configure Dell OS10 switches for lossless RoCEv2 with PFC, ECN, WRED, and DSCP-to-traffic-class mapping. Priority 3 for RDMA traffic classes 24 and 26.
- [Deploy Fish Audio TTS on Kubernetes](https://kubernetes.recipes/recipes/ai/deploy-fish-audio-tts-kubernetes/): Deploy Fish Audio S2-Pro 5B text-to-speech model on Kubernetes for high-quality voice synthesis with multi-speaker support and streaming audio.
- [Deploy GLM-5 754B on Kubernetes](https://kubernetes.recipes/recipes/ai/deploy-glm5-754b-kubernetes/): Deploy Zhipu AI GLM-5 754B model on Kubernetes with vLLM. One of the largest open-weight models with multi-node tensor parallelism across 8+ GPUs.
- [Deploy Granite 4.0 Speech on Kubernetes](https://kubernetes.recipes/recipes/ai/deploy-granite-speech-kubernetes/): Deploy IBM Granite 4.0 1B Speech model on Kubernetes for automatic speech recognition. Lightweight 2B model runs on CPU or small GPU for STT workloads.
- [Deploy Kimi K2.5 1.1T MoE on Kubernetes](https://kubernetes.recipes/recipes/ai/deploy-kimi-k25-kubernetes/): Deploy Moonshot AI Kimi-K2.5 1.1T MoE multimodal model on Kubernetes. The largest open MoE model with 2.69M downloads for frontier AI tasks.
- [Deploy Llama 2 70B on Kubernetes](https://kubernetes.recipes/recipes/ai/deploy-llama2-70b-kubernetes/): Deploy Meta Llama 2 70B on Kubernetes with multi-GPU tensor parallelism, vLLM serving, and production-ready health checks and resource limits.
- [Deploy Llama 3.1 8B Instruct on K8s](https://kubernetes.recipes/recipes/ai/deploy-llama31-8b-kubernetes/): Deploy Meta Llama 3.1 8B Instruct on Kubernetes with vLLM. Production-ready single-GPU deployment with 128K context, tool calling, and autoscaling.
- [Deploy LTX Video Generation on K8s](https://kubernetes.recipes/recipes/ai/deploy-ltx-video-kubernetes/): Deploy Lightricks LTX-2.3 image-to-video model on Kubernetes for AI video generation with batch processing and S3 output storage.
- [Deploy MiniMax M2.5 229B on Kubernetes](https://kubernetes.recipes/recipes/ai/deploy-minimax-m25-kubernetes/): Deploy MiniMax M2.5 229B model on Kubernetes with vLLM. High-performance LLM with 485K downloads, optimized for multi-turn conversation and long context.
- [Deploy Mistral 7B with NVIDIA NIM](https://kubernetes.recipes/recipes/ai/deploy-mistral-nvidia-nim/): Step-by-step guide to deploy Mistral-7B using NVIDIA NIM with TensorRT-LLM backend on Kubernetes for optimized GPU inference.
- [Deploy Mistral 7B with vLLM on Kubernetes](https://kubernetes.recipes/recipes/ai/deploy-mistral-vllm-kubernetes/): Step-by-step guide to deploy Mistral-7B-v0.1 using vLLM as an OpenAI-compatible inference server on Kubernetes with GPU fractioning.
- [Deploy Multinode NIM Models on Kubernetes](https://kubernetes.recipes/recipes/ai/deploy-multinode-nim-models-kubernetes/): Run large language models across multiple GPU nodes with NVIDIA NIM. Tensor parallelism, NCCL, InfiniBand, and Kubernetes Job orchestration.
- [Deploy NVIDIA Nemotron 120B MoE on K8s](https://kubernetes.recipes/recipes/ai/deploy-nemotron-120b-kubernetes/): Deploy NVIDIA Nemotron-3-Super-120B-A12B MoE model on Kubernetes. 120B total parameters with 12B active for enterprise-grade inference.
- [Deploy Microsoft Phi-4 on Kubernetes](https://kubernetes.recipes/recipes/ai/deploy-phi4-kubernetes/): Deploy Microsoft Phi-4 small language model on Kubernetes with vLLM. Efficient 14B model with GPT-4 level reasoning on a single GPU.
- [Deploy Phi-4 Reasoning Vision on K8s](https://kubernetes.recipes/recipes/ai/deploy-phi4-reasoning-vision-kubernetes/): Deploy Microsoft Phi-4-reasoning-vision-15B on Kubernetes for multimodal chain-of-thought reasoning with visual understanding on a single GPU.
- [Deploy Qwen3 235B MoE on Kubernetes](https://kubernetes.recipes/recipes/ai/deploy-qwen3-235b-moe-kubernetes/): Deploy Alibaba Qwen3-235B-A22B mixture-of-experts model on Kubernetes. Only 22B parameters active per token for efficient 235B-class inference.
- [Deploy Qwen3 Coder 80B on Kubernetes](https://kubernetes.recipes/recipes/ai/deploy-qwen3-coder-kubernetes/): Deploy Qwen3-Coder-Next 80B on Kubernetes for code generation, review, and refactoring. Production-ready AI coding assistant with multi-GPU serving.
- [Deploy Qwen3 TTS on Kubernetes](https://kubernetes.recipes/recipes/ai/deploy-qwen3-tts-kubernetes/): Deploy Qwen3-TTS-12Hz-1.7B-CustomVoice on Kubernetes for text-to-speech with custom voice cloning. 1.13M downloads, lightweight single-GPU deployment.
- [Deploy Qwen3.5 35B MoE on Kubernetes](https://kubernetes.recipes/recipes/ai/deploy-qwen35-35b-moe-kubernetes/): Deploy Alibaba Qwen3.5-35B-A3B mixture-of-experts multimodal model on Kubernetes. 35B total parameters with only 3B active for ultra-efficient inference.
- [Deploy Qwen3.5 397B MoE on Kubernetes](https://kubernetes.recipes/recipes/ai/deploy-qwen35-397b-moe-kubernetes/): Deploy Alibaba Qwen3.5-397B-A17B MoE multimodal model on Kubernetes. 397B total parameters with only 17B active per token for frontier VLM inference.
- [Deploy Qwen3.5 9B Multimodal on K8s](https://kubernetes.recipes/recipes/ai/deploy-qwen35-9b-kubernetes/): Deploy Alibaba Qwen3.5-9B vision-language model on Kubernetes with vLLM. Process images and text with a single GPU deployment.
- [RetinaNet Object Detection on K8s](https://kubernetes.recipes/recipes/ai/deploy-retinanet-kubernetes/): Deploy RetinaNet object detection model on Kubernetes with Triton Inference Server, TensorRT optimization, and batch processing pipelines.
- [Deploy Sarvam 105B on Kubernetes](https://kubernetes.recipes/recipes/ai/deploy-sarvam-105b-kubernetes/): Deploy Sarvam 105B multilingual LLM on Kubernetes with vLLM. India's largest open language model with native support for 10+ Indic languages.
- [Stable Diffusion XL on Kubernetes](https://kubernetes.recipes/recipes/ai/deploy-stable-diffusion-xl-kubernetes/): Deploy Stable Diffusion XL for image generation on Kubernetes with TensorRT acceleration, queued batch processing, and S3 output storage.
- [Deploy Whisper Speech-to-Text on K8s](https://kubernetes.recipes/recipes/ai/deploy-whisper-kubernetes/): Deploy OpenAI Whisper for speech-to-text on Kubernetes with faster-whisper, batch transcription Jobs, and real-time streaming endpoints.
- [Kubernetes Deployment Strategies: Blue-Green, Canary](https://kubernetes.recipes/recipes/deployments/deployment-strategies/): Kubernetes deployment strategies compared: rolling update, recreate, blue-green and canary. YAML, traffic splitting, Argo Rollouts, Flagger and rollback.
- [Deployment vs StatefulSet in Kubernetes](https://kubernetes.recipes/recipes/deployments/deployment-vs-statefulset/): Choose between Deployment and StatefulSet for your Kubernetes workloads. Compare identity, storage, ordering, scaling, and use cases for each controller.
- [DGX H100 GPU Topology nvidia-smi](https://kubernetes.recipes/recipes/ai/dgx-h100-gpu-topology-nvidia-smi/): Inspect DGX H100 GPU topology with nvidia-smi topo -m. NVSwitch NV18 links, cross-socket detection, PCIe hierarchy, and NCCL performance validation.
- [DGX H100 nvidia-smi topo -m Guide](https://kubernetes.recipes/recipes/ai/dgx-h100-nvidia-smi-topo-kubernetes/): Read nvidia-smi topo -m output on DGX H100 systems. Understand NVLink, NVSwitch, PCIe topology, GPU-to-GPU bandwidth, and NUMA affinity for Kubernetes.
- [Diagnose NVIDIA Memory-Only Kernel Modules](https://kubernetes.recipes/recipes/troubleshooting/diagnose-nvidia-memory-only-modules/): Understand why lsmod shows NVIDIA modules loaded but modinfo fails, and how the GPU Operator's proprietary driver container inserts modules without.
- [Digital Provenance and Content Authenticity](https://kubernetes.recipes/recipes/security/digital-provenance-content-authenticity-kubernetes/): Implement digital provenance on Kubernetes with C2PA content credentials. Verify AI-generated content, sign media pipelines.
- [Disable PCIe ACS for GPU-Direct P2P](https://kubernetes.recipes/recipes/ai/disable-acs-pcie-gpu-direct-p2p/): Disable PCIe ACS so GPUs and RDMA NICs can use GPUDirect P2P DMA: BIOS setting, setpci and pcie_acs_override, IOMMU passthrough, and nvidia-smi topo checks.
- [Disable GDS and Enable IOMMU Passthrough on K8s GPUs](https://kubernetes.recipes/recipes/ai/disable-gds-enable-iommu-passthrough-kubernetes/): Disable GPUDirect Storage (GDS) when not needed and configure IOMMU passthrough mode for GPU and NIC device assignment. Kernel parameters, BIOS settings, VFIO
- [DisaggregatedSet for Multi-Role LLM Inference](https://kubernetes.recipes/recipes/ai/disaggregatedset-leaderworkerset-llm-inference-kubernetes/): Deploy disaggregated LLM inference on Kubernetes with DisaggregatedSet and LeaderWorkerSet. Separate prefill and decode phases across GPU pools
- [Disconnected Environments OpenShift](https://kubernetes.recipes/recipes/deployments/disconnected-environments-openshift-guide/): Complete guide to OpenShift disconnected and air-gapped environments. Mirror registry, oc-mirror, OLM, OSUS, IDMS, upgrades, and enclave support overview.
- [Distributed fio Storage Benchmark K8s](https://kubernetes.recipes/recipes/storage/distributed-fio-kubernetes-openshift/): Run distributed fio benchmarks on Kubernetes and OpenShift to test storage performance at scale. Covers fio-distributed with k8s Jobs, Red Hat dbench.
- [Distributed Inference Kubernetes](https://kubernetes.recipes/recipes/ai/distributed-inference-kubernetes/): Deploy distributed LLM inference with tensor parallelism across multiple GPUs and pipeline parallelism across nodes on Kubernetes.
- [Distributed Inference with Run:ai](https://kubernetes.recipes/recipes/ai/distributed-inference-runai-kubernetes/): Deploy distributed AI inference with NVIDIA Run:ai on Kubernetes. Single-node Knative, multinode LeaderWorkerSet, NIM, autoscaling, and observability.
- [Distributed Multi-GPU Inference on Kubernetes](https://kubernetes.recipes/recipes/ai/distributed-multi-gpu-inference-kubernetes/): Deploy distributed inference across multiple GPUs and nodes on Kubernetes. Covers tensor parallelism, pipeline parallelism, vLLM, and NIM multi-GPU serving.
- [Distributed Training TensorFlow PyTorch](https://kubernetes.recipes/recipes/ai/distributed-training-tensorflow-pytorch-kubernetes/): Run distributed training jobs on Kubernetes with TensorFlow and PyTorch. Training Operator, multi-worker strategies, NCCL configuration.
- [Kubernetes dnsPolicy: ClusterFirst, dnsConfig, ndots](https://kubernetes.recipes/recipes/networking/dns-policies-configuration/): Kubernetes pod dnsPolicy explained: ClusterFirst, ClusterFirstWithHostNet, Default, None. dnsConfig nameservers, search domains, ndots tuning, limits.
- [Debug DNS Resolution Failures in Pods](https://kubernetes.recipes/recipes/troubleshooting/dns-resolution-failure-pods/): Troubleshoot pods unable to resolve DNS names. Check CoreDNS health, ndots configuration, search domains, and NetworkPolicies blocking UDP port 53 DNS traffic.
- [Run DOCA Bench on OpenShift with SR-IOV and Privileged SCC](https://kubernetes.recipes/recipes/networking/doca-bench-openshift-sriov-privileged/): Run NVIDIA DOCA Bench as a Kubernetes Job on OpenShift with SR-IOV VF allocation, privileged SCC, and huge pages to benchmark BlueField DPU from x86 pods.
- [NVIDIA DOCA Driver Container in Kubernetes](https://kubernetes.recipes/recipes/configuration/doca-driver-container-kubernetes/): Deploy and configure NVIDIA DOCA Driver containers via NicClusterPolicy for RDMA, NFS-RDMA, and precompiled driver builds.
- [DOCA Driver on OpenShift with DTK](https://kubernetes.recipes/recipes/configuration/doca-driver-openshift-dtk/): Build and deploy precompiled NVIDIA DOCA Driver containers on OpenShift using DriverToolKit, MachineConfig, and upgrade lifecycle.
- [DOCA Perftest RDMA Benchmarking](https://kubernetes.recipes/recipes/ai/doca-perftest-rdma-benchmark-kubernetes/): Run NVIDIA DOCA perftest on Kubernetes to benchmark RDMA bandwidth and latency between GPU nodes. Traffic patterns, GPUDirect memory modes.
- [DOCA Telemetry BlueField Kubernetes](https://kubernetes.recipes/recipes/observability/doca-telemetry-bluefield-kubernetes/): Collect NVIDIA BlueField DPU telemetry in Kubernetes using DOCA Telemetry libraries. Monitor adaptive retransmission, PCC, diagnostics, and PCI metrics.
- [Domain-Specific Language Models on Kubernetes](https://kubernetes.recipes/recipes/ai/domain-specific-language-models-kubernetes/): Deploy and fine-tune domain-specific LLMs on Kubernetes. Legal, healthcare, finance, and code models with LoRA fine-tuning, NIM serving, and RAG pipelines.
- [Multi-Node NVLink with ComputeDomains](https://kubernetes.recipes/recipes/configuration/dra-computedomains-nvlink/): Configure ComputeDomains for robust and secure Multi-Node NVLink (MNNVL) workloads on NVIDIA GB200 and similar systems using DRA
- [Dynamic Resource Allocation for NVIDIA GPUs](https://kubernetes.recipes/recipes/configuration/dra-gpu-allocation/): Learn to use Kubernetes Dynamic Resource Allocation (DRA) for flexible GPU allocation, sharing, and configuration with the NVIDIA DRA Driver
- [MIG GPU Partitioning with DRA on Kubernetes](https://kubernetes.recipes/recipes/configuration/dra-mig-partitioning/): Dynamically partition NVIDIA A100 and H100 GPUs using Multi-Instance GPU (MIG) technology with Dynamic Resource Allocation for flexible workload isolation
- [Mixed Accelerator Workloads with DRA](https://kubernetes.recipes/recipes/configuration/dra-mixed-accelerators/): Orchestrate heterogeneous accelerator workloads combining GPUs, TPUs, FPGAs, and custom AI chips using Dynamic Resource Allocation
- [TPU Allocation Dynamic Resource Allocation](https://kubernetes.recipes/recipes/configuration/dra-tpu-allocation/): Configure Google Cloud TPUs in Kubernetes using DRA for flexible allocation, multi-slice workloads, and optimized machine learning training
- [Dual-Fabric Mellanox: GPU InfiniBand + Storage Ethernet](https://kubernetes.recipes/recipes/networking/dual-fabric-mellanox-gpu-storage-ethernet-infiniband/): Design and configure dual-fabric network architecture with separate Mellanox NICs for GPU communication (InfiniBand) and storage traffic (Ethernet). Covers
- [How to Configure Dynamic Volume Provisioning](https://kubernetes.recipes/recipes/storage/dynamic-volume-provisioning/): Set up dynamic volume provisioning in Kubernetes with StorageClasses. Learn to configure provisioners for AWS EBS, GCP PD, Azure Disk, and NFS.
- [ECN MachineConfig OpenShift Nodes](https://kubernetes.recipes/recipes/networking/ecn-machineconfig-openshift-nodes/): Enable ECN (Explicit Congestion Notification) on OpenShift nodes via MachineConfig for lossless RoCEv2 RDMA networking. Sysctl and Mellanox NIC configuration.
- [EDR Flexera Agents Kubernetes Deploy](https://kubernetes.recipes/recipes/security/edr-flexera-kubernetes-agents/): Deploy EDR and Flexera agents on Kubernetes with DaemonSets. Priority classes, host path access, exclusion paths, and security agent lifecycle.
- [emptyDir Volumes: Sharing, Lifecycle, and Memory-Backed](https://kubernetes.recipes/recipes/storage/emptydir-volume-sharing-lifecycle-memory-backed/): Master emptyDir volumes for CKA/CKAD exam prep. Share data between containers, understand volume lifecycle across restarts vs Pod deletion, and configure
- [Enable GPUDirect Storage on OpenShift](https://kubernetes.recipes/recipes/storage/enable-gpudirect-storage-openshift/): Configure GPUDirect Storage (GDS) with the NVIDIA GPU Operator on OpenShift, including the Open Kernel Module requirement and nvidia-fs verification.
- [Enable NIC Feature Discovery on Kubernetes](https://kubernetes.recipes/recipes/networking/enable-nic-feature-discovery/): Enable NIC Feature Discovery through NicClusterPolicy and verify the node labels required by SR-IOV and RDMA GPU networking workflows on Kubernetes.
- [Configure Environment Variables and ConfigMaps](https://kubernetes.recipes/recipes/configuration/environment-variables-configmaps/): Manage application configuration with environment variables and ConfigMaps. Learn injection methods, mounting as files, and dynamic configuration updates.
- [etcd Backup and Restore: Kubernetes and OpenShift](https://kubernetes.recipes/recipes/storage/etcd-backup-restore-kubernetes/): Back up and restore etcd on kubeadm and OpenShift: etcdctl/etcdutl snapshots, cluster-backup.sh, automated CronJobs to S3, test restores, and DR steps.
- [Fix etcd High Latency and Slow API Server](https://kubernetes.recipes/recipes/troubleshooting/etcd-high-latency-troubleshooting/): Debug etcd performance issues causing slow kubectl responses and API server timeouts. Covers disk I/O, compaction, defragmentation, and leader elections.
- [Debug etcd Performance Issues in Kubernetes](https://kubernetes.recipes/recipes/troubleshooting/etcd-performance-troubleshooting/): Diagnose slow etcd causing API latency and leader election storms. Check disk IOPS, compaction, defrag, and network latency.
- [ETS, PFC & DSCP Trust on Mellanox Quickstart](https://kubernetes.recipes/recipes/networking/ets-pfc-dscp-trust-mellanox-quickstart/): Quick reference for enabling ETS queues, PFC, DSCP trust, and DSCP-to-priority mapping on Mellanox ConnectX NICs. Three commands for lossless RoCE Ethernet.
- [External DNS for Kubernetes: Setup Guide](https://kubernetes.recipes/recipes/networking/external-dns-kubernetes/): Automate DNS record management with ExternalDNS for Kubernetes. Sync Service and Ingress hostnames to Route53, CloudFlare, Google Cloud DNS, and 30+ providers.
- [External Secrets Operator: Vault, AWS, Azure, GCP](https://kubernetes.recipes/recipes/security/external-secrets-operator/): Install External Secrets Operator and sync secrets from HashiCorp Vault, AWS Secrets Manager, Azure Key Vault or GCP into Kubernetes Secrets, with templates.
- [External Secrets Operator on OpenShift](https://kubernetes.recipes/recipes/security/external-secrets-operator-openshift/): Manage Kubernetes secrets from external vaults using External Secrets Operator on OpenShift. Covers ExternalSecret CRD, SecretStore configuration, and GitOps
- [Falco Runtime Security for Kubernetes](https://kubernetes.recipes/recipes/security/falco-runtime-security/): Deploy Falco for Kubernetes runtime threat detection: modern eBPF install, default rules, custom rules, Falcosidekick alert routing and automated response.
- [Feast Feature Store Kubernetes](https://kubernetes.recipes/recipes/ai/feast-feature-store-kubernetes/): Deploy Feast feature store on Kubernetes for ML feature management. Offline and online stores, feature serving, point-in-time joins.
- [Fix fio libaio Silent Exit on OpenShift crun](https://kubernetes.recipes/recipes/troubleshooting/fio-libaio-seccomp-crun-openshift/): Debug fio instantly exiting with no output on crun-based OpenShift nodes. The root cause is seccomp blocking libaio syscalls — fix with psync or unconfined.
- [Fio NFS Benchmark on OpenShift Nodes](https://kubernetes.recipes/recipes/storage/fio-nfs-benchmark-openshift/): Run fio NFS storage benchmarks on OpenShift using parallel pods with hostPath mounts. Measure IOPS, bandwidth, and latency across multiple NFS endpoints.
- [PScale NFS and SMB Storage Benchmarking](https://kubernetes.recipes/recipes/storage/fio-pscale-nfs-smb-benchmark-kubernetes/): Benchmark NFS and SMB storage performance on Kubernetes using fio clients in Pods. Covers multi-client parallel testing, bandwidth measurement, and IOPS
- [Fix 502 Bad Gateway in Kubernetes Ingress](https://kubernetes.recipes/recipes/troubleshooting/fix-502-bad-gateway-kubernetes/): Fix 502 Bad Gateway in Kubernetes and NGINX Ingress: empty endpoints, selector/port mismatch, rollout termination races, timeouts, and protocol mismatch.
- [Fix Expired Certificates in Kubernetes](https://kubernetes.recipes/recipes/troubleshooting/fix-certificate-expiration-cluster/): Renew expired certificates causing API server failures and kubelet disconnections. Manual and automatic renewal for kubeadm and OpenShift.
- [Fix etcd Leader Election Timeout](https://kubernetes.recipes/recipes/troubleshooting/fix-etcd-leader-election-timeout/): Troubleshoot etcd leader election timeouts in K8s. Disk latency, network partition, heartbeat interval, and recovery steps.
- [Fix Certificate Errors Kubernetes](https://kubernetes.recipes/recipes/troubleshooting/fix-kubernetes-certificate-errors/): Troubleshoot TLS certificate errors in K8s. x509 unknown authority, expired certs, cert-manager issues, and custom CA bundles.
- [Fix Pod cgroup Memory Errors K8s](https://kubernetes.recipes/recipes/troubleshooting/fix-kubernetes-pod-cgroup-errors/): Fix cgroup memory limit and OOM errors in Kubernetes pods. Covers cgroup v2 migration, memory.max, swap settings, and kernel tuning for stable workloads.
- [Fix Service Not Reachable in Kubernetes](https://kubernetes.recipes/recipes/troubleshooting/fix-kubernetes-service-not-reachable/): Debug Kubernetes Service connectivity issues. Endpoint selection, kube-proxy rules, DNS resolution, and NetworkPolicy blocks.
- [Fix NVIDIA Peer Memory Driver Not Detected](https://kubernetes.recipes/recipes/troubleshooting/fix-nvidia-peermem-not-detected/): Diagnose and resolve the 'NVIDIA peer memory driver not detected' error when running GPU workloads with RDMA on Kubernetes and OpenShift.
- [Flexera License Management Kubernetes](https://kubernetes.recipes/recipes/configuration/flexera-license-management-kubernetes/): Manage software licenses in Kubernetes with Flexera. FlexNet Manager, container license tracking, GPU software metering, and compliance for enterprise K8s.
- [Flux GitOps on Kubernetes: Bootstrap to HelmRelease](https://kubernetes.recipes/recipes/deployments/flux-gitops/): Flux CD GitOps: bootstrap on GitHub/GitLab, GitRepository + Kustomization, HelmRelease, image automation, multi-tenant repo layout, alerts and debugging.
- [Flux vs ArgoCD: Kubernetes GitOps Compared](https://kubernetes.recipes/recipes/deployments/flux-vs-argocd-gitops-comparison/): Compare Flux and ArgoCD for Kubernetes GitOps. Architecture, multi-tenancy, Helm support, UI, scalability, and when to choose each for production GitOps de.
- [FSDP LoRA Fine-Tuning LLMs on Kubernetes](https://kubernetes.recipes/recipes/ai/fsdp-lora-finetuning-kubernetes/): Fine-tune large language models with FSDP and LoRA on Kubernetes. Covers memory-efficient loading, checkpoint strategies, and multi-node H200 training.
- [Full GitOps Pipeline k3s to Production](https://kubernetes.recipes/recipes/deployments/full-gitops-pipeline-k3s/): End-to-end GitOps pipeline: git push triggers Gitea Actions build, pushes to quay.io, Octopus Deploy creates release with ephemeral preview.
- [Gateway API HTTPRoutes TLS on k3s](https://kubernetes.recipes/recipes/networking/gateway-api-httproutes-tls-k3s/): Configure Gateway API HTTPRoutes with TLS termination on k3s using Cilium. Route traffic to multiple services with wildcard certificates and HTTP-to-HTTPS .
- [GPU Operator GDS with NVMe and NFS RDMA](https://kubernetes.recipes/recipes/storage/gds-nvme-nfs-rdma/): Configure GPUDirect Storage for local NVMe drives and NFS over RDMA in Kubernetes, including cuFile verification and performance benchmarking.
- [GenAI-Perf: Benchmark LLM Inference on Kubernetes](https://kubernetes.recipes/recipes/ai/genai-perf-nvidia-inference-benchmarking/): Benchmark Triton, vLLM and NIM endpoints with NVIDIA GenAI-Perf on Kubernetes: TTFT, ITL, throughput, concurrency sweeps, GPU telemetry, and AIPerf migration.
- [Gitea Actions Runner Push to Quay](https://kubernetes.recipes/recipes/deployments/gitea-actions-runner-quay-push/): Deploy Gitea Actions runner on k3s to build container images and push to quay.io. DinD-less builds with Kaniko, automated CI pipelines for every git push.
- [Gitea PostgreSQL Valkey on k3s](https://kubernetes.recipes/recipes/deployments/gitea-postgresql-valkey-k3s/): Deploy self-hosted Gitea with PostgreSQL and Valkey (Redis fork) on k3s. Complete Git forge with Actions CI runner, container registry, and package management.
- [GitLab Runner Helm Kubernetes Executor](https://kubernetes.recipes/recipes/deployments/gitlab-runner-helm-kubernetes-executor/): Deploy GitLab Runner on Kubernetes with Helm. Configure concurrent jobs, internal registry, PodMonitor metrics, scale-to-zero, security contexts.
- [GitOps for AI Workloads on Kubernetes](https://kubernetes.recipes/recipes/ai/gitops-ai-workloads-kubernetes/): Deploy AI models with GitOps on Kubernetes. Version ML models in Git, ArgoCD for model rollouts, Flux for GPU cluster sync.
- [GitOps Bootstrap for Bare-Metal GPU Clusters](https://kubernetes.recipes/recipes/deployments/gitops-bootstrap-bare-metal-gpu/): Bootstrap bare-metal GPU clusters with ArgoCD and Kustomize in air-gapped environments with NVIDIA GPU and Network Operators.
- [GKE OIDC Issuer Workload Identity](https://kubernetes.recipes/recipes/security/gke-oidc-issuer-workload-identity/): Enable OIDC issuer on GKE with --enable-oidc-issuer. Configure workload identity federation for cross-cloud auth and external IdP integration.
- [Enable GPUDirect Storage in ClusterPolicy](https://kubernetes.recipes/recipes/ai/gpu-clusterpolicy-gds-enable/): Enable NVIDIA GPUDirect Storage (GDS) in the GPU Operator ClusterPolicy for direct GPU-to-NVMe data paths. Driver module configuration and verification.
- [GPU Feature Discovery Node Labels](https://kubernetes.recipes/recipes/ai/gpu-feature-discovery-kubernetes/): Configure NVIDIA GPU Feature Discovery for automatic node labeling on Kubernetes. GPU model, driver version, CUDA, and MIG labels for scheduling.
- [GPU Node Affinity Scheduling K8s](https://kubernetes.recipes/recipes/ai/gpu-node-affinity-scheduling/): Schedule GPU workloads with node affinity and topology on Kubernetes. GPU type selection, multi-GPU locality, and NUMA-aware pod placement.
- [GPU Operator ClusterPolicy RDMA and GDS Configuration](https://kubernetes.recipes/recipes/ai/gpu-operator-clusterpolicy-rdma-gds-iommu/): Configure NVIDIA GPU Operator ClusterPolicy to disable RDMA and enable GPUDirect Storage (GDS). Control nvidia-peermem, nvidia-fs modules, driver
- [GPU Operator ClusterPolicy Complete Reference](https://kubernetes.recipes/recipes/configuration/gpu-operator-clusterpolicy-reference/): Complete reference for the NVIDIA GPU Operator ClusterPolicy CRD covering driver, toolkit, device plugin, MOFED, GDS, MIG, and DCGM configuration options.
- [GPU Operator GPUDirect Storage GDS Module](https://kubernetes.recipes/recipes/storage/gpu-operator-gds-module/): Enable the GPUDirect Storage GDS module in the NVIDIA GPU Operator ClusterPolicy for direct GPU-to-storage data transfers bypassing CPU and system memory.
- [NVIDIA GPU Operator MOFED Driver Configuration](https://kubernetes.recipes/recipes/configuration/gpu-operator-mofed-driver/): Configure the NVIDIA GPU Operator to deploy Mellanox OFED drivers for high-performance RDMA networking on Kubernetes GPU nodes with InfiniBand and RoCE support.
- [GPU Operator Node Validation Metrics](https://kubernetes.recipes/recipes/observability/gpu-operator-node-status-exporter-metrics/): Monitor NVIDIA GPU Operator node validation with gpu_operator_node_driver_ready and status exporter metrics. Prometheus alerts for GPU node health.
- [GPU Operator Canary Upgrade Strategy](https://kubernetes.recipes/recipes/deployments/gpu-operator-upgrade-canary/): Safely upgrade NVIDIA GPU Operator using canary node pools, 48-hour bake periods, validation gates, and Git-based rollback.
- [GPU Tenant Bootstrap Bundle for Kubernetes](https://kubernetes.recipes/recipes/security/gpu-tenant-bootstrap-bundle/): Provision GPU tenants with a single Kustomize bundle containing namespace, RBAC, NetworkPolicy, quotas, and HAProxy VIP config.
- [Per-Tenant GPU Monitoring and Chargeback](https://kubernetes.recipes/recipes/observability/gpu-tenant-monitoring-chargeback/): Build per-tenant GPU monitoring dashboards with queue time, utilization, thermal metrics, and GPU-hour chargeback on Kubernetes.
- [GPU Tenant SLO Observability on Kubernetes](https://kubernetes.recipes/recipes/observability/gpu-tenant-slo-observability/): Define and monitor GPU tenant SLOs for queue time, inference latency, GPU utilization, and job completion rate with Prometheus alerting.
- [Configure GPU Time-Slicing on Kubernetes](https://kubernetes.recipes/recipes/ai/gpu-time-slicing-kubernetes/): Configure NVIDIA GPU time-slicing with the GPU Operator: device plugin ConfigMap, replicas, per-node profiles, verification, and memory-sharing caveats.
- [GPU Cluster Upgrade Version Matrix](https://kubernetes.recipes/recipes/configuration/gpu-upgrade-version-matrix/): Maintain a version compatibility matrix for GPU Operator, Network Operator, drivers, firmware, CUDA, and OpenShift for safe upgrades.
- [GPUDirect RDMA via DMA-BUF on Kubernetes](https://kubernetes.recipes/recipes/networking/gpudirect-rdma-dma-buf/): Configure GPUDirect RDMA using DMA-BUF kernel subsystem for zero-copy GPU-to-GPU transfers over InfiniBand and RoCE networks.
- [GPUDirect RDMA Setup and Verification on Kubernetes](https://kubernetes.recipes/recipes/ai/gpudirect-rdma-setup-verification-kubernetes/): Enable and verify GPUDirect RDMA for GPU-to-NIC direct data transfer on Kubernetes. Install nvidia-peermem, configure DMA-BUF, verify RDMA paths, troubleshoot
- [GPUDirect Storage on Kubernetes](https://kubernetes.recipes/recipes/ai/gpudirect-storage-gds-kubernetes/): Configure NVIDIA GPUDirect Storage (GDS) for direct data path between NVMe/NFS storage and GPU memory bypassing CPU. Covers Magnum IO, cuFile API, GDS driver
- [Grafana Dashboard 6417 Kubernetes Pods](https://kubernetes.recipes/recipes/observability/grafana-dashboard-6417-kubernetes/): Import Grafana dashboard 6417 for Kubernetes pod monitoring. Configure Prometheus data source, visualize CPU, memory, network, and disk usage per pod.
- [Grafana Kubernetes Monitoring Dashboards Guide](https://kubernetes.recipes/recipes/observability/grafana-kubernetes-monitoring-dashboards/): Deploy and configure Grafana dashboards for Kubernetes monitoring including dashboard 6417 for pod metrics, dashboard 315 for cluster overview, and custom
- [gVisor on Kubernetes: RuntimeClass Sandbox (runsc)](https://kubernetes.recipes/recipes/security/gvisor-container-runtime/): Run sandboxed pods with gVisor on Kubernetes: install runsc, configure containerd, create the gvisor RuntimeClass, GKE Sandbox, enforcement, verification.
- [H200 NVL 8-GPU Topology Bandwidth Tiers for Kubernetes](https://kubernetes.recipes/recipes/ai/h200-nvl4-topology-bandwidth-tiers-kubernetes/): Map the three bandwidth tiers of 8× H200 NVL GPU nodes—NVLink (~337 GB/s), PCIe+UPI (~50 GB/s), RoCE (~35 GB/s)—for NCCL topology-aware NUMA scheduling.
- [HAProxy Keepalived Multi-Tenant GPU Ingress](https://kubernetes.recipes/recipes/networking/haproxy-keepalived-multi-tenant-gpu/): Configure HAProxy with Keepalived VIPs for per-tenant GPU cluster ingress with Jinja2 templates and per-tenant access logging.
- [How to Integrate HashiCorp Vault with K8s](https://kubernetes.recipes/recipes/security/hashicorp-vault-kubernetes/): Securely manage secrets with HashiCorp Vault in Kubernetes. Learn to inject secrets into pods using the Vault Agent Injector and CSI Provider.
- [Helm before-hook-creation Hook](https://kubernetes.recipes/recipes/helm/helm-before-hook-creation/): Use Helm before-hook-creation for database migrations and pre-install checks. Complete hook lifecycle, delete policies, and ordering.
- [How to Create and Use Helm Charts](https://kubernetes.recipes/recipes/helm/helm-chart-basics/): Master Helm, the Kubernetes package manager. Learn to create charts, manage releases, and template your deployments for reusability.
- [Helm Chart Dependencies: Condition, Alias, Lock](https://kubernetes.recipes/recipes/helm/helm-chart-dependencies-guide/): Helm subchart dependencies in Chart.yaml: condition and tags flags, alias for multiple instances, import-values, Chart.lock, and dependency update vs build.
- [Helm Chart Development from Scratch](https://kubernetes.recipes/recipes/helm/helm-chart-development-guide/): Build production-ready Helm charts with templates, values, helpers, hooks, tests, and CI validation. Complete guide from chart create to publishing.
- [How to Create Helm Chart Repositories](https://kubernetes.recipes/recipes/helm/helm-chart-repositories/): Set up and manage Helm chart repositories. Learn to host charts on GitHub Pages, S3, GCS, and OCI registries for team distribution.
- [Helm Chart Testing & CI/CD Pipeline Integration](https://kubernetes.recipes/recipes/helm/helm-chart-testing-ci/): Test Helm charts automatically with ct (chart-testing), helm unittest, and GitHub Actions. Validate templates, lint values.
- [Helm Hook Delete Policy Explained](https://kubernetes.recipes/recipes/helm/helm-hook-delete-policy/): Configure Helm hook delete policies: before-hook-creation, hook-succeeded, hook-failed. Control Job cleanup after install, upgrade, and test hooks.
- [Helm Hooks: pre-install, post-upgrade, Weights](https://kubernetes.recipes/recipes/helm/helm-hooks-lifecycle/): Helm hooks explained: pre-install, post-upgrade, pre-delete and test hooks for migrations and backups, ordered by hook weights, cleaned by hook-delete-policy.
- [Helm Install: Deploy Charts Guide](https://kubernetes.recipes/recipes/helm/helm-install-chart-guide/): Install Helm charts on Kubernetes with helm install, upgrade, rollback, and values customization. Repository management, OCI registries, and release lifecycle.
- [Helm Library Charts: Reusable Templates Guide](https://kubernetes.recipes/recipes/helm/helm-library-charts/): Build a Helm library chart (type: library) to share Deployment, Service and label templates across charts, consume it as a dependency, and override safely.
- [Helm OCI Registry: Push, Pull and Install Charts](https://kubernetes.recipes/recipes/helm/helm-oci-registry-charts/): Push, pull and install Helm charts from OCI registries (GHCR, ECR, ACR, Harbor): helm registry login, helm push, Argo CD and Flux OCI sources, CI release.
- [Helm Rollback and History Guide](https://kubernetes.recipes/recipes/helm/helm-rollback-history-guide/): Roll back Helm releases and manage revision history. Diagnose failed upgrades, compare revisions, and automate rollback.
- [Helm Secrets Mgmt with SOPS & Age Encryption](https://kubernetes.recipes/recipes/helm/helm-secrets-sops-management/): Encrypt Helm values files using SOPS with Age or GPG keys. Manage secrets in Git safely with helm-secrets plugin for transparent encrypt/decrypt workflows.
- [Helm Sprig add1 trim merge Functions](https://kubernetes.recipes/recipes/helm/helm-sprig-add1-trim-merge-functions/): Helm Sprig add1 function increments integers in templates. Plus trim for whitespace removal and merge for combining dictionaries in Helm charts.
- [Helm Sprig cat Function: Concatenate Strings](https://kubernetes.recipes/recipes/helm/helm-sprig-cat-function/): Helm Sprig cat function concatenates strings with spaces between arguments. Syntax, why cat inserts spaces, conditionals, and template examples.
- [Helm Sprig Functions Complete Reference](https://kubernetes.recipes/recipes/helm/helm-sprig-functions-complete-reference/): Complete reference for Helm Sprig template functions including cat, print, join, tostring, add1, trim, quote, default, and more. Examples and common patterns
- [Helm Sprig join Function: List to String](https://kubernetes.recipes/recipes/helm/helm-sprig-join-function/): Helm Sprig join function converts lists to delimited strings. Join list example with CSV output, label values, and multi-value template patterns.
- [Helm Sprig print quote default Functions](https://kubernetes.recipes/recipes/helm/helm-sprig-print-quote-default-functions/): Helm Sprig print function concatenates without spaces, quote wraps in double quotes, default provides fallback values. Template examples and patterns.
- [Helm Sprig toString Function Guide](https://kubernetes.recipes/recipes/helm/helm-sprig-tostring-function/): Helm Sprig toString function converts values to strings in templates. Handle integers, booleans, lists, and nil values safely in Helm charts.
- [Helm Sprig Functions: cat, print, toString](https://kubernetes.recipes/recipes/helm/helm-templating-sprig/): Helm Sprig function reference with copy-paste examples and output: cat, print, toString, add1, quote, join, default, and regexReplaceAll explained.
- [Fix Helm Upgrade Failed and Rollback](https://kubernetes.recipes/recipes/helm/helm-upgrade-failed-troubleshooting/): Debug failed Helm releases stuck in pending-upgrade or failed state. Covers atomic upgrades, manual rollback, secret storage cleanup, and history limits.
- [Helm Values Override Patterns Explained](https://kubernetes.recipes/recipes/helm/helm-values-override-patterns/): Master Helm values override patterns. CLI flags, multiple files, JSON values, and precedence rules for complex deployments.
- [Hermes Agent Self-Hosted AI on Kubernetes](https://kubernetes.recipes/recipes/ai/hermes-agent-self-hosted-kubernetes/): Deploy Hermes Agent (Nous Research) on Kubernetes as a persistent self-hosted AI agent with memory, automated skill creation, multi-platform
- [Kubernetes HPA: Set Max Replicas and Scale](https://kubernetes.recipes/recipes/autoscaling/horizontal-pod-autoscaler/): Configure Kubernetes HPA (autoscaling/v2): averageUtilization, set or change maxReplicas, CPU, memory and custom metrics, and scaling behavior.
- [Hugo nginx Static Site on a k3s Cluster](https://kubernetes.recipes/recipes/deployments/hugo-nginx-static-site-k3s/): Deploy a Hugo static site with nginx on k3s. Multi-stage build, Brotli compression, security headers, and automated redeployment on git push via Gitea Actions.
- [ib_write_bw RDMA Bandwidth Testing](https://kubernetes.recipes/recipes/networking/ib-write-bw-rdma-benchmark-kubernetes/): Run ib_write_bw from perftest on Kubernetes GPU nodes: SR-IOV device selection, GPUDirect RDMA, multi-QP scaling, full CLI reference, and RoCE tuning.
- [Identify Mellanox NIC Models from Linux](https://kubernetes.recipes/recipes/troubleshooting/identify-mellanox-nic-models/): Map interface names to PCI addresses and Mellanox model generations to build accurate SR-IOV policies and GPU networking configurations on Kubernetes.
- [IDMS ITMS ICSP Disconnected OpenShift](https://kubernetes.recipes/recipes/configuration/idms-itms-disconnected-openshift/): Configure ImageDigestMirrorSet, ImageTagMirrorSet, and ImageContentSourcePolicy for disconnected OpenShift. Redirect image pulls to your mirror registry.
- [Kubernetes Image Pull Optimization: Faster Pod Starts](https://kubernetes.recipes/recipes/configuration/image-pull-optimization-kubernetes/): Speed up Kubernetes image pulls: pre-pull DaemonSets, containerd registry mirrors, lazy pulling with eStargz/SOCI, parallel pulls, and GC tuning for AI images.
- [How to Configure Image Pull Secrets](https://kubernetes.recipes/recipes/configuration/image-pull-secrets/): Pull container images from private registries using image pull secrets. Configure authentication for Docker Hub, GCR, ECR, ACR, and private registries.
- [Fix ImagePullBackOff and ErrImagePull in Kubernetes](https://kubernetes.recipes/recipes/troubleshooting/imagepullbackoff-troubleshooting/): Fix ImagePullBackOff, ErrImagePull and Back-off pulling image: wrong tags, pull secrets, Docker Hub rate limits, registry CA and network errors.
- [Fix OpenShift ImageStream Import Errors](https://kubernetes.recipes/recipes/troubleshooting/imagestream-import-errors/): Debug ImageStream import failures in OpenShift. Resolve DNS errors, auth issues, TLS problems, and registry rate limiting.
- [InfiniBand vs Ethernet for AI on Kubernetes](https://kubernetes.recipes/recipes/networking/infiniband-ethernet-ai-kubernetes/): Compare InfiniBand and Ethernet networking for GPU AI workloads on Kubernetes, including RDMA, RoCE, latency, and throughput considerations.
- [InfiniBand Subnet Manager OpenSM on Kubernetes](https://kubernetes.recipes/recipes/networking/infiniband-subnet-manager-opensm-kubernetes/): Deploy and manage InfiniBand Subnet Manager (OpenSM) on Kubernetes for GPU cluster fabric management. Covers SM architecture, UFM integration, partition
- [Fix Ingress 502 and 503 Gateway Errors](https://kubernetes.recipes/recipes/networking/ingress-502-503-troubleshooting/): Debug 502 Bad Gateway and 503 Service Unavailable from Kubernetes ingress controllers. Fix backend health and timeout issues.
- [Secure Ingress with SSL/TLS Certificates](https://kubernetes.recipes/recipes/networking/ingress-tls-certificates/): Configure TLS termination for Kubernetes Ingress using cert-manager and Let's Encrypt. Automate certificate issuance and renewal.
- [Migrate Ingress to Gateway API ingress2gateway](https://kubernetes.recipes/recipes/networking/ingress2gateway-migration/): Migrate Ingress to Gateway API using ingress2gateway. Convert HTTPRoute and TLSRoute with zero-downtime parallel migration.
- [Install ArgoCD on AlmaLinux: Step-by-Step](https://kubernetes.recipes/recipes/deployments/install-argocd-alma-linux/): Deploy ArgoCD on Kubernetes running on AlmaLinux. GitOps continuous delivery with automated sync, self-healing, and multi-cluster support.
- [Install ArgoCD on Amazon Linux](https://kubernetes.recipes/recipes/deployments/install-argocd-amazon-linux/): Deploy ArgoCD on Kubernetes running on Amazon Linux. GitOps continuous delivery with automated sync, self-healing, and multi-cluster support.
- [Install ArgoCD on Arch Linux: Step-by-Step](https://kubernetes.recipes/recipes/deployments/install-argocd-arch-linux/): Deploy ArgoCD on Kubernetes running on Arch Linux. GitOps continuous delivery with automated sync, self-healing, and multi-cluster support.
- [Install ArgoCD on CentOS Stream](https://kubernetes.recipes/recipes/deployments/install-argocd-centos-stream/): Deploy ArgoCD on Kubernetes running on CentOS Stream. GitOps continuous delivery with automated sync, self-healing, and multi-cluster support.
- [Install ArgoCD on Debian: Step-by-Step Guide](https://kubernetes.recipes/recipes/deployments/install-argocd-debian/): Deploy ArgoCD on Kubernetes running on Debian. GitOps continuous delivery with automated sync, self-healing, and multi-cluster support.
- [Install ArgoCD on Fedora: Step-by-Step Guide](https://kubernetes.recipes/recipes/deployments/install-argocd-fedora/): Deploy ArgoCD on Kubernetes running on Fedora. GitOps continuous delivery with automated sync, self-healing, and multi-cluster support.
- [Install ArgoCD on openSUSE: Step-by-Step](https://kubernetes.recipes/recipes/deployments/install-argocd-opensuse/): Deploy ArgoCD on Kubernetes running on openSUSE. GitOps continuous delivery with automated sync, self-healing, and multi-cluster support.
- [Install ArgoCD on Oracle Linux](https://kubernetes.recipes/recipes/deployments/install-argocd-oracle-linux/): Deploy ArgoCD on Kubernetes running on Oracle Linux. GitOps continuous delivery with automated sync, self-healing, and multi-cluster support.
- [Install ArgoCD on RHEL: Step-by-Step Guide](https://kubernetes.recipes/recipes/deployments/install-argocd-rhel/): Deploy ArgoCD on Kubernetes running on RHEL. GitOps continuous delivery with automated sync, self-healing, and multi-cluster support.
- [Install ArgoCD on Rocky Linux Step-by-Step](https://kubernetes.recipes/recipes/deployments/install-argocd-rocky-linux/): Deploy ArgoCD on Kubernetes running on Rocky Linux. GitOps continuous delivery with automated sync, self-healing, and multi-cluster support.
- [Install ArgoCD on SUSE SLES: Step-by-Step](https://kubernetes.recipes/recipes/deployments/install-argocd-suse-sles/): Deploy ArgoCD on Kubernetes running on SUSE SLES. GitOps continuous delivery with automated sync, self-healing, and multi-cluster support.
- [Install ArgoCD on Ubuntu: Step-by-Step Guide](https://kubernetes.recipes/recipes/deployments/install-argocd-ubuntu/): Deploy ArgoCD on Kubernetes running on Ubuntu. GitOps continuous delivery with automated sync, self-healing, and multi-cluster support.
- [Install Helm on AlmaLinux: Setup Guide](https://kubernetes.recipes/recipes/helm/install-helm-alma-linux/): Install Helm 3 on AlmaLinux and configure chart repositories. Covers package manager install, script install, and shell completion for AlmaLinux 8/9.
- [Install Helm on Amazon Linux: Setup Guide](https://kubernetes.recipes/recipes/helm/install-helm-amazon-linux/): Install Helm on Amazon Linux 2023 and AL2. Three install methods, chart repository setup, shell completion, and troubleshooting for Amazon Linux environments.
- [Install Helm on Arch Linux: Setup Guide](https://kubernetes.recipes/recipes/helm/install-helm-arch-linux/): Install Helm 3 on Arch Linux and configure chart repositories. Covers package manager install, script install, and shell completion for Arch Linux rolling.
- [Install Helm on CentOS Stream Setup Guide](https://kubernetes.recipes/recipes/helm/install-helm-centos-stream/): Install Helm 3 on CentOS Stream and configure chart repositories. Covers package manager install, script install, and shell completion for CentOS Stream 9.
- [Install Helm on Debian: Setup Guide](https://kubernetes.recipes/recipes/helm/install-helm-debian/): Install Helm 3 on Debian and configure chart repositories. Covers package manager install, script install, and shell completion for Debian 11/12.
- [Install Helm on Fedora: Setup Guide](https://kubernetes.recipes/recipes/helm/install-helm-fedora/): Install Helm 3 on Fedora and configure chart repositories. Covers package manager install, script install, and shell completion for Fedora 39/40.
- [Install Helm on openSUSE: Setup Guide](https://kubernetes.recipes/recipes/helm/install-helm-opensuse/): Install Helm 3 on openSUSE with package manager or script. Configure chart repos and shell completion for openSUSE Leap 15 / Tumbleweed.
- [Install Helm on Oracle Linux: Setup Guide](https://kubernetes.recipes/recipes/helm/install-helm-oracle-linux/): Install Helm 3 on Oracle Linux and configure chart repositories. Covers package manager install, script install, and shell completion for Oracle Linux 8/9.
- [Install Helm on RHEL: Complete Setup Guide](https://kubernetes.recipes/recipes/helm/install-helm-rhel/): Install Helm 3 on RHEL and configure chart repositories. Covers package manager install, script install, and shell completion for RHEL 8/9.
- [Install Helm on Rocky Linux: Setup Guide](https://kubernetes.recipes/recipes/helm/install-helm-rocky-linux/): Install Helm 3 on Rocky Linux and configure chart repositories. Covers package manager install, script install, and shell completion for Rocky Linux 8/9.
- [Install Helm on SUSE SLES: Setup Guide](https://kubernetes.recipes/recipes/helm/install-helm-suse-sles/): Install Helm 3 on SUSE SLES and configure chart repositories. Covers package manager install, script install, and shell completion for SLES 15.
- [Install Helm on Ubuntu 22.04/24.04 (2 Methods)](https://kubernetes.recipes/recipes/helm/install-helm-ubuntu/): Install Helm 3 on Ubuntu with the official script or apt repository. Step-by-step commands, verification, and shell completion for Ubuntu 22.04 and 24.04.
- [Install Kubernetes on AlmaLinux](https://kubernetes.recipes/recipes/deployments/install-kubernetes-alma-linux/): Step-by-step guide to install Kubernetes on AlmaLinux with kubeadm. Covers containerd, kubeadm init, CNI setup, and worker node joining for AlmaLinux 8/9.
- [Install Kubernetes on Amazon Linux](https://kubernetes.recipes/recipes/deployments/install-kubernetes-amazon-linux/): Install Kubernetes on Amazon Linux with kubeadm. Covers containerd setup, kubeadm init, Calico CNI, and worker node joining for Amazon Linux 2023.
- [Install Kubernetes on Arch Linux](https://kubernetes.recipes/recipes/deployments/install-kubernetes-arch-linux/): Step-by-step guide to install Kubernetes on Arch Linux with kubeadm. Covers containerd, kubeadm init, CNI setup, and worker node joining for Arch Linux rolling.
- [Install Kubernetes on CentOS Stream](https://kubernetes.recipes/recipes/deployments/install-kubernetes-centos-stream/): Step-by-step guide to install Kubernetes on CentOS Stream with kubeadm. Covers containerd, kubeadm init, CNI setup, and worker node joining for CentOS Stream 9.
- [Install Kubernetes on Debian: Setup Guide](https://kubernetes.recipes/recipes/deployments/install-kubernetes-debian/): Step-by-step guide to install Kubernetes on Debian with kubeadm. Covers containerd, kubeadm init, CNI setup, and worker node joining for Debian 11/12.
- [Install Kubernetes on Fedora: Setup Guide](https://kubernetes.recipes/recipes/deployments/install-kubernetes-fedora/): Step-by-step guide to install Kubernetes on Fedora with kubeadm. Covers containerd, kubeadm init, CNI setup, and worker node joining for Fedora 39/40.
- [Install Kubernetes on Fedora with kubeadm](https://kubernetes.recipes/recipes/configuration/install-kubernetes-fedora-kubeadm/): Step-by-step guide to install Kubernetes on Fedora Linux using kubeadm. Disable swap, configure containerd, install kubeadm kubelet kubectl.
- [Install Kubernetes on openSUSE](https://kubernetes.recipes/recipes/deployments/install-kubernetes-opensuse/): Install Kubernetes on openSUSE with kubeadm. Covers containerd setup, kubeadm init, Calico CNI, and worker node joining for openSUSE Leap 15 / Tumbleweed.
- [Install Kubernetes on Oracle Linux](https://kubernetes.recipes/recipes/deployments/install-kubernetes-oracle-linux/): Step-by-step guide to install Kubernetes on Oracle Linux with kubeadm. Covers containerd, kubeadm init, CNI setup, and worker node joining for Oracle Linux 8/9.
- [Install Kubernetes on RHEL: Setup Guide](https://kubernetes.recipes/recipes/deployments/install-kubernetes-rhel/): Step-by-step guide to install Kubernetes on RHEL with kubeadm. Covers containerd, kubeadm init, CNI setup, and worker node joining for RHEL 8/9.
- [Install Kubernetes on Rocky Linux](https://kubernetes.recipes/recipes/deployments/install-kubernetes-rocky-linux/): Step-by-step guide to install Kubernetes on Rocky Linux with kubeadm. Covers containerd, kubeadm init, CNI setup, and worker node joining for Rocky Linux 8/9.
- [Install Kubernetes on SUSE SLES](https://kubernetes.recipes/recipes/deployments/install-kubernetes-suse-sles/): Step-by-step guide to install Kubernetes on SUSE SLES with kubeadm. Covers containerd, kubeadm init, CNI setup, and worker node joining for SLES 15.
- [Install Kubernetes on Ubuntu: Setup Guide](https://kubernetes.recipes/recipes/deployments/install-kubernetes-ubuntu/): Step-by-step guide to install Kubernetes on Ubuntu with kubeadm. Covers containerd, kubeadm init, CNI setup, and worker node joining for Ubuntu 22.04/24.04.
- [Integrate DisaggregatedSet with llm-d on Kubernetes](https://kubernetes.recipes/recipes/ai/integrate-disaggregatedset-llmd-kubernetes/): Deploy disaggregated LLM inference using DisaggregatedSet and llm-d on Kubernetes. Install LWS then DS controller, model prefill/decode roles, wire llm-d
- [Inter-Node Tensor Parallelism on Kubernetes](https://kubernetes.recipes/recipes/ai/inter-node-tensor-parallelism-kubernetes/): Split a single LLM across multiple physical servers using tensor parallelism. Configure vLLM, NIM, and Ray for inter-node TP with NCCL over RDMA or TCP.
- [IOMMU BIOS and Kernel Config for NCCL GPU-Direct](https://kubernetes.recipes/recipes/ai/iommu-bios-kernel-nccl-gpu-direct/): Configure IOMMU at BIOS and kernel level to enable NCCL GPU-Direct RDMA on Kubernetes. Covers Intel VT-d, AMD-Vi, kernel parameters, passthrough
- [IOMMU on K8s: GPU Passthrough and SR-IOV](https://kubernetes.recipes/recipes/configuration/iommu-gpu-passthrough-sriov-kubernetes/): Enable and configure IOMMU for GPU passthrough, SR-IOV, and VFIO on Kubernetes. Kernel parameters, IOMMU groups, device isolation, and troubleshooting guide.
- [IOMMU Kernel Parameters for Kubernetes GPU Nodes](https://kubernetes.recipes/recipes/ai/iommu-kernel-parameters-gpu-nodes-kubernetes/): Configure IOMMU kernel parameters for optimal GPU and RDMA performance on Kubernetes. Compare intel_iommu, amd_iommu, and iommu settings, passthrough vs off vs
- [ITMS External-to-External Registry Mirroring](https://kubernetes.recipes/recipes/configuration/itms-external-registry-mapping/): Configure OpenShift ImageTagMirrorSet to map external registries to your private registry. Mirror Docker Hub, GHCR, Quay.io, and NVIDIA NGC.
- [Mirror the AI/GPU Platform Stack for Disconnected OpenShift](https://kubernetes.recipes/recipes/configuration/itms-idms-ai-platform-disconnected-mirror/): Configure ImageDigestMirrorSet and ImageTagMirrorSet to mirror Run:ai, Kubeflow, MPI Operator, and Dell CSM images for air-gapped GPU platforms.
- [ITMS Race Condition with Ingress Controllers](https://kubernetes.recipes/recipes/troubleshooting/itms-ingress-controller-race-condition/): Resolve the ITMS race condition where ImageTagMirrorSet rollouts deadlock with hostNetwork ingress controllers during MCO drain.
- [ITMS Updates registries.conf via MachineConfig](https://kubernetes.recipes/recipes/configuration/itms-registries-conf-machineconfig/): How ITMS and IDMS update /etc/containers/registries.conf on immutable CoreOS nodes via MCO and MachineConfig. Full chain deep-dive.
- [Jaeger Tracing on Kubernetes: Setup Guide](https://kubernetes.recipes/recipes/observability/jaeger-distributed-tracing/): Set up Jaeger tracing on Kubernetes step by step: install the operator, deploy Jaeger, and instrument Python, Node.js, and Go apps to trace microservices.
- [Fix Kubernetes Job Failures and Retries](https://kubernetes.recipes/recipes/troubleshooting/job-failure-troubleshooting/): Debug Kubernetes Jobs stuck in backoff or hitting retry limits. Covers backoffLimit, activeDeadlineSeconds, and CronJob overlap.
- [Journald Verify Config Kubernetes Nodes](https://kubernetes.recipes/recipes/troubleshooting/journald-verify-config-kubernetes/): Validate journald configuration on Kubernetes nodes. Fix journal corruption, tune storage limits, configure persistence, and troubleshoot systemd-journald.
- [OpenShift Storage Benchmark fio Config Profiles](https://kubernetes.recipes/recipes/storage/k8s-fio-config-profiles-openshift/): Benchmark OpenShift and Kubernetes storage using fio with reusable YAML config profiles for random and sequential read/write I/O patterns.
- [K8s-IO Benchmark CLI for fio and HammerDB](https://kubernetes.recipes/recipes/storage/k8s-io-benchmark-cli-fio-hammerdb/): Run distributed fio and HammerDB storage benchmarks on Kubernetes with K8s-IO, a lightweight Go CLI tool that replaces heavy benchmark operators.
- [GPU Sharing and Bin Packing with KAI Scheduler](https://kubernetes.recipes/recipes/ai/kai-scheduler-gpu-sharing/): Maximize GPU utilization with KAI Scheduler GPU sharing, fractional GPUs, and bin packing strategies for Kubernetes AI workloads.
- [Installing NVIDIA KAI Scheduler AI Workloads](https://kubernetes.recipes/recipes/ai/kai-scheduler-installation/): Deploy KAI Scheduler for optimized GPU resource allocation in Kubernetes AI/ML clusters with hierarchical queues and batch scheduling
- [Batch Scheduling PodGroups in KAI Scheduler](https://kubernetes.recipes/recipes/ai/kai-scheduler-podgroups/): Implement gang scheduling for distributed training jobs using KAI Scheduler PodGroups to ensure all-or-nothing pod scheduling
- [KAI Scheduler Hierarchical Queues & Fairness](https://kubernetes.recipes/recipes/ai/kai-scheduler-queues/): Configure hierarchical queues in KAI Scheduler for multi-tenant GPU clusters with quotas, limits, and Dominant Resource Fairness (DRF)
- [Kairos k3s on Hetzner CPX42: Immutable Bootstrap](https://kubernetes.recipes/recipes/deployments/kairos-k3s-hetzner-immutable-bootstrap/): Deploy an immutable Kairos-based k3s cluster on Hetzner Cloud CPX42. Automated provisioning with cloud-init, immutable OS upgrades.
- [KAITO AI Model Inference Kubernetes](https://kubernetes.recipes/recipes/ai/kaito-ai-inference-kubernetes/): Deploy AI models with KAITO (Kubernetes AI Toolchain Operator) for automated GPU provisioning, model serving, and inference workload management.
- [Karpenter Node Autoscaling on EKS (v1 API)](https://kubernetes.recipes/recipes/autoscaling/karpenter-node-autoscaling/): Install Karpenter v1 on EKS: NodePools, EC2NodeClass, spot fallback, consolidation, GPU pools, and how it compares to Cluster Autoscaler.
- [Kata Containers RuntimeClass Kubernetes](https://kubernetes.recipes/recipes/security/kata-containers-runtimeclass-kubernetes/): Deploy Kata Containers with Kubernetes RuntimeClass for hardware-isolated pods. VM-based sandboxing, microVM configuration, and multi-runtime clusters.
- [Katib Hyperparameter Tuning Kubernetes](https://kubernetes.recipes/recipes/ai/katib-hyperparameter-tuning-kubernetes/): Automate hyperparameter tuning with Katib on Kubernetes. Bayesian optimization, random search, grid search, early stopping.
- [KEDA vs HPA: Event-Driven Autoscaling Explained](https://kubernetes.recipes/recipes/autoscaling/keda-vs-hpa-autoscaling-comparison/): Compare KEDA and HPA for Kubernetes autoscaling. Scale on Kafka lag, Prometheus metrics, queue depth, cron, and custom events. KEDA vs HPA decision guide.
- [How to Run Kubernetes in Docker (kind)](https://kubernetes.recipes/recipes/troubleshooting/kind-local-kubernetes/): Create local Kubernetes clusters using kind (Kubernetes in Docker). Set up multi-node clusters, configure networking, and test applications locally.
- [Configure Knative Ingress Networking](https://kubernetes.recipes/recipes/networking/knative-ingress-networking/): Set up Knative Serving ingress with Kourier, Istio, or Contour. Custom domains, TLS, path routing, and external visibility.
- [KnativeServing for AI Inference OpenShift](https://kubernetes.recipes/recipes/ai/knative-serving-nim-inference-openshift/): Configure KnativeServing with scale-to-zero, GPU scheduling features, Kourier ingress, and custom domain templates for AI inference workloads on OpenShift.
- [KServe Model Serving Kubernetes](https://kubernetes.recipes/recipes/ai/kserve-model-serving-kubernetes/): Deploy ML models with KServe for serverless inference on Kubernetes. InferenceService, scale-to-zero, canary rollouts, model transformers.
- [KubeCon EU 2026 Book Giveaway Recap](https://kubernetes.recipes/recipes/configuration/kubecon-eu-2026-book-giveaway-success/): Recap of the Kubernetes Recipes book giveaway at KubeCon EU 2026 Amsterdam. Photos from the signing sessions, community highlights, and how to get your copy.
- [KubeCon EU 2026 Book Signing Events](https://kubernetes.recipes/recipes/configuration/kubecon-eu-2026-book-signing/): Join Luca Berton at two KubeCon Amsterdam events: Signal Overflow at Booking.com HQ (Mon 23 Mar) and book signing at vCluster booth #521 (Tue 24 Mar).
- [kubectl apply vs create: Key Differences](https://kubernetes.recipes/recipes/configuration/kubectl-apply-vs-create/): Understand when to use kubectl apply vs kubectl create. Declarative vs imperative, last-applied annotation, server-side apply, and GitOps workflows.
- [kubectl Cheat Sheet: Essential Commands](https://kubernetes.recipes/recipes/configuration/kubectl-cheat-sheet/): Complete kubectl cheat sheet with essential commands for pods, deployments, services, debugging, and cluster management. Copy-paste ready examples.
- [kubectl Config: Manage Contexts and Clusters](https://kubernetes.recipes/recipes/configuration/kubectl-config-context-management/): Manage kubectl contexts with kubectl config commands. Switch clusters, delete contexts, rename entries, and merge multiple kubeconfig files safely.
- [kubectl cp Copy Files to and from Pods](https://kubernetes.recipes/recipes/troubleshooting/kubectl-cp-copy-files-pods/): Copy files between local machine and Kubernetes pods with kubectl cp. Supports containers, namespaces, tar-based transfer, and common troubleshooting.
- [kubectl create secret docker-registry](https://kubernetes.recipes/recipes/configuration/kubectl-create-secret-docker-registry/): Create Kubernetes Docker registry secrets with --docker-password-stdin. Authenticate to private registries and configure imagePullSecrets securely.
- [kubectl describe: Read Pod Events Guide](https://kubernetes.recipes/recipes/troubleshooting/kubectl-describe-pod-events/): Use kubectl describe pod to read events, conditions, and container states. Diagnose scheduling failures, image pulls, crashes, and probe failures.
- [kubectl exec: Run Commands in Pods](https://kubernetes.recipes/recipes/troubleshooting/kubectl-exec-into-pod/): Use kubectl exec to run commands inside running pods. Interactive shell, multi-container pods, debugging techniques, and security considerations.
- [kubectl get pods: Output Formats Guide](https://kubernetes.recipes/recipes/deployments/kubectl-get-pods-examples/): Master kubectl get pods with output formats, label selectors, field selectors, and custom columns. Wide output, JSON, YAML, and jsonpath examples.
- [kubectl logs View Pod Logs Guide](https://kubernetes.recipes/recipes/troubleshooting/kubectl-logs-view-pod-logs/): View and stream Kubernetes pod logs with kubectl logs. Multi-container pods, previous crashes, label selectors, timestamps, and log aggregation patterns.
- [kubectl rollout restart Deployment](https://kubernetes.recipes/recipes/deployments/kubectl-rollout-restart-deployment/): Restart Kubernetes Deployments, StatefulSets, and DaemonSets with kubectl rollout restart. Zero-downtime rolling restart without changing pod spec.
- [kubectl run: Create Pod from Command Line](https://kubernetes.recipes/recipes/deployments/kubectl-run-pod-command/): Use kubectl run to create pods and deployments from the command line. Dry-run output, resource limits, environment variables, and CKA exam patterns.
- [Distrib. Training Kubeflow Training Operator](https://kubernetes.recipes/recipes/ai/kubeflow-distributed-training/): Run multi-node distributed PyTorch and TensorFlow training jobs using Kubeflow Training Operator with NCCL, RDMA, and shared storage.
- [Kubeflow ML Platform Setup Kubernetes](https://kubernetes.recipes/recipes/ai/kubeflow-ml-platform-kubernetes/): Deploy Kubeflow as a production-ready ML platform on Kubernetes. Notebooks, pipelines, training operators, and model serving with KServe for end-to-end MLO.
- [Kubeflow MPIJob Worker SSH Setup for GPU Training](https://kubernetes.recipes/recipes/ai/kubeflow-mpijob-worker-ssh-gpu-training/): Configure SSH daemon in Kubeflow MPIJob worker pods for multi-node GPU training. Covers SSHD setup in containers, host key generation, authorized keys from MPI
- [Kubeflow Operator: Full ML Platform](https://kubernetes.recipes/recipes/ai/kubeflow-operator-platform/): Deploy the complete Kubeflow platform on Kubernetes with the Kubeflow Operator. Covers Pipelines, Notebooks, KServe, Katib, and multi-tenant ML workflows.
- [Kubeflow Training Operator on Kubernetes](https://kubernetes.recipes/recipes/ai/kubeflow-training-operator/): Install Kubeflow Training Operator for distributed ML training with PyTorchJob, TFJob, and MPIJob on GPU-enabled Kubernetes clusters.
- [Fix Kubelet NotReady and Node Pressure Issues](https://kubernetes.recipes/recipes/troubleshooting/kubelet-not-ready-troubleshooting/): Debug kubelet NotReady status, node pressure conditions, and eviction issues. Covers disk pressure, memory pressure, PID pressure, and network not ready.
- [Kubernetes 1.35 and 1.36 Upgrade Checklist](https://kubernetes.recipes/recipes/configuration/kubernetes-1-35-1-36-upgrade-checklist/): Kubernetes 1.35 and 1.36 upgrade checklist with deprecated APIs, removed features, new GA capabilities, and step-by-step migration guide for production clu.
- [Kubernetes 1.36 Constrained Impersonation](https://kubernetes.recipes/recipes/security/kubernetes-1-36-constrained-impersonation/): Use constrained impersonation in Kubernetes 1.36 to limit which identities a user can impersonate. Tighter RBAC control for multi-tenant clusters.
- [Kubernetes 1.36 CSI Differential Snapshots](https://kubernetes.recipes/recipes/storage/kubernetes-1-36-csi-differential-snapshots/): Use CSI differential snapshots in Kubernetes 1.36 to track changed blocks between snapshots. Enables incremental backups and faster disaster recovery.
- [Kubernetes 1.36 Declarative Type Validation](https://kubernetes.recipes/recipes/configuration/kubernetes-1-36-declarative-validation/): Kubernetes 1.36 introduces declarative validation for native API types using validation-gen. Replaces hand-written validation code with struct tag annotations.
- [Kubernetes 1.36 DRA for GPU and TPU Management](https://kubernetes.recipes/recipes/ai/kubernetes-1-36-dra-gpu-management/): Use Dynamic Resource Allocation in Kubernetes 1.36 for advanced GPU/TPU management with partitionable devices, device taints, and tolerations.
- [Kubernetes 1.36 External SA Token Signing](https://kubernetes.recipes/recipes/security/kubernetes-1-36-external-sa-token-signing/): Delegate ServiceAccount token signing to external KMS or HSM systems in Kubernetes 1.36. Improve security with hardware-backed key management.
- [Migrate from externalIPs in Kubernetes 1.36](https://kubernetes.recipes/recipes/networking/kubernetes-1-36-externalips-deprecation/): Service externalIPs are deprecated in Kubernetes 1.36 due to CVE-2020-8554. Migrate to Gateway API, LoadBalancer services, or MetalLB for external access.
- [Kubernetes 1.36 Gang Scheduling](https://kubernetes.recipes/recipes/ai/kubernetes-1-36-gang-scheduling/): Use gang scheduling in Kubernetes 1.36 to schedule Pod groups atomically. Essential for distributed ML training, MPI jobs, and Spark workloads.
- [Migrate from gitRepo Volume in Kubernetes 1.36](https://kubernetes.recipes/recipes/configuration/kubernetes-1-36-gitrepo-removal/): The gitRepo volume plugin is permanently removed in Kubernetes 1.36. Migrate to init containers or OCI volumes to avoid broken deployments.
- [Kubernetes 1.36 Graceful Leader Transition](https://kubernetes.recipes/recipes/configuration/kubernetes-1-36-graceful-leader-transition/): Configure graceful leader transitions in Kubernetes 1.36 control plane components. Eliminate brief outages during leader election failovers.
- [Kubernetes 1.36 L3 Cache Topology in CPU Manager](https://kubernetes.recipes/recipes/configuration/kubernetes-1-36-l3-cache-topology-cpu-manager/): Configure L3 cache topology awareness in Kubernetes 1.36 CPU Manager. Allocate CPUs sharing L3 cache for better performance in latency-sensitive workloads.
- [Kubernetes 1.36 Memory QoS with cgroups v2](https://kubernetes.recipes/recipes/configuration/kubernetes-1-36-memory-qos-cgroups-v2/): Configure memory quality of service with cgroups v2 in Kubernetes 1.36. Set memory.min and memory.high for guaranteed memory and throttling before OOM kills.
- [Kubernetes 1.36 Mixed Version Proxy](https://kubernetes.recipes/recipes/configuration/kubernetes-1-36-mixed-version-proxy/): Use the Mixed Version Proxy in Kubernetes 1.36 to handle API version skew during rolling upgrades. Ensures API availability across mixed control plane versions.
- [Kubernetes 1.36 Native Histogram Metrics](https://kubernetes.recipes/recipes/observability/kubernetes-1-36-native-histograms/): Enable Prometheus native histograms in Kubernetes 1.36 for higher-resolution metrics with lower storage cost. Covers all control plane components.
- [Kubernetes 1.36 OCI Volume Source](https://kubernetes.recipes/recipes/storage/kubernetes-1-36-oci-volume-source/): Use OCI VolumeSource in Kubernetes 1.36 to pull OCI artifacts directly into Pod volumes. No init containers needed for ML models, configs, or data.
- [Kubernetes 1.36 Pod Certificates (mTLS)](https://kubernetes.recipes/recipes/security/kubernetes-1-36-pod-certificates/): Use Pod Certificates in Kubernetes 1.36 to authenticate Pods to the API server via mTLS. Built-in X.509 certificate provisioning without external tools.
- [Kubernetes 1.36 Pod-Level Resource Limits](https://kubernetes.recipes/recipes/configuration/kubernetes-1-36-pod-level-resources/): Set resource requests and limits at the Pod level in Kubernetes 1.36 instead of per-container. Simplifies multi-container Pod resource management.
- [Kubernetes 1.36 RestartAllContainers for ML](https://kubernetes.recipes/recipes/ai/kubernetes-1-36-restart-all-containers/): Use the RestartAllContainers policy in Kubernetes 1.36 to restart all Pod containers in-place when a worker fails, avoiding costly ML training rescheduling.
- [Kubernetes 1.36 SELinux Mount-Time Labeling](https://kubernetes.recipes/recipes/security/kubernetes-1-36-selinux-mount-labeling/): Configure SELinux mount-time volume labeling in Kubernetes 1.36 to eliminate slow recursive relabeling and speed up Pod startup times dramatically.
- [Kubernetes 1.36 SPDY to WebSocket Migration](https://kubernetes.recipes/recipes/networking/kubernetes-1-36-spdy-to-websockets/): Kubernetes 1.36 continues migrating kubectl exec/attach/port-forward from SPDY to WebSockets. Understand the changes and troubleshoot connection issues.
- [Kubernetes 1.36 Statusz and Flagz Endpoints](https://kubernetes.recipes/recipes/troubleshooting/kubernetes-1-36-statusz-flagz/): Use /statusz and /flagz debug endpoints in Kubernetes 1.36 control plane components. Inspect runtime status and effective flag values without log parsing.
- [Kubernetes 1.36 Topology-Aware Scheduling](https://kubernetes.recipes/recipes/ai/kubernetes-1-36-topology-aware-scheduling/): Use topology-aware workload scheduling in Kubernetes 1.36 to place Pods on nodes with optimal GPU, NUMA, and network topology for ML training.
- [Kubernetes 1.36 User Namespaces in Pods](https://kubernetes.recipes/recipes/security/kubernetes-1-36-user-namespaces/): Enable user namespaces in Kubernetes 1.36 for rootless containers and stronger Pod isolation. Map container root to unprivileged host UIDs.
- [Kubernetes 1.36 VolumeGroupSnapshot GA](https://kubernetes.recipes/recipes/storage/kubernetes-1-36-volume-group-snapshot/): Use VolumeGroupSnapshot in Kubernetes 1.36 to take crash-consistent snapshots of multiple volumes atomically. Now GA and production-ready.
- [K8s Admission Controllers List Guide](https://kubernetes.recipes/recipes/configuration/kubernetes-admission-controller-list/): Complete list of Kubernetes admission controllers. Enable and disable controllers, PodSecurity, ResourceQuota, and custom validating webhooks guide.
- [Kubernetes Admission Controllers and Webhooks](https://kubernetes.recipes/recipes/security/kubernetes-admission-controllers-guide/): Build validating and mutating admission webhooks for Kubernetes. Policy enforcement with OPA Gatekeeper, Kyverno, and custom webhooks.
- [K8s Admission Webhooks: Validate and Mutate](https://kubernetes.recipes/recipes/security/kubernetes-admission-webhooks-guide/): Build Kubernetes validating and mutating admission webhooks. Webhook configuration, TLS setup, failure policies, and common patterns for policy enforcement.
- [Kubernetes Affinity and Anti-Affinity Guide](https://kubernetes.recipes/recipes/configuration/kubernetes-affinity-anti-affinity/): Schedule pods with Kubernetes node affinity, pod affinity, and anti-affinity rules. Spread across zones, co-locate related services, and optimize
- [Kubernetes Node and Pod Affinity Guide](https://kubernetes.recipes/recipes/configuration/kubernetes-affinity-guide/): Configure node affinity, pod affinity, and anti-affinity rules for advanced Kubernetes scheduling. Control pod placement across zones, nodes, and topologies.
- [AI Cost Management on Kubernetes](https://kubernetes.recipes/recipes/ai/kubernetes-ai-cost-management/): Control AI infrastructure costs on Kubernetes with GPU utilization tracking, chargeback per team, spot instance strategies, right-sizing recommendations.
- [K8s AI Gateway: Inference Extension Guide](https://kubernetes.recipes/recipes/ai/kubernetes-ai-gateway-inference-extension/): Use the Kubernetes AI Gateway and Inference Extension to route LLM traffic. Model-aware routing, load balancing across inference backends.
- [AI Inference Optimization Kubernetes](https://kubernetes.recipes/recipes/ai/kubernetes-ai-inference-optimization/): Optimize AI inference performance on Kubernetes. Request batching, KV cache tuning, speculative decoding, continuous batching.
- [Kubernetes AI Infrastructure Scaling](https://kubernetes.recipes/recipes/ai/kubernetes-ai-infrastructure-scaling/): What changes when Kubernetes AI inference infrastructure grows from 10K to 100K requests per second, including routing, caching, autoscaling, GPUs, and SLOs.
- [Kubernetes for AI Search and Discoverability](https://kubernetes.recipes/recipes/ai/kubernetes-ai-search-discoverability/): Deploy AI-searchable services on Kubernetes: llms.txt implementation, RAG-optimized APIs, structured data for AI chatbots, and infrastructure patterns
- [AI Workload Monitoring Kubernetes](https://kubernetes.recipes/recipes/observability/kubernetes-ai-workload-monitoring/): Monitor AI and GPU workloads on Kubernetes with DCGM Exporter, Prometheus, and Grafana. GPU utilization, memory usage, inference latency.
- [Kubernetes Alerting Best Practices](https://kubernetes.recipes/recipes/observability/kubernetes-alerting-best-practices/): Design effective Kubernetes alerts that reduce noise and catch real issues. Covers severity tiers, golden signals, runbook links, and fatigue prevention.
- [Alertmanager Routing, Grouping, and Silences](https://kubernetes.recipes/recipes/observability/kubernetes-alertmanager-routing-silences/): Configure Alertmanager routing trees, receiver integrations, inhibition rules, silences, and alert grouping for production Kubernetes monitoring stacks.
- [Kubernetes Annotations Complete Guide](https://kubernetes.recipes/recipes/configuration/kubernetes-annotations-guide/): Use Kubernetes annotations for metadata, automation, and controller config. Common patterns for ingress annotations, Helm labels, and triggers.
- [Kubernetes API Aggregation Layer](https://kubernetes.recipes/recipes/configuration/kubernetes-api-aggregation/): Extend the Kubernetes API with custom API servers using the aggregation layer to add new resource types and functionality without modifying core components
- [Kubernetes API Deprecation Migration Guide](https://kubernetes.recipes/recipes/configuration/kubernetes-api-deprecation-migration/): Migrate deprecated Kubernetes APIs before cluster upgrades. Detect deprecated resources with pluto, kubent, and kubectl convert.
- [API Priority and Fairness K8s Guide](https://kubernetes.recipes/recipes/configuration/kubernetes-api-priority-fairness/): Configure Kubernetes API Priority and Fairness to protect the API server. Covers FlowSchemas, PriorityLevelConfigurations, and request concurrency tuning.
- [kubectl explain: API Resource Reference](https://kubernetes.recipes/recipes/configuration/kubernetes-api-resources-explain/): Use kubectl explain and api-resources to discover Kubernetes API objects. Field documentation, resource versions, short names, and API group exploration.
- [Kubernetes API Versions Explained](https://kubernetes.recipes/recipes/configuration/kubernetes-api-versions-explained/): Understand K8s API versions: alpha, beta, stable. API deprecation policy, migration strategy, and kubectl api-versions usage.
- [Argo Rollouts Canary Blue-Green K8s](https://kubernetes.recipes/recipes/deployments/kubernetes-argo-rollouts-guide/): Progressive delivery with Argo Rollouts on Kubernetes. Canary, blue-green, analysis templates, and experiment-based promotion for safe deployments.
- [Argo Workflows on Kubernetes: DAGs, Artifacts, Cron](https://kubernetes.recipes/recipes/deployments/kubernetes-argo-workflows-guide/): Argo Workflows guide: install, steps vs DAG pipelines, artifact passing, parameters and conditionals, retries, CronWorkflow, WorkflowTemplates and RBAC.
- [ArgoCD Sync Waves and Hooks Guide](https://kubernetes.recipes/recipes/deployments/kubernetes-argocd-sync-waves/): Configure ArgoCD sync waves for ordered deployments. Wave ordering, sync hooks, resource health checks, and dependency management patterns.
- [K8s Audit Logging for Enterprise Compliance](https://kubernetes.recipes/recipes/security/kubernetes-audit-logging-compliance/): Configure API server audit logging for SOC2, HIPAA, and PCI-DSS compliance. Structured audit policies, log shipping, and alerting on suspicious activity.
- [Kubernetes Audit Logging Configuration](https://kubernetes.recipes/recipes/security/kubernetes-audit-logging-configuration/): Configure Kubernetes audit logging: audit policy levels and rules, kube-apiserver flags, webhook backend, OpenShift audit profiles, and jq queries.
- [Kubernetes Blue-Green and Canary Deployment Strategies](https://kubernetes.recipes/recipes/deployments/kubernetes-blue-green-canary-deployments/): Implement blue-green and canary deployment strategies on Kubernetes. Zero-downtime releases using Service label switching, traffic splitting, progressive
- [Blue-Green Deployment in Kubernetes](https://kubernetes.recipes/recipes/deployments/kubernetes-blue-green-deployment/): Implement blue-green deployments in Kubernetes for instant rollback. Covers Service selector switching, Argo Rollouts blue-green, and comparison with canary
- [Calico NetworkPolicy K8s Guide](https://kubernetes.recipes/recipes/networking/kubernetes-calico-networkpolicy/): Configure Calico NetworkPolicy for K8s. GlobalNetworkPolicy, host endpoints, application layer policies, and DNS policy rules.
- [Canary Deployments with Flagger](https://kubernetes.recipes/recipes/deployments/kubernetes-canary-deployment-flagger/): Automate canary deployments in Kubernetes using Flagger with Istio, Linkerd, or NGINX ingress. Progressive traffic shifting, metric analysis.
- [Canary Deployment in Kubernetes: Step by Step](https://kubernetes.recipes/recipes/deployments/kubernetes-canary-deployment-guide/): Canary deployments in Kubernetes: native replica-ratio canary, Gateway API and Istio weights, header routing, Argo Rollouts analysis and automated rollback.
- [Certificate Expiration Management K8s](https://kubernetes.recipes/recipes/security/kubernetes-cert-expiration-management/): Monitor and manage Kubernetes certificate expiration. kubeadm cert check, cert-manager alerts, auto-renewal, and preventing expired certificate outages.
- [cert-manager Advanced Configuration](https://kubernetes.recipes/recipes/security/kubernetes-cert-manager-advanced/): Advanced cert-manager patterns for Kubernetes. Wildcard certificates, DNS-01 challenges, certificate rotation, cross-namespace sharing.
- [cert-manager: Automated TLS Certificates](https://kubernetes.recipes/recipes/security/kubernetes-cert-manager-guide/): Automate TLS certificate management with cert-manager in Kubernetes. Let's Encrypt integration, Issuer configuration, wildcard certificates, and automatic
- [K8s Certificate Rotation and Management](https://kubernetes.recipes/recipes/security/kubernetes-certificate-management/): Manage Kubernetes cluster certificates with kubeadm. Check expiration, renew certificates, configure auto-rotation, and troubleshoot TLS errors.
- [Kubernetes Certificate Signing Requests](https://kubernetes.recipes/recipes/security/kubernetes-certificate-signing-request/): Use the Kubernetes CSR API to issue, approve, and manage TLS certificates. Automate certificate workflows for services, users, and kubelet rotation.
- [K8s Change Mgmt for Enterprise Operations](https://kubernetes.recipes/recipes/configuration/kubernetes-change-management-enterprise/): Implement ITIL-aligned change management for Kubernetes with approval gates, maintenance windows, rollback procedures, and change audit trails.
- [Kubernetes CI/CD Pipeline with GitHub Actions](https://kubernetes.recipes/recipes/deployments/kubernetes-ci-cd-pipeline/): Build a complete CI/CD pipeline for Kubernetes with GitHub Actions. Covers Docker build, image push, Helm deploy, and automated rollback on failure.
- [Cilium Network Policies Kubernetes](https://kubernetes.recipes/recipes/security/kubernetes-cilium-network-policies/): Advanced network policies with Cilium on Kubernetes. L7 HTTP-aware policies, DNS-based egress, identity-based security, cluster-wide policies.
- [Cilium: eBPF-Powered K8s Networking](https://kubernetes.recipes/recipes/networking/kubernetes-cilium-networking-guide/): Deploy Cilium CNI in Kubernetes for eBPF-based networking, network policies, service mesh, and observability with Hubble.
- [Cluster API: Declarative K8s Management](https://kubernetes.recipes/recipes/configuration/kubernetes-cluster-api-guide/): Manage Kubernetes cluster lifecycle with Cluster API. Provision, upgrade, and scale clusters declaratively using management and infrastructure providers.
- [Advanced Cluster Autoscaler Config & Tuning](https://kubernetes.recipes/recipes/autoscaling/kubernetes-cluster-autoscaler-advanced/): Fine-tune the Kubernetes Cluster Autoscaler with expanders, priority-based scaling, mixed instance policies, and GPU node pool autoscaling for production c.
- [Kubernetes Cluster Autoscaler: Setup and Tuning](https://kubernetes.recipes/recipes/autoscaling/kubernetes-cluster-autoscaler-configuration/): Install and tune Kubernetes Cluster Autoscaler on EKS, GKE and AKS: scale-down flags, expanders, max-node-provision-time, GPU scale-to-zero.
- [Kubernetes Cluster Upgrade with kubeadm: Step-by-Step](https://kubernetes.recipes/recipes/configuration/kubernetes-cluster-upgrade/): Upgrade Kubernetes safely: version skew, etcd backup, kubeadm control plane and worker upgrades, drain/uncordon, EKS/AKS/GKE commands, and rollback.
- [Kubernetes ClusterIP Service Explained](https://kubernetes.recipes/recipes/networking/kubernetes-clusterip-service/): Understand Kubernetes ClusterIP services for internal communication. How kube-proxy routes traffic, DNS resolution, and when ClusterIP is the right service
- [Kubernetes CNI Comparison: Cilium vs Calico vs Flannel](https://kubernetes.recipes/recipes/networking/kubernetes-cni-comparison-guide/): Compare Kubernetes CNI plugins in 2026: Cilium, Calico, Flannel, OVN-Kubernetes and Multus. Dataplane, NetworkPolicy, encryption, BGP, and how to choose.
- [Essential Kubernetes Commands Reference](https://kubernetes.recipes/recipes/configuration/kubernetes-commands/): Master the most used Kubernetes commands for daily operations. Complete kubectl reference for pods, deployments, services, debugging, and cluster management.
- [Create ConfigMap from File in Kubernetes](https://kubernetes.recipes/recipes/configuration/kubernetes-configmap-from-file/): Create Kubernetes ConfigMaps from files, directories, and env files with kubectl. Mount as volumes or inject as environment variables in pods.
- [Kubernetes ConfigMap: Create, Mount, Update](https://kubernetes.recipes/recipes/configuration/kubernetes-configmap-guide/): Create Kubernetes ConfigMaps from files, literals and directories, mount them as volumes or env vars, update them safely, and apply ConfigMap best practices.
- [K8s ConfigMap Hot Reload Without Restart](https://kubernetes.recipes/recipes/configuration/kubernetes-configmap-hot-reload/): Reload Kubernetes ConfigMaps without pod restarts, and fix ConfigMap changes not applied to pods: volume auto-update, subPath, Reloader, checksums.
- [ConfigMap Reload Patterns Kubernetes](https://kubernetes.recipes/recipes/configuration/kubernetes-configmap-reload-patterns/): Implement automatic ConfigMap reload in Kubernetes using volume projection, Reloader operator, checksum annotations, and inotify sidecars.
- [Immutable ConfigMaps and Secrets](https://kubernetes.recipes/recipes/configuration/kubernetes-configmap-secret-immutable/): Use immutable ConfigMaps and Secrets for performance and safety in Kubernetes. Reduce API server load, prevent accidental changes.
- [ConfigMap subPath Update Fix K8s](https://kubernetes.recipes/recipes/configuration/kubernetes-configmap-subpath-updates/): Handle ConfigMap subPath mount limitations in Kubernetes. Why subPath mounts don't auto-update, workarounds, and alternative patterns.
- [Kubernetes startupProbe Configuration Guide](https://kubernetes.recipes/recipes/configuration/kubernetes-container-probes-startup/): Configure startupProbe for slow-starting containers to prevent premature kills. Understand interaction with liveness and readiness probes.
- [Container Runtime Comparison K8s](https://kubernetes.recipes/recipes/configuration/kubernetes-container-runtime-comparison/): Compare Kubernetes container runtimes: containerd vs CRI-O vs Kata Containers. Performance, security, and use cases for each runtime in production.
- [K8s Container Runtimes: containerd vs CRI-O](https://kubernetes.recipes/recipes/configuration/kubernetes-container-runtime-guide/): Compare Kubernetes container runtimes containerd and CRI-O. Configuration, crictl debugging, runtime class for gVisor and Kata, and migration from Docker.
- [Continuous Profiling with Pyroscope](https://kubernetes.recipes/recipes/observability/kubernetes-continuous-profiling-pyroscope/): Deploy Pyroscope on Kubernetes for continuous CPU and memory profiling. Identify performance bottlenecks in production without overhead.
- [CoreDNS Custom Config Kubernetes](https://kubernetes.recipes/recipes/networking/kubernetes-coredns-custom-config/): Customize CoreDNS on Kubernetes for advanced DNS needs. Forward zones, stub domains, custom records, caching tuning, and DNS debugging.
- [CoreDNS Customization Guide Kubernetes](https://kubernetes.recipes/recipes/networking/kubernetes-coredns-customization-guide/): Customize CoreDNS with forward zones, rewrite rules, cache tuning, and stub domains. Troubleshoot DNS resolution failures and optimize query performance in.
- [Cosign Image Signing Kubernetes](https://kubernetes.recipes/recipes/security/kubernetes-cosign-image-verification/): Verify container image signatures with Cosign and Sigstore on Kubernetes. Policy enforcement with Kyverno, supply chain security, and SBOM attestation.
- [Kubernetes Cost Monitoring with Kubecost](https://kubernetes.recipes/recipes/observability/kubernetes-cost-monitoring-kubecost/): Monitor and optimize Kubernetes costs with Kubecost. Track per-namespace and per-deployment spend with cloud billing integration and savings tips.
- [How to Optimize Kubernetes Costs](https://kubernetes.recipes/recipes/configuration/kubernetes-cost-optimization/): Reduce cloud costs in Kubernetes clusters. Right-size resources, use spot instances, implement autoscaling, and monitor spending effectively.
- [Kubernetes Cost Optimization Strategies](https://kubernetes.recipes/recipes/configuration/kubernetes-cost-optimization-strategies/): Comprehensive cost reduction strategies for Kubernetes clusters: right-sizing, spot instances, autoscaling, idle resource detection, namespace budgets, and GPU
- [CRD Development Kubernetes Guide](https://kubernetes.recipes/recipes/configuration/kubernetes-crd-development-guide/): Design and implement Kubernetes Custom Resource Definitions. Schema validation, status subresource, printer columns, conversion webhooks.
- [K8s Custom Resources: CRD Development](https://kubernetes.recipes/recipes/configuration/kubernetes-crd-guide/): Create Kubernetes Custom Resource Definitions with schema validation, additional printer columns, subresources, and conversion webhooks.
- [Fix CreateContainerError in Kubernetes](https://kubernetes.recipes/recipes/troubleshooting/kubernetes-createcontainererror-troubleshoot/): Troubleshoot Kubernetes CreateContainerError with step-by-step debugging. ConfigMap mounts, Secret references, volume permissions, and container runtime issues.
- [Kubernetes CronJob: Schedule, Timezone, Best Practices](https://kubernetes.recipes/recipes/deployments/kubernetes-cronjob-best-practices/): Kubernetes CronJob guide: cron syntax, timeZone, concurrencyPolicy, startingDeadlineSeconds, history and TTL cleanup, manual runs, suspend and monitoring.
- [Crossplane Infrastructure as Code on Kubernetes](https://kubernetes.recipes/recipes/configuration/kubernetes-crossplane-infrastructure/): Provision AWS, GCP and Azure resources from Kubernetes with Crossplane: providers, ProviderConfig, XRDs, pipeline Compositions, claims and GitOps.
- [Build Custom CSI Drivers Kubernetes](https://kubernetes.recipes/recipes/storage/kubernetes-csi-driver-development/): Develop custom Container Storage Interface drivers for Kubernetes. CSI spec, controller and node plugins, volume lifecycle, and testing with csi-sanity.
- [K8s CSI Drivers: Container Storage Guide](https://kubernetes.recipes/recipes/storage/kubernetes-csi-driver-guide/): Install and configure Kubernetes CSI drivers for persistent storage. CSI architecture, StorageClass provisioners, snapshots, and volume expansion patterns.
- [Kubernetes CSI VolumeSnapshot: Create and Restore](https://kubernetes.recipes/recipes/storage/kubernetes-csi-snapshots-restore/): Create and restore Kubernetes CSI volume snapshots: snapshot controller, VolumeSnapshotClass, restore to a new PVC, PVC cloning, scheduled snapshots.
- [Custom Metrics Autoscaling in Kubernetes](https://kubernetes.recipes/recipes/autoscaling/kubernetes-custom-metrics-autoscaling/): Scale Kubernetes pods on custom application metrics with Prometheus Adapter. Configure HPA with custom and external metrics beyond CPU and memory.
- [Custom Metrics with Prometheus Adapter](https://kubernetes.recipes/recipes/autoscaling/kubernetes-custom-metrics-prometheus/): Expose application metrics to Kubernetes HPA via Prometheus Adapter. Configure custom.metrics.k8s.io for HTTP requests per second, queue depth.
- [Custom Scheduler Kubernetes Guide](https://kubernetes.recipes/recipes/deployments/kubernetes-custom-scheduler-guide/): Build and deploy custom Kubernetes schedulers for specialized workloads. Scheduler profiles, extender webhooks, scoring plugins.
- [Kubernetes DaemonSet: One Pod Per Node Guide](https://kubernetes.recipes/recipes/deployments/kubernetes-daemonset-guide/): Kubernetes DaemonSet: run one pod on every node (or a subset). Log collectors, node-exporter, GPU nodes, control-plane tolerations, maxSurge updates.
- [DaemonSet Update Strategies Kubernetes](https://kubernetes.recipes/recipes/deployments/kubernetes-daemonset-update-strategies/): Configure DaemonSet rolling updates with maxUnavailable, OnDelete strategy, partition rollouts, and canary updates for node-level workloads like log collec.
- [Kubernetes DaemonSet Update Strategies](https://kubernetes.recipes/recipes/deployments/kubernetes-daemonset-update-strategy/): Configure DaemonSet rolling updates with maxUnavailable and maxSurge. Understand OnDelete vs RollingUpdate strategies for node-level workloads.
- [Dapr: Microservice Building Blocks on K8s](https://kubernetes.recipes/recipes/deployments/kubernetes-dapr-microservices-guide/): Deploy Dapr in Kubernetes for service invocation, state management, pub/sub messaging, and secrets using a language-agnostic sidecar architecture.
- [Kubernetes Day 2: Where the Leverage Kicks In](https://kubernetes.recipes/recipes/deployments/kubernetes-day-2-operations-leverage/): Why Kubernetes pays off after initial setup. Day 2 operations leverage: auto-scaling, self-healing, rolling updates, observability.
- [Debug Kubernetes Pods: Complete Guide](https://kubernetes.recipes/recipes/troubleshooting/kubernetes-debug-pods/): Debug Kubernetes pods with kubectl debug, ephemeral containers, and netshoot. Troubleshoot distroless images, network issues, and crashed pods step by step.
- [Kubernetes Debugging Toolkit and Commands](https://kubernetes.recipes/recipes/troubleshooting/kubernetes-debugging-toolkit/): Essential kubectl debugging commands and tools for Kubernetes troubleshooting. Covers ephemeral containers, debug pods, network debugging, and log analysis.
- [Deploy a New App in 5 Minutes on Kubernetes](https://kubernetes.recipes/recipes/deployments/kubernetes-deploy-new-app-five-minutes/): Deploy a production-ready application in 5 minutes on an existing Kubernetes cluster. Deployment, Service, Ingress, TLS, autoscaling.
- [Kubernetes Deployment Complete Guide](https://kubernetes.recipes/recipes/deployments/kubernetes-deployment-guide/): Create and manage Kubernetes Deployments for stateless applications. Covers replicas, selectors, rolling updates, rollback, and deployment strategies.
- [Kubernetes Disaster Recovery for Enterprise](https://kubernetes.recipes/recipes/configuration/kubernetes-disaster-recovery-enterprise/): Kubernetes disaster recovery with Velero backup and restore. Cross-region replication, etcd snapshots, multi-cluster failover, and RTO/RPO strategies.
- [Kubernetes Disaster Recovery Planning](https://kubernetes.recipes/recipes/configuration/kubernetes-disaster-recovery-plan/): Build a Kubernetes disaster recovery plan with etcd backups, Velero, cross-region replication, and RTO/RPO targets for production clusters.
- [Kubernetes DNS Configuration: CoreDNS and dnsPolicy](https://kubernetes.recipes/recipes/networking/kubernetes-dns-configuration/): Configure Kubernetes DNS: service/pod record formats, dnsPolicy and dnsConfig, CoreDNS forwarding, stub domains, static hosts, ndots, and NodeLocal DNSCache.
- [DNS Debugging Kubernetes Guide](https://kubernetes.recipes/recipes/troubleshooting/kubernetes-dns-debugging-guide/): Debug Kubernetes DNS issues systematically. CoreDNS troubleshooting, ndots configuration, search domains, and resolving slow DNS lookups.
- [Kubernetes DNS: How Service Discovery Works](https://kubernetes.recipes/recipes/networking/kubernetes-dns-how-it-works/): Understand Kubernetes DNS resolution with CoreDNS. Service discovery, pod DNS, headless services, custom DNS policies, and troubleshooting DNS failures.
- [K8s DNS for Services: Resolution Guide](https://kubernetes.recipes/recipes/networking/kubernetes-dns-services-guide/): Understand Kubernetes DNS for Services and Pods. Service discovery patterns, FQDN format, headless services, DNS policies, ndots configuration.
- [Kubernetes DNS Policy ClusterFirstWithHostNet](https://kubernetes.recipes/recipes/networking/kubernetes-dnspolicy-clusterfirstwithhostnet/): Configure Kubernetes DNS policies: ClusterFirst, ClusterFirstWithHostNet, Default, and None. Fix DNS resolution for hostNetwork pods and custom nameservers.
- [Docker Registry Secret kubectl](https://kubernetes.recipes/recipes/configuration/kubernetes-docker-registry-secret/): Create Kubernetes docker-registry secrets with kubectl. --docker-password-stdin, .dockerconfigjson format, and automating registry authentication.
- [Kubernetes Downward API: Complete Guide](https://kubernetes.recipes/recipes/configuration/kubernetes-downward-api-guide/): Expose pod and container metadata to applications using the Downward API. Environment variables, volume files, fieldRef, resourceFieldRef, and common patterns.
- [Kubernetes Downward API: Pod Metadata in Env](https://kubernetes.recipes/recipes/configuration/kubernetes-downward-api-pod-metadata/): Expose pod metadata to containers using Kubernetes Downward API. Access pod name, namespace, node name, labels, annotations.
- [Dynamic Resource Allocation for GPUs](https://kubernetes.recipes/recipes/ai/kubernetes-dynamic-resource-allocation-gpu/): Use Kubernetes Dynamic Resource Allocation to schedule GPUs. DRA ResourceClaims, partitionable devices, GPU sharing, and structured parameters for accelerators.
- [Logging in Kubernetes with the EFK Stack](https://kubernetes.recipes/recipes/observability/kubernetes-efk-logging-stack/): Set up EFK logging on Kubernetes: Elasticsearch (ECK or StatefulSet), Fluentd DaemonSet with CRI parsing and K8s metadata, Kibana, retention and KQL.
- [K8s Volumes: emptyDir and hostPath Guide](https://kubernetes.recipes/recipes/storage/kubernetes-emptydir-hostpath-volumes/): Configure Kubernetes emptyDir and hostPath volumes for temporary storage and host filesystem access. Memory-backed tmpfs, size limits.
- [emptyDir tmpfs Kubernetes Guide](https://kubernetes.recipes/recipes/storage/kubernetes-emptydir-tmpfs-guide/): Configure emptyDir volumes with memory-backed tmpfs on Kubernetes. Size limits, memory accounting, sidecar sharing, and ephemeral cache patterns.
- [Kubernetes EndpointSlices Explained](https://kubernetes.recipes/recipes/networking/kubernetes-endpoint-slices/): Understand Kubernetes EndpointSlices for scalable service endpoint management. How they improve on Endpoints objects for large clusters with thousands of pods.
- [K8s EndpointSlice and Service Discovery](https://kubernetes.recipes/recipes/networking/kubernetes-endpoint-slices-discovery/): Understand Kubernetes EndpointSlice for scalable service discovery. DNS resolution, headless services, external services, and endpoint conditions.
- [EndpointSlices and Service Topology](https://kubernetes.recipes/recipes/networking/kubernetes-endpoint-slices-service-topology/): Understand EndpointSlices for scalable service discovery in Kubernetes. Covers topology-aware routing and traffic localization for large clusters.
- [EndpointSlice Service Discovery](https://kubernetes.recipes/recipes/networking/kubernetes-endpointslice-service-discovery/): Understand Kubernetes EndpointSlices for scalable service discovery. Compare with legacy Endpoints and configure topology-aware routing.
- [How to Use Kubernetes EndpointSlices](https://kubernetes.recipes/recipes/networking/kubernetes-endpointslices/): Understand and manage EndpointSlices for scalable service discovery. Configure endpoint slicing, troubleshoot connectivity, and optimize large clusters.
- [K8s Capacity Planning for Enterprise Clusters](https://kubernetes.recipes/recipes/autoscaling/kubernetes-enterprise-capacity-planning/): Right-size enterprise clusters with data-driven capacity planning. Forecast resource needs, optimize bin-packing, and plan for growth with metrics.
- [Enterprise GitOps at Scale with Fleet Mgmt](https://kubernetes.recipes/recipes/deployments/kubernetes-enterprise-gitops-fleet/): Manage hundreds of Kubernetes clusters with ArgoCD ApplicationSets, Flux multi-cluster, and fleet-wide policy enforcement using GitOps principles.
- [Enterprise Container Image Governance](https://kubernetes.recipes/recipes/security/kubernetes-enterprise-image-governance/): Enforce image policies with admission controllers. Require signed images, block public registries, and automate vulnerability scanning gates.
- [Automated Secret Rotation on Kubernetes](https://kubernetes.recipes/recipes/security/kubernetes-enterprise-secret-rotation/): Implement zero-downtime secret rotation with External Secrets Operator, HashiCorp Vault dynamic secrets, and rolling restarts for enterprise compliance.
- [Enterprise Service Mesh mTLS & Observability](https://kubernetes.recipes/recipes/networking/kubernetes-enterprise-service-mesh-mtls/): Deploy Istio service mesh for enterprise mTLS, traffic management, circuit breaking, and distributed tracing across microservices on Kubernetes.
- [Env Variables from ConfigMap K8s](https://kubernetes.recipes/recipes/configuration/kubernetes-env-configmap-secrets/): Inject environment variables from ConfigMaps and Secrets in Kubernetes. envFrom, valueFrom, configMapKeyRef, and secretKeyRef patterns.
- [Kubernetes EnvFrom ConfigMap Environment Variables](https://kubernetes.recipes/recipes/configuration/kubernetes-envfrom-configmap-environment-variables/): Inject all ConfigMap keys as environment variables using envFrom in Kubernetes pods. Configure configMapRef, secretRef, prefix options, and selective key
- [envFrom ConfigMapRef Kubernetes](https://kubernetes.recipes/recipes/configuration/kubernetes-envfrom-configmapref/): Inject all ConfigMap keys as environment variables using envFrom configMapRef in Kubernetes. Bulk injection, prefix, and selective key patterns.
- [Kubernetes Environment Variables Guide](https://kubernetes.recipes/recipes/configuration/kubernetes-environment-variables/): Set Kubernetes environment variables with envFrom, configMapRef, secretKeyRef, and the Downward API. Variable ordering, fieldRef, and best practices.
- [Kubernetes Ephemeral Containers: kubectl debug Guide](https://kubernetes.recipes/recipes/troubleshooting/kubernetes-ephemeral-containers-debugging/): Debug running and distroless pods with kubectl debug ephemeral containers: --target, --copy-to for CrashLoopBackOff, node debugging, profiles and RBAC.
- [Ephemeral Storage Management Guide](https://kubernetes.recipes/recipes/storage/kubernetes-ephemeral-storage-management/): Manage ephemeral storage in Kubernetes with emptyDir size limits, ephemeral-storage requests and limits, and eviction thresholds.
- [Generic Ephemeral Volumes in Kubernetes](https://kubernetes.recipes/recipes/storage/kubernetes-ephemeral-volumes-generic/): Use generic ephemeral volumes for per-pod temporary storage with CSI driver features. Scratch space, caching, and temp data without pre-provisioned PVCs.
- [etcd Deep Dive: K8s Data Store Operations](https://kubernetes.recipes/recipes/configuration/kubernetes-etcd-deep-dive-guide/): Master etcd operations for Kubernetes. Backup and restore, compaction, defragmentation, health checks, member management, and performance tuning for production.
- [etcd Maintenance Operations Kubernetes](https://kubernetes.recipes/recipes/storage/kubernetes-etcd-maintenance-operations/): Perform etcd maintenance for Kubernetes clusters. Defragmentation, compaction, snapshot backup, member health checks, and performance monitoring with etcdctl.
- [Kubernetes etcd Operations and Maintenance](https://kubernetes.recipes/recipes/configuration/kubernetes-etcd-operations-guide/): Manage etcd for Kubernetes: backup, restore, compaction, defragmentation, member management, and disaster recovery procedures.
- [etcd Performance Tuning Kubernetes](https://kubernetes.recipes/recipes/configuration/kubernetes-etcd-performance-tuning/): Tune etcd for Kubernetes cluster performance. Disk IOPS requirements, compaction, defragmentation, and monitoring etcd health metrics.
- [How to Use Kubernetes Events for Monitoring](https://kubernetes.recipes/recipes/observability/kubernetes-events-monitoring/): Monitor cluster activity through Kubernetes events. Capture, filter, and alert on events for troubleshooting and operational visibility.
- [kubectl exec: Run Commands Inside K8s Pods](https://kubernetes.recipes/recipes/troubleshooting/kubernetes-exec-into-pod/): Use kubectl exec to run commands inside Kubernetes pods. Covers interactive sessions, multi-container pods, and ephemeral container debugging.
- [Extended Resources & RDMA Shared Device Plugin](https://kubernetes.recipes/recipes/networking/kubernetes-extended-resources-rdma-shared-device/): Kubernetes extended resources for RDMA devices using the shared device plugin. Advertise and schedule InfiniBand and RoCE NICs without SR-IOV using k8s-rdm.
- [ExternalDNS Automation Kubernetes](https://kubernetes.recipes/recipes/networking/kubernetes-external-dns-automation/): Automate DNS record management with ExternalDNS on Kubernetes. Route53, CloudDNS, and Azure DNS integration for Ingress, Service, and Gateway resources.
- [Falco Rules for Kubernetes: Complete Guide](https://kubernetes.recipes/recipes/security/kubernetes-falco-rules-guide/): Write custom Falco rules for K8s runtime security. Syscall detection, container escape alerts, and cryptomining detection.
- [Flux Sources Config Kubernetes](https://kubernetes.recipes/recipes/deployments/kubernetes-flux-sources-guide/): Configure Flux source controllers for GitOps on Kubernetes. GitRepository, HelmRepository, OCIRepository, and Bucket sources for multi-source deployments.
- [fsGroupChangePolicy OnRootMismatch: Fix Slow Mounts](https://kubernetes.recipes/recipes/storage/kubernetes-fsgroupchangepolicy/): Fix slow pod starts from recursive fsGroup chown: set fsGroupChangePolicy OnRootMismatch, check CSI fsGroupPolicy, and handle OpenShift SCC fsGroup.
- [Kubernetes Gateway API: HTTPRoute Examples](https://kubernetes.recipes/recipes/networking/kubernetes-gateway-api/): Gateway API examples: GatewayClass, Gateway listeners, HTTPRoute, GRPCRoute, TLSRoute, canary weights, HTTPS redirect, ReferenceGrant, Ingress migration.
- [Gateway API gRPC Routes on Kubernetes](https://kubernetes.recipes/recipes/networking/kubernetes-gateway-api-grpc-routes/): Configure Kubernetes Gateway API GRPCRoute for gRPC traffic routing. Service-level matching, header-based routing, and traffic splitting for gRPC services.
- [Gateway API HTTPRoute Kubernetes](https://kubernetes.recipes/recipes/networking/kubernetes-gateway-api-httproute/): Configure HTTPRoute for Kubernetes Gateway API. Path matching, header-based routing, traffic splitting, URL rewriting, and request mirroring.
- [K8s Golden Signals: SLI and SLO Monitoring](https://kubernetes.recipes/recipes/observability/kubernetes-golden-signals-sli-slo/): Implement Google SRE golden signals on Kubernetes. Define SLIs, set SLO targets, configure error budgets, and build SLO dashboards with Prometheus and Sloth.
- [Goldilocks VPA Dashboard for Resource Optimization](https://kubernetes.recipes/recipes/autoscaling/kubernetes-goldilocks-vpa-dashboard/): Deploy Goldilocks to visualize VPA recommendations across all workloads and identify over-provisioned or under-provisioned containers with actionable
- [K8s GPU Limits Requests Configuration](https://kubernetes.recipes/recipes/ai/kubernetes-gpu-limits-requests/): Configure GPU resource limits and requests in Kubernetes pod specs. nvidia.com/gpu resource, fractional GPUs, MIG slices, and multi-GPU allocation.
- [GPU Node Provisioning Kubernetes](https://kubernetes.recipes/recipes/ai/kubernetes-gpu-node-provisioning/): Automate GPU node provisioning for Kubernetes with Karpenter, Cluster Autoscaler, and cloud-specific node pools for AI and ML workloads.
- [GPU Operator Advanced Configuration](https://kubernetes.recipes/recipes/ai/kubernetes-gpu-operator-advanced-config/): Advanced NVIDIA GPU Operator configuration on Kubernetes. Driver containers, CUDA toolkit, GDS, GPUDirect RDMA, MIG manager, DCGM Exporter.
- [GPU Sharing on Kubernetes: Time-Slicing vs MIG vs MPS](https://kubernetes.recipes/recipes/ai/kubernetes-gpu-sharing-mps-mig/): Compare NVIDIA GPU sharing on Kubernetes: time-slicing vs MIG vs MPS vs full GPU. Isolation, supported GPUs, GPU Operator config, and when to use each.
- [K8s Graceful Shutdown & terminationGracePeriod](https://kubernetes.recipes/recipes/deployments/kubernetes-graceful-shutdown-guide/): Kubernetes graceful shutdown: terminationGracePeriodSeconds (default 30s), preStop sleep, SIGTERM handlers, and fixing the endpoint-removal race for zero 502s.
- [Grafana Dashboards for Kubernetes Guide](https://kubernetes.recipes/recipes/observability/kubernetes-grafana-dashboards-guide/): Import and customize Grafana dashboards for Kubernetes monitoring. Dashboard 315, 6417, kube-prometheus-stack, and custom panel creation.
- [Sandboxed Runtimes on k3s: gVisor and Kata with RuntimeClass](https://kubernetes.recipes/recipes/security/kubernetes-gvisor-kata-containers-runtimeclass/): Deploy sandboxed container runtimes on k3s and Kubernetes with RuntimeClass: gVisor (runsc), Kata Containers, WASM, and alternatives. Isolate untrusted
- [Harbor: Private Container Registry on K8s](https://kubernetes.recipes/recipes/security/kubernetes-harbor-registry-guide/): Deploy Harbor container registry in Kubernetes for private image hosting. Vulnerability scanning, image replication, RBAC, Helm chart repository.
- [Kubernetes Headless Service Explained](https://kubernetes.recipes/recipes/networking/kubernetes-headless-service/): Create Kubernetes headless services for StatefulSet DNS, direct pod addressing, and service discovery. Understand when clusterIP None is the right choice.
- [Kubernetes Health Checks Best Practices](https://kubernetes.recipes/recipes/configuration/kubernetes-health-checks/): Design effective Kubernetes health checks with liveness, readiness, and startup probes. Avoid common anti-patterns like database checks in liveness probes.
- [Helm Chart Testing CI/CD Guide](https://kubernetes.recipes/recipes/helm/kubernetes-helm-chart-testing/): Test Helm charts with helm test, helm lint, chart-testing, and conftest. Unit tests, integration tests, and CI/CD pipeline integration for chart quality.
- [Helm vs Kustomize: Which to Use](https://kubernetes.recipes/recipes/helm/kubernetes-helm-vs-kustomize/): Compare Helm and Kustomize for Kubernetes configuration management. Covers templating vs overlays, use cases, pros and cons, and when to use both together.
- [DNS Autoscaling and CoreDNS Scaling](https://kubernetes.recipes/recipes/networking/kubernetes-horizontal-dns-autoscaling/): Scale CoreDNS horizontally with dns-autoscaler and proportional autoscaling. Tune cache size, configure node-local DNS cache.
- [HPA v2 Multiple Metrics Scaling Guide](https://kubernetes.recipes/recipes/autoscaling/kubernetes-horizontal-pod-autoscaler-v2/): Configure HorizontalPodAutoscaler v2 with CPU, memory, custom, and external metrics. Control scaling behavior with stabilization windows.
- [K8s Horizontal Scaling: Manual and Auto](https://kubernetes.recipes/recipes/autoscaling/kubernetes-horizontal-scaling-patterns/): Scale Kubernetes workloads horizontally with kubectl scale, HPA, and KEDA. Covers replica management and event-driven scaling strategies.
- [Kubernetes hostPath Volume Guide](https://kubernetes.recipes/recipes/storage/kubernetes-hostpath-volume/): Use hostPath volumes to mount node filesystem paths into pods. Types, security risks, use cases for DaemonSets, and safer alternatives like local PVs.
- [hostPath vs PVC Kubernetes Guide](https://kubernetes.recipes/recipes/storage/kubernetes-hostpath-vs-pvc/): Compare hostPath and PVC storage options for Kubernetes. Security risks of hostPath, node affinity constraints, and when to use each storage type.
- [HPA Behavior and Scaling Policies](https://kubernetes.recipes/recipes/autoscaling/kubernetes-hpa-behavior-scaling-policies/): Configure HPA scaling behavior with stabilization windows, scaling policies, and rate limiting. Fine-tune scale-up and scale-down speed.
- [HPA Container Resource Metrics](https://kubernetes.recipes/recipes/autoscaling/kubernetes-hpa-container-resource-metrics/): Configure HPA to scale based on individual container metrics instead of pod-level averages. Target specific containers in multi-container pods.
- [K8s HPA: Autoscale on CPU and Memory](https://kubernetes.recipes/recipes/autoscaling/kubernetes-hpa-cpu-memory-guide/): Configure Kubernetes HorizontalPodAutoscaler to scale on CPU and memory utilization. Target utilization, minReplicas, maxReplicas, and scaling behavior.
- [HPA Custom Metrics Scaling Guide](https://kubernetes.recipes/recipes/autoscaling/kubernetes-hpa-custom-metrics-guide/): Scale Kubernetes workloads on custom Prometheus metrics with HPA. Prometheus Adapter, external metrics, and request-rate-based scaling for web services.
- [Kubernetes HPA Custom Metrics with Prometheus](https://kubernetes.recipes/recipes/autoscaling/kubernetes-hpa-custom-metrics-prometheus/): Configure Kubernetes HPA with custom Prometheus metrics. Prometheus Adapter setup, custom and external metrics, scaling on request latency, queue depth.
- [Kubernetes HPA Custom Metrics Prometheus Adapter](https://kubernetes.recipes/recipes/autoscaling/kubernetes-hpa-custom-metrics-prometheus-adapter/): Configure Kubernetes Horizontal Pod Autoscaler with custom Prometheus metrics via the Prometheus Adapter. Scale on request latency, queue depth, GPU
- [HPA Prometheus Custom Metrics K8s](https://kubernetes.recipes/recipes/autoscaling/kubernetes-hpa-prometheus-adapter/): Configure HPA with custom Prometheus metrics using prometheus-adapter on Kubernetes. Custom and external metrics, query mapping, and scaling on business KPIs.
- [Kubernetes imagePullPolicy Guide](https://kubernetes.recipes/recipes/configuration/kubernetes-image-pull-policy-always-never/): Configure imagePullPolicy correctly: Always, Never, and IfNotPresent behavior. Understand digest pinning and tag mutability implications.
- [Trivy Image Scanning Kubernetes](https://kubernetes.recipes/recipes/security/kubernetes-image-scanning-trivy/): Scan container images with Trivy on K8s. Admission webhook, CI/CD integration, CIS benchmarks, and vulnerability reporting.
- [imagePullSecrets Pod Config K8s](https://kubernetes.recipes/recipes/configuration/kubernetes-imagepullsecrets-guide/): Configure imagePullSecrets for pulling from private container registries on Kubernetes. Docker registry secrets, service account default.
- [K8s imagePullSecrets: Private Registry Auth](https://kubernetes.recipes/recipes/configuration/kubernetes-imagepullsecrets-private-registry/): Configure imagePullSecrets for pulling container images from private registries. Create docker-registry secrets, attach to pods and ServiceAccounts.
- [Kubernetes Ingress Guide: Routing, TLS, Controllers](https://kubernetes.recipes/recipes/networking/kubernetes-ingress-complete-guide/): Kubernetes Ingress in practice: host/path routing, pathType, TLS with cert-manager, rewrites, canary, auth, NGINX annotations, and Ingress vs Gateway API.
- [Kubernetes Ingress Fundamentals](https://kubernetes.recipes/recipes/networking/kubernetes-ingress-fundamentals/): Configure Kubernetes Ingress for HTTP routing, TLS termination, and path-based routing. NGINX Ingress Controller setup, annotations, and multi-service routing.
- [K8s Ingress NGINX: Routing and TLS](https://kubernetes.recipes/recipes/networking/kubernetes-ingress-nginx-guide/): Configure Kubernetes Ingress with NGINX controller. Path-based routing, TLS termination, annotations, rate limiting, and multiple hosts with examples.
- [Ingress Path Routing Kubernetes](https://kubernetes.recipes/recipes/networking/kubernetes-ingress-path-routing/): Configure Kubernetes Ingress for path-based and host-based routing. PathType Prefix vs Exact, rewrite rules, and multi-service routing patterns.
- [K8s Ingress Rate Limit NGINX Config](https://kubernetes.recipes/recipes/networking/kubernetes-ingress-rate-limit-nginx/): Configure rate limiting on Kubernetes NGINX Ingress. limit-rps, limit-burst-multiplier annotations, per-client limits, and webhook protection patterns.
- [Kubernetes Ingress TLS Certificate with cert-manager](https://kubernetes.recipes/recipes/networking/kubernetes-ingress-tls-cert-manager/): Automate TLS certificate management on Kubernetes with cert-manager. Let's Encrypt integration, ClusterIssuer configuration, automatic renewal, wildcard
- [Kubernetes Init Containers: Patterns and Examples](https://kubernetes.recipes/recipes/deployments/kubernetes-init-containers-patterns-examples/): Kubernetes init containers explained with YAML: wait for dependencies, migrations, git clone, permissions, certs, resource rules, sidecars and debugging.
- [Kubernetes IPPool Management Guide](https://kubernetes.recipes/recipes/networking/kubernetes-ippool-management-guide/): Configure IP address pools in Kubernetes with Whereabouts, NV-IPAM, MetalLB, and Calico IPPool for secondary networks and LoadBalancer IPs.
- [Istio Traffic Management Kubernetes](https://kubernetes.recipes/recipes/networking/kubernetes-istio-traffic-management/): Advanced Istio traffic management on Kubernetes. VirtualService routing, DestinationRule load balancing, traffic mirroring, fault injection.
- [Jaeger Tracing Kubernetes Guide](https://kubernetes.recipes/recipes/observability/kubernetes-jaeger-tracing-guide/): Deploy Jaeger for distributed tracing on Kubernetes. Collector, storage backends, sampling strategies, and trace analysis for microservice debugging.
- [Kubernetes Job Completions and Parallelism](https://kubernetes.recipes/recipes/deployments/kubernetes-job-completion-parallelism/): Kubernetes Job completions vs parallelism with YAML: indexed jobs, work queues, backoffLimit, backoffLimitPerIndex, podFailurePolicy and successPolicy.
- [Job Completion Patterns Kubernetes](https://kubernetes.recipes/recipes/deployments/kubernetes-job-completion-patterns/): Configure Kubernetes Jobs with indexed completions, work queues, parallel processing, backoff limits, and TTL cleanup for batch workloads.
- [Kubernetes Jobs and CronJobs: Complete Guide](https://kubernetes.recipes/recipes/deployments/kubernetes-job-cronjob-guide/): Kubernetes Jobs and CronJobs with YAML: run-to-completion, backoffLimit, parallelism, Indexed jobs, podFailurePolicy, TTL cleanup and cron scheduling.
- [Job TTL Cleanup Kubernetes Guide](https://kubernetes.recipes/recipes/deployments/kubernetes-job-ttl-cleanup/): Automate Kubernetes Job cleanup with TTL controller. ttlSecondsAfterFinished, CronJob history limits, and preventing completed Job accumulation.
- [KEDA: Event-Driven Autoscaling for Kubernetes](https://kubernetes.recipes/recipes/autoscaling/kubernetes-keda-event-driven-autoscaling/): Install KEDA with Helm and scale on Kafka lag, RabbitMQ, Prometheus, cron or HTTP. ScaledObject, ScaledJob, TriggerAuthentication, scale-to-zero.
- [KEDA Scalers Guide for Kubernetes](https://kubernetes.recipes/recipes/autoscaling/kubernetes-keda-scalers-guide/): Configure KEDA scalers for event-driven autoscaling on Kubernetes. Covers Kafka, RabbitMQ, Prometheus, and cron trigger configuration.
- [KIND Local Kubernetes Dev Guide](https://kubernetes.recipes/recipes/deployments/kubernetes-kind-local-development/): Use KIND for local Kubernetes development. Multi-node clusters, ingress setup, load balancer, persistent storage, and CI/CD integration.
- [Kubernetes Kind Local Development Cluster](https://kubernetes.recipes/recipes/configuration/kubernetes-kind-local-development-cluster/): Create local Kubernetes clusters with kind (Kubernetes in Docker). Multi-node clusters, ingress setup, local registry, port mapping, volume mounts, and CI/CD
- [Knative: Serverless Workloads on Kubernetes](https://kubernetes.recipes/recipes/deployments/kubernetes-knative-serverless-guide/): Run serverless containers with Knative Serving and Eventing on Kubernetes. Auto-scaling to zero, traffic splitting, revision management.
- [kubeadm init: Bootstrap K8s Cluster](https://kubernetes.recipes/recipes/configuration/kubernetes-kubeadm-init-guide/): Bootstrap a Kubernetes cluster with kubeadm init and join. Control plane setup, worker node joining, pod network installation.
- [K8s kubeadm Upgrade: Step-by-Step Guide](https://kubernetes.recipes/recipes/configuration/kubernetes-kubeadm-upgrade-guide/): Upgrade Kubernetes clusters with kubeadm from one minor version to the next. Control plane upgrade, worker node drain, kubelet upgrade, and rollback procedures.
- [kubectl debug: Advanced Pod Debugging](https://kubernetes.recipes/recipes/troubleshooting/kubernetes-kubectl-debug-guide/): Use kubectl debug for ephemeral containers, node debugging, and pod copy debugging. Debug distroless images, share process namespaces, and node-level access.
- [kubectl Plugins: Extend with Krew](https://kubernetes.recipes/recipes/configuration/kubernetes-kubectl-plugins-guide/): Install kubectl plugins with Krew package manager. Essential plugins for debugging, resource management, and cluster operations. Build custom kubectl plugins.
- [kubectl wait: Script K8s Operations](https://kubernetes.recipes/recipes/configuration/kubernetes-kubectl-wait-scripting/): Use kubectl wait for scripting Kubernetes operations. Wait for pod ready, job completion, deployment rollout, and custom conditions in CI/CD pipelines.
- [Kubeflow PyTorchJob Training K8s](https://kubernetes.recipes/recipes/ai/kubernetes-kubeflow-pytorchjob/): Run distributed PyTorch training on Kubernetes with Kubeflow PyTorchJob. ElasticPolicy, nproc_per_node, RDMA configuration, and multi-GPU scaling.
- [K8s Kubelet Configuration and Tuning](https://kubernetes.recipes/recipes/configuration/kubernetes-kubelet-configuration/): Configure Kubernetes kubelet with KubeletConfiguration API. Resource reservation, eviction thresholds, image garbage collection, and node allocatable settings.
- [Kubernetes kustomization.yaml Guide](https://kubernetes.recipes/recipes/configuration/kubernetes-kustomization-yaml-guide/): Write kustomization.yaml files for Kubernetes resource management. Overlays, patches, generators, transformers, and multi-environment deployment patterns.
- [Kustomize Advanced Patterns Kubernetes](https://kubernetes.recipes/recipes/configuration/kubernetes-kustomize-advanced-patterns/): Advanced Kustomize patterns for Kubernetes configuration management. Strategic merge patches, JSON patches, components, replacements.
- [Kustomize Guide: Overlays, Patches, Generators](https://kubernetes.recipes/recipes/configuration/kubernetes-kustomize-guide/): Kustomize for Kubernetes: base/overlay layout, kustomization.yaml (kustomize.config.k8s.io/v1beta1), patches, images, replicas, ConfigMap generators, fixes.
- [Kustomize Overlays Guide Kubernetes](https://kubernetes.recipes/recipes/configuration/kubernetes-kustomize-overlays-guide/): Manage Kubernetes manifests with Kustomize overlays. Base and overlay patterns, strategic merge patches, JSON patches, ConfigMap generators.
- [Kyverno: K8s Policy Engine Without Code](https://kubernetes.recipes/recipes/security/kubernetes-kyverno-policy-guide/): Enforce Kubernetes policies with Kyverno. Validate, mutate, and generate resources using YAML policies. Image verification, label enforcement.
- [Kubernetes Labels vs Annotations: Best Practices](https://kubernetes.recipes/recipes/configuration/kubernetes-labels-annotations-best-practices/): Kubernetes labels vs annotations: when to use each, recommended app.kubernetes.io labels, naming rules, common annotations, and selector pitfalls.
- [Kubernetes Labels Best Practices (app.kubernetes.io)](https://kubernetes.recipes/recipes/configuration/kubernetes-labels-best-practices/): Kubernetes labels best practices: recommended app.kubernetes.io labels (name, part-of, component), naming rules, selector patterns, cost-allocation labels.
- [Kubernetes Labels and Selectors Guide](https://kubernetes.recipes/recipes/configuration/kubernetes-labels-selectors-guide/): Kubernetes labels and selectors: kubectl label, -l equality and set-based queries, matchLabels vs matchExpressions, field selectors, recommended labels.
- [Kubernetes Lease Objects & Leader Election](https://kubernetes.recipes/recipes/deployments/kubernetes-leases/): What a Kubernetes Lease is, how the control plane uses it for leader election, and copy-paste client-go code plus kubectl get lease checks for HA controllers.
- [K8s Let's Encrypt Ingress with cert-manager](https://kubernetes.recipes/recipes/security/kubernetes-letsencrypt-ingress-cert-manager/): Automate TLS certificates for Kubernetes Ingress using cert-manager and Let's Encrypt. ClusterIssuer setup, HTTP-01 and DNS-01 challenges, and auto-renewal.
- [Let's Encrypt Ingress Kubernetes](https://kubernetes.recipes/recipes/networking/kubernetes-letsencrypt-ingress-setup/): Set up Let's Encrypt TLS certificates for Kubernetes Ingress with cert-manager. HTTP-01 challenge, automatic renewal, and HTTPS redirect configuration.
- [Kubernetes LimitRange Defaults](https://kubernetes.recipes/recipes/configuration/kubernetes-limit-range-defaults/): Set default resource requests and limits per namespace with LimitRange. Enforce min/max constraints and prevent unbounded resource consumption.
- [Linkerd: Lightweight K8s Service Mesh](https://kubernetes.recipes/recipes/networking/kubernetes-linkerd-service-mesh-guide/): Deploy Linkerd service mesh in Kubernetes for mTLS, traffic splitting, retries, and observability. A lighter, zero-config alternative to Istio.
- [Kubernetes Liveness Probe Best Practices](https://kubernetes.recipes/recipes/configuration/kubernetes-liveness-probe-best-practices/): Configure Kubernetes liveness probes correctly. Best practices for httpGet, exec, and tcpSocket probes. Avoid database checks, thundering herd.
- [Kubernetes Probes Liveness Readiness](https://kubernetes.recipes/recipes/deployments/kubernetes-liveness-readiness-startup-probes/): Configure liveness, readiness, and startup probes in Kubernetes. HTTP, TCP, exec, and gRPC probe types with real-world tuning for production workloads.
- [Kubernetes Load Balancing Strategies](https://kubernetes.recipes/recipes/networking/kubernetes-load-balancing/): Configure Kubernetes load balancing with Services, Ingress, and Gateway API. Round-robin, session affinity, weighted routing, and traffic policy.
- [Kubernetes LoadBalancer Service Guide](https://kubernetes.recipes/recipes/networking/kubernetes-loadbalancer/): Expose Kubernetes services with LoadBalancer type for production traffic. Covers cloud providers, MetalLB for bare-metal, health checks, and cost optimization.
- [K8s Local Development with Minikube and Kind](https://kubernetes.recipes/recipes/deployments/kubernetes-local-development/): Set up local Kubernetes clusters for development with Minikube, Kind, and k3d. Covers installation, configuration, local registries, and hot-reload workflows.
- [Local Persistent Volumes Kubernetes](https://kubernetes.recipes/recipes/storage/kubernetes-local-persistent-volumes/): Configure local persistent volumes on Kubernetes for high-performance storage. Node affinity, local-path-provisioner, and SSD-backed database workloads.
- [Kubernetes Log Aggregation with Grafana Loki](https://kubernetes.recipes/recipes/observability/kubernetes-log-aggregation-loki/): Aggregate Kubernetes logs with Grafana Loki 3.x: Helm install, Alloy collector (Promtail replacement), S3 storage, retention, LogQL queries and log alerts.
- [EFK Stack: Kubernetes Centralized Logging](https://kubernetes.recipes/recipes/observability/kubernetes-logging-elk-stack/): Deploy EFK stack for Kubernetes centralized logging. Elasticsearch, Fluentd, Kibana setup, log collection, parsing, and retention policies.
- [Kubernetes Logging Fluent Bit Guide](https://kubernetes.recipes/recipes/observability/kubernetes-logging-fluentbit-guide/): Deploy Fluent Bit for centralized Kubernetes logging. DaemonSet configuration, parsing, filtering, and forwarding logs to Elasticsearch, Loki, or S3.
- [MetalLB Bare Metal Load Balancer](https://kubernetes.recipes/recipes/networking/kubernetes-metallb-bare-metal-lb/): Deploy MetalLB for LoadBalancer services on bare-metal Kubernetes. L2 mode, BGP mode, IP address pools, and integration with Cilium and Gateway API.
- [K8s Metrics Server: Install and Configure](https://kubernetes.recipes/recipes/observability/kubernetes-metrics-server-setup/): Install and configure Kubernetes Metrics Server for kubectl top, HPA autoscaling, and resource monitoring. Troubleshoot common metrics-server errors and TL.
- [K8s Metrics Server: kubectl top Guide](https://kubernetes.recipes/recipes/observability/kubernetes-metrics-server-top/): Install Kubernetes Metrics Server for kubectl top and HPA. Resource usage monitoring, troubleshooting metrics, and custom metrics integration.
- [K8s Monitoring with Prometheus and Grafana](https://kubernetes.recipes/recipes/observability/kubernetes-monitoring-guide/): Set up Kubernetes monitoring with Prometheus and Grafana. Covers kube-prometheus-stack, custom dashboards, alerting rules, and key metrics to monitor.
- [Multi-Cluster Service Mesh Kubernetes](https://kubernetes.recipes/recipes/networking/kubernetes-multi-cluster-service-mesh/): Connect multiple Kubernetes clusters with service mesh federation. Istio multi-cluster, Linkerd multi-cluster, cross-cluster service discovery.
- [Multi-Container Pod Patterns: Sidecar, Ambassador, Adapter](https://kubernetes.recipes/recipes/deployments/kubernetes-multi-container-patterns/): Kubernetes multi-container pod patterns with YAML: sidecar log shipper, ambassador DB proxy and pooler, adapter metrics/log transformer, native sidecars.
- [Kubernetes Multi-Tenancy Patterns](https://kubernetes.recipes/recipes/security/kubernetes-multi-tenancy/): Implement multi-tenancy in Kubernetes with namespaces, RBAC, quotas, network policies, and virtual clusters. Covers soft and hard tenancy models.
- [Kubernetes Multi-Tenancy for Enterprise Teams](https://kubernetes.recipes/recipes/security/kubernetes-multi-tenancy-enterprise/): Implement secure multi-tenancy with namespace isolation, ResourceQuotas, NetworkPolicies, hierarchical namespaces, and vCluster for strong isolation.
- [Multidimensional Pod Autoscaler (MPA)](https://kubernetes.recipes/recipes/autoscaling/kubernetes-multidimensional-pod-autoscaler/): Configure Google's Multidimensional Pod Autoscaler to scale both horizontally and vertically simultaneously. Combines HPA and VPA logic in one controller.
- [Multi-Tenancy Namespaces Kubernetes](https://kubernetes.recipes/recipes/security/kubernetes-multitenancy-namespaces/): Implement multi-tenancy on Kubernetes with namespaces. Resource quotas, network policies, RBAC isolation, and hierarchical namespaces for team separation.
- [Kubernetes Namespaces: Management and Best Practices](https://kubernetes.recipes/recipes/configuration/kubernetes-namespace-guide/): Create and manage Kubernetes namespaces for teams and environments: naming strategy, ResourceQuota, LimitRange, RBAC, NetworkPolicy and a provisioning template.
- [Namespace Templates: Instant Envs in K8s](https://kubernetes.recipes/recipes/configuration/kubernetes-namespace-template-instant-environments/): Create production-ready namespace templates for instant environment provisioning. One command deploys namespace, RBAC, quotas, network policies, and monitoring.
- [NATS: Lightweight Messaging for Kubernetes](https://kubernetes.recipes/recipes/configuration/kubernetes-nats-messaging-guide/): Deploy NATS messaging in Kubernetes for pub/sub, request/reply, and JetStream streaming. A high-performance alternative to Kafka for cloud-native apps.
- [Network Debugging Tools Kubernetes](https://kubernetes.recipes/recipes/troubleshooting/kubernetes-network-debugging-tools/): Debug Kubernetes networking with tcpdump, netshoot, iptables tracing, conntrack inspection, and DNS resolution testing techniques.
- [Network Observability with Cilium Hubble](https://kubernetes.recipes/recipes/observability/kubernetes-network-observability-cilium-hubble/): Monitor Kubernetes network traffic with Cilium Hubble. Service maps, DNS visibility, HTTP flow logs, network policy auditing, and Hubble UI dashboards.
- [Kubernetes Egress Network Policies](https://kubernetes.recipes/recipes/networking/kubernetes-network-policy-egress/): Control outbound traffic from pods with egress NetworkPolicies. Allow DNS, block internet access, and restrict pod-to-pod communication by namespace.
- [NetworkPolicy Recipes Cookbook K8s](https://kubernetes.recipes/recipes/security/kubernetes-network-policy-recipes/): Common Kubernetes NetworkPolicy recipes. Default deny, allow DNS, namespace isolation, database access, and external egress patterns for zero-trust networking.
- [Kubernetes Security Checklist for Production](https://kubernetes.recipes/recipes/security/kubernetes-network-security-checklist/): Production security checklist for Kubernetes clusters. Covers RBAC, network policies, pod security, secrets encryption, audit logging, and image scanning.
- [How to Troubleshoot Kubernetes Networking](https://kubernetes.recipes/recipes/troubleshooting/kubernetes-network-troubleshooting/): Debug and resolve Kubernetes networking issues systematically. Learn to diagnose DNS problems, service connectivity, network policies, and CNI issues.
- [Kubernetes NetworkPolicy Default Deny Examples](https://kubernetes.recipes/recipes/security/kubernetes-networkpolicy-default-deny/): Kubernetes NetworkPolicy default deny examples for ingress and egress: deny all, allow DNS, ingress controller, namespace isolation and a zero-trust baseline.
- [Kubernetes NetworkPolicy Default Deny Egress](https://kubernetes.recipes/recipes/security/kubernetes-networkpolicy-default-deny-egress/): Implement Kubernetes NetworkPolicy default deny egress rules. Block all outbound traffic, then allow specific destinations: DNS, external APIs.
- [Kubernetes NetworkPolicy Examples and Guide](https://kubernetes.recipes/recipes/networking/kubernetes-networkpolicy-guide/): Copy-paste Kubernetes NetworkPolicy examples: default deny, allow DNS, allow by label or namespace, ingress controller, database access and egress CIDRs.
- [NetworkPolicy Zero Trust Kubernetes](https://kubernetes.recipes/recipes/security/kubernetes-networkpolicy-zero-trust/): Implement zero-trust networking with Kubernetes NetworkPolicies. Default-deny ingress and egress, namespace isolation, DNS egress rules, and Cilium L7 policies.
- [NFS Dynamic Provisioner Kubernetes](https://kubernetes.recipes/recipes/storage/kubernetes-nfs-provisioner-guide/): Deploy NFS dynamic provisioner for ReadWriteMany storage on Kubernetes. NFS CSI driver, StorageClass configuration, and performance tuning with nconnect.
- [K8s Node Affinity and Pod Scheduling](https://kubernetes.recipes/recipes/deployments/kubernetes-node-affinity-guide/): Configure Kubernetes node affinity, pod affinity, and anti-affinity rules. nodeSelector, requiredDuringScheduling, preferredDuringScheduling, and topology.
- [Kubernetes Node Affinity Guide](https://kubernetes.recipes/recipes/configuration/kubernetes-node-affinity-required-preferred/): Schedule pods to specific nodes with requiredDuringScheduling and preferredDuringScheduling node affinity. Control placement with expressions and weights.
- [Node Affinity Scheduling Kubernetes](https://kubernetes.recipes/recipes/deployments/kubernetes-node-affinity-scheduling/): Configure node affinity rules for Kubernetes pod scheduling. Required vs preferred affinity, label selectors, and combining with taints and tolerations.
- [kubectl cordon, drain and uncordon Nodes](https://kubernetes.recipes/recipes/deployments/kubernetes-node-drain-cordon/): Safely cordon, drain and uncordon Kubernetes nodes for maintenance: drain flags, PDB-blocked drains, bare pods, and a maintenance script.
- [Node Maintenance and Drain Operations](https://kubernetes.recipes/recipes/deployments/kubernetes-node-maintenance-drain/): Safely drain Kubernetes nodes for maintenance with cordon, drain, and uncordon. Handle PodDisruptionBudgets, DaemonSets, and local storage.
- [Fix Node NotReady Status in Kubernetes](https://kubernetes.recipes/recipes/troubleshooting/kubernetes-node-notready-troubleshooting/): Troubleshoot Kubernetes nodes in NotReady state. Kubelet issues, disk pressure, network problems, certificate expiration, and recovery procedures.
- [Check Kubernetes Node Status with kubectl](https://kubernetes.recipes/recipes/troubleshooting/kubernetes-node-status-check/): Check and troubleshoot Kubernetes node status with kubectl. Node conditions (Ready, MemoryPressure, DiskPressure), NotReady debugging, and capacity monitoring.
- [Fix Untolerated Taint node-role master](https://kubernetes.recipes/recipes/troubleshooting/kubernetes-node-untolerated-taint-master/): Fix 'node untolerated taint node-role.kubernetes.io/master' scheduling error. Remove or tolerate control plane taints to schedule pods on master nodes.
- [Kubernetes NodePort Service Explained](https://kubernetes.recipes/recipes/networking/kubernetes-nodeport/): Expose Kubernetes services externally with NodePort. Understand port ranges, security implications, and when to use NodePort vs LoadBalancer vs Ingress.
- [Kubernetes OIDC Authentication with kubelogin](https://kubernetes.recipes/recipes/security/kubernetes-oidc-authentication-guide/): Set up OIDC SSO for the Kubernetes API server with Keycloak, Dex, or Azure/Entra ID: API server flags, kubelogin kubeconfig, and RBAC bindings for IdP groups.
- [K8s OIDC Integration with Enterprise SSO](https://kubernetes.recipes/recipes/security/kubernetes-oidc-enterprise-sso/): Configure Kubernetes API server OIDC authentication with Keycloak, Azure AD, or Okta for enterprise single sign-on and group-based RBAC.
- [Kubernetes OOMKilled (Exit Code 137): Debug and Fix](https://kubernetes.recipes/recipes/troubleshooting/kubernetes-oomkilled-troubleshooting/): Fix OOMKilled (exit code 137) in Kubernetes: find the killed container, container vs node OOM, memory limits, JVM/Node.js heap sizing, leaks, and VPA.
- [OPA Gatekeeper Policy Enforcement](https://kubernetes.recipes/recipes/security/kubernetes-opa-gatekeeper-policies/): Enforce policies with OPA Gatekeeper on Kubernetes. ConstraintTemplates, Constraints, dry-run mode, audit, and common policies for security compliance.
- [Kubernetes and OpenShift Major Version Upgrade](https://kubernetes.recipes/recipes/deployments/kubernetes-openshift-major-minor-upgrade/): Upgrade Kubernetes minor versions (1.31→1.32) and OpenShift (4.16→4.17, EUS-to-EUS). API deprecation migration, etcd backup.
- [Kubernetes and OpenShift Patch Updates](https://kubernetes.recipes/recipes/deployments/kubernetes-openshift-patch-update/): Apply patch updates to Kubernetes and OpenShift clusters safely. Patch version upgrades for control plane, kubeadm, kubelet.
- [Kubernetes and OpenShift Upgrade Strategy](https://kubernetes.recipes/recipes/deployments/kubernetes-openshift-upgrade-overview/): Complete upgrade strategy for Kubernetes and OpenShift clusters. Understand patch, minor, and major versions, upgrade paths.
- [OpenTelemetry Auto-Instrumentation](https://kubernetes.recipes/recipes/observability/kubernetes-opentelemetry-auto-instrumentation/): Configure OpenTelemetry Operator auto-instrumentation to inject tracing into pods without code changes. Supports Java, Python, Node.js, .NET, and Go.
- [OpenTelemetry Collector Kubernetes](https://kubernetes.recipes/recipes/observability/kubernetes-opentelemetry-collector/): Deploy the OpenTelemetry Collector on Kubernetes for unified observability. Traces, metrics, and logs pipeline configuration, auto-instrumentation.
- [OpenTelemetry in Kubernetes: Traces and Metrics](https://kubernetes.recipes/recipes/observability/kubernetes-opentelemetry-guide/): Deploy OpenTelemetry Collector in Kubernetes for distributed tracing and metrics. Auto-instrumentation, OTLP export, Jaeger integration.
- [Kubernetes Operator Pattern: CRDs and Controllers](https://kubernetes.recipes/recipes/deployments/kubernetes-operator-pattern/): The Kubernetes operator pattern explained: CRDs, reconcile loops, owner references, finalizers, status, and building operators with Kubebuilder or Operator SDK.
- [Pod Disruption Budget Best Practices](https://kubernetes.recipes/recipes/deployments/kubernetes-pdb-best-practices/): Configure PodDisruptionBudgets for high availability on Kubernetes. minAvailable vs maxUnavailable, voluntary disruptions, and upgrade coordination.
- [PDB Rolling Update Coordination K8s](https://kubernetes.recipes/recipes/deployments/kubernetes-pdb-rolling-updates/): Coordinate PodDisruptionBudgets with rolling updates on Kubernetes. minAvailable vs maxUnavailable, voluntary disruptions, and upgrade-safe configurations.
- [PersistentVolumeClaim PVC Guide K8s](https://kubernetes.recipes/recipes/storage/kubernetes-persistent-volume-claims/): Create and manage PersistentVolumeClaims on Kubernetes. Access modes, storage classes, volume expansion, and namespace-scoped PVC lifecycle.
- [Persistent Volume Expansion Kubernetes](https://kubernetes.recipes/recipes/storage/kubernetes-persistent-volume-expansion/): Expand PersistentVolumeClaims online without downtime. allowVolumeExpansion, filesystem resize, StatefulSet PVC expansion.
- [Kubernetes PV and PVC: Persistent Volume Guide](https://kubernetes.recipes/recipes/storage/kubernetes-persistent-volume-guide/): Kubernetes PersistentVolumes and PVCs: dynamic vs static (NFS) provisioning, access modes, reclaim policy, PV lifecycle, Block mode, and expansion.
- [PersistentVolume Reclaim Policies](https://kubernetes.recipes/recipes/storage/kubernetes-persistent-volume-reclaim-policy/): Understand Retain, Delete, and Recycle reclaim policies for PersistentVolumes. Manage PV lifecycle after PVC deletion and recover bound volumes.
- [K8s PersistentVolumeClaimSpec Reference](https://kubernetes.recipes/recipes/storage/kubernetes-persistentvolumeclaimspec/): Complete PersistentVolumeClaimSpec reference for Kubernetes. accessModes, storageClassName, resources, selector, volumeMode, and dataSource explained.
- [Platform Engineering: Golden Paths in K8s](https://kubernetes.recipes/recipes/configuration/kubernetes-platform-engineering-golden-paths/): Build golden paths for developers on Kubernetes. Internal developer platform with Backstage, self-service namespaces, pre-built Helm charts.
- [Pod Affinity and Anti-Affinity Guide](https://kubernetes.recipes/recipes/deployments/kubernetes-pod-affinity-antiaffinity/): Configure pod affinity and anti-affinity rules for Kubernetes scheduling. Co-locate cache with app, spread replicas across nodes.
- [Pod Disruption Budget Strategies](https://kubernetes.recipes/recipes/deployments/kubernetes-pod-disruption-budget-strategies/): Configure PodDisruptionBudgets for zero-downtime maintenance. MinAvailable vs maxUnavailable strategies for stateful workloads, GPU training.
- [Kubernetes Pod Lifecycle: Phases and States](https://kubernetes.recipes/recipes/configuration/kubernetes-pod-lifecycle-guide/): Kubernetes pod lifecycle explained: pod phases (Pending, Running, Succeeded, Failed, Unknown), container states, restart policies, hooks, and stuck pods.
- [Kubernetes Pod Priority and Preemption (PriorityClass)](https://kubernetes.recipes/recipes/deployments/kubernetes-pod-priority-preemption/): Kubernetes pod priority and preemption: PriorityClass YAML, preemptionPolicy Never, system-critical classes, tier design, quotas and preemption monitoring.
- [K8s Pod Resource Monitoring with Grafana](https://kubernetes.recipes/recipes/observability/kubernetes-pod-resource-monitoring-grafana/): Monitor Kubernetes pod CPU and memory with Grafana dashboards. Prometheus queries for resource usage, request vs limit tracking.
- [K8s Pod Security Admission Standards](https://kubernetes.recipes/recipes/security/kubernetes-pod-security-admission/): Configure Kubernetes Pod Security Admission with enforce, audit, and warn modes. Privileged, baseline, and restricted profiles for namespace-level pod security.
- [Pod Topology Spread Advanced Patterns](https://kubernetes.recipes/recipes/deployments/kubernetes-pod-topology-spread-advanced/): Advanced topology spread constraints for Kubernetes. Multi-zone HA, GPU rack awareness, combined with affinity rules, and minDomains for scaling clusters.
- [kubectl Port-Forward: Access Pods and Services](https://kubernetes.recipes/recipes/configuration/kubernetes-port-forwarding-guide/): Use kubectl port-forward to access Kubernetes pods, services, and deployments from your local machine. Debug, test, and access internal services securely.
- [K8s PriorityClass and Missing Pod Priority](https://kubernetes.recipes/recipes/configuration/kubernetes-priorityclass-missing-pod-priority/): Fix missing pod priority in Kubernetes. PriorityClass configuration, preemption behavior, system-critical classes, and scheduling order for GPU workloads.
- [Kubernetes Liveness and Readiness Probes Guide](https://kubernetes.recipes/recipes/deployments/kubernetes-probes-liveness-readiness/): Configure Kubernetes liveness, readiness, and startup probes for health checks. HTTP, TCP, exec probes, timing parameters, and failure threshold tuning.
- [Kubernetes Projected Volumes Explained](https://kubernetes.recipes/recipes/storage/kubernetes-projected-volumes/): Combine Secrets, ConfigMaps, Downward API and bound ServiceAccount tokens into one projected volume mount. YAML, permissions, token audience, gotchas.
- [Prometheus Alerting Rules Kubernetes](https://kubernetes.recipes/recipes/observability/kubernetes-prometheus-alerting-rules/): Write effective Prometheus alerting rules for Kubernetes. Alertmanager routing, inhibition, silence, and production-ready alert templates for CPU, memory.
- [Prometheus Monitoring on Kubernetes: Setup Guide](https://kubernetes.recipes/recipes/observability/kubernetes-prometheus-monitoring-guide/): Monitor Kubernetes with Prometheus: kube-prometheus-stack Helm install, ServiceMonitor/PodMonitor, PrometheusRule alerts, recording rules and key PromQL.
- [PV Reclaim Policy Retain vs Delete](https://kubernetes.recipes/recipes/storage/kubernetes-pv-reclaim-policies/): Understand Kubernetes PersistentVolume reclaim policies. Retain vs Delete vs Recycle, recovering data from released PVs.
- [K8s QoS Classes: Guaranteed vs Burstable](https://kubernetes.recipes/recipes/configuration/kubernetes-qos-classes-guide/): Understand Kubernetes QoS classes for pod eviction priority. Guaranteed, Burstable, and BestEffort resource configurations and eviction behavior under pressure.
- [Kubernetes Rate Limiting with Gateway API](https://kubernetes.recipes/recipes/networking/kubernetes-rate-limiting-gateway-api/): Implement rate limiting for Kubernetes services using Gateway API, Istio, Kong, NGINX, and Envoy. Protect APIs from abuse
- [Kubernetes Rate Limiting Guide](https://kubernetes.recipes/recipes/networking/kubernetes-rate-limiting-guide/): Implement rate limiting in Kubernetes with Ingress annotations, Gateway API, Envoy filters, and application-level middleware. Protect APIs from abuse.
- [RBAC Audit Review Kubernetes Guide](https://kubernetes.recipes/recipes/security/kubernetes-rbac-audit-guide/): Audit Kubernetes RBAC permissions for security compliance. Identify over-permissioned roles, service account privileges, and least-privilege enforcement.
- [K8s RBAC: Roles, ClusterRoles, and Bindings](https://kubernetes.recipes/recipes/security/kubernetes-rbac-guide/): Configure Kubernetes RBAC with Roles, ClusterRoles, RoleBindings, and service accounts. Least privilege access control for users, groups, and applications.
- [RBAC Least Privilege Kubernetes](https://kubernetes.recipes/recipes/security/kubernetes-rbac-least-privilege/): Configure Kubernetes RBAC with least-privilege Roles, ClusterRoles, and service account bindings. Audit permissions, restrict secrets access.
- [Kubernetes RBAC Role ClusterRole](https://kubernetes.recipes/recipes/security/kubernetes-rbac-role-clusterrole/): Configure RBAC in Kubernetes with Roles, ClusterRoles, RoleBindings, and ClusterRoleBindings. Least-privilege access for users, groups, and service accounts.
- [K8s RBAC: Role and RoleBinding Guide](https://kubernetes.recipes/recipes/security/kubernetes-rbac-role-rolebinding/): Configure Kubernetes RBAC with Role, ClusterRole, RoleBinding, and ClusterRoleBinding. Service account permissions, least privilege, and audit examples.
- [Fix RBAC Permission Errors K8s](https://kubernetes.recipes/recipes/troubleshooting/kubernetes-rbac-troubleshooting/): Debug Kubernetes RBAC permission errors. kubectl auth can-i, impersonation testing, ClusterRole aggregation, and common permission mistakes.
- [Readiness Liveness Startup Probes](https://kubernetes.recipes/recipes/deployments/kubernetes-readiness-liveness-startup/): Configure Kubernetes health probes correctly. When to use each probe type, common mistakes, and production-ready probe configurations.
- [Readiness Probe Kubernetes Guide](https://kubernetes.recipes/recipes/deployments/kubernetes-readiness-probe-guide/): Configure readiness probes correctly on Kubernetes. HTTP, TCP, exec probes, failure threshold tuning, and why readiness probes should never check databases.
- [400 Kubernetes Recipes Milestone Recap](https://kubernetes.recipes/recipes/configuration/kubernetes-recipes-400-pages-milestone/): Kubernetes Recipes reaches 400 articles. Explore new AI/GPU infrastructure, NVIDIA networking, ArgoCD GitOps, OpenShift, and RHACS security recipes.
- [Kubernetes Release Cycle and Version Support](https://kubernetes.recipes/recipes/configuration/kubernetes-release-cycle-version-support/): Kubernetes release cycle explained: 3 releases per year, 14-month support window, patch cadence, version skew policy, and upgrade planning timeline.
- [Kubernetes ReplicaSet Explained with Examples](https://kubernetes.recipes/recipes/deployments/kubernetes-replicaset-guide/): What a Kubernetes ReplicaSet does, how selectors and ownerReferences work, pod-template-hash, scaling, and why you should use a Deployment instead.
- [Kubernetes LimitRange and ResourceQuota](https://kubernetes.recipes/recipes/configuration/kubernetes-request-limit-ranges/): Configure LimitRange and ResourceQuota in Kubernetes namespaces. Set default resource requests, enforce limits, and prevent resource exhaustion across teams.
- [Kubernetes Resiliency Patterns Guide](https://kubernetes.recipes/recipes/deployments/kubernetes-resiliency-patterns/): Build resilient Kubernetes apps with PDBs, topology spread, anti-affinity, health probes, and graceful shutdown patterns.
- [Resource Format 200m 256Mi Syntax](https://kubernetes.recipes/recipes/configuration/kubernetes-resource-format-syntax/): Understand Kubernetes resource format: CPU millicores (200m, 500m, 1) and memory units (256Mi, 1Gi). Syntax reference for requests, limits.
- [Kubernetes CPU 200m Memory 256Mi: Limits Format](https://kubernetes.recipes/recipes/configuration/kubernetes-resource-limits-cpu-memory-format/): Use Kubernetes container resources limits with cpu 200m and memory 256Mi. Copy YAML syntax, convert millicores and Mi/Gi, and avoid Mi vs M mistakes.
- [CPU and Memory Limits Deep Dive](https://kubernetes.recipes/recipes/configuration/kubernetes-resource-management-limits-cpu-memory/): Deep dive into Kubernetes CPU and memory management. CFS bandwidth throttling, OOMKill scoring, cgroup v2 behavior, memory.high vs memory.
- [K8s Resource Optimization Strategies](https://kubernetes.recipes/recipes/autoscaling/kubernetes-resource-optimization/): Kubernetes resource optimization strategies and best practices. Right-size pods with VPA, Goldilocks dashboards, and resource allocation techniques.
- [Kubernetes Right-Sizing and Cost Optimization](https://kubernetes.recipes/recipes/autoscaling/kubernetes-resource-optimization-strategies/): Optimize Kubernetes resource allocation with right-sizing, VPA recommendations, bin packing, request-to-limit ratios, and cost reduction best practices.
- [Kubernetes ResourceQuota and LimitRange Examples](https://kubernetes.recipes/recipes/configuration/kubernetes-resource-quota-limitrange/): Kubernetes ResourceQuota and LimitRange YAML: namespace CPU/memory/GPU quotas, object counts, storage-class and priority scopes, default container limits.
- [Kubernetes Resource Requests and Limits Guide](https://kubernetes.recipes/recipes/configuration/kubernetes-resource-requests-limits/): Set CPU and memory requests and limits in Kubernetes: QoS classes, CPU throttling vs OOMKilled, LimitRange defaults, GPUs, and right-sizing.
- [Reusable CI/CD Pipeline Templates for K8s](https://kubernetes.recipes/recipes/deployments/kubernetes-reusable-pipeline-templates/): Build once, deploy anything. Reusable CI/CD pipeline templates for Kubernetes using GitHub Actions, GitLab CI, and Tekton.
- [Kubernetes Rolling Update: maxSurge & maxUnavailable](https://kubernetes.recipes/recipes/deployments/kubernetes-rolling-update-strategy/): Kubernetes Deployment rolling update strategy: maxSurge and maxUnavailable math, zero-downtime settings, preStop drain, rollout status and rollback.
- [Kubernetes Route and Ingress Management Guide](https://kubernetes.recipes/recipes/networking/kubernetes-route-management-openshift-ingress/): Manage OpenShift Routes and Kubernetes Ingress resources. TLS termination, path-based routing, weighted traffic splitting.
- [Sealed Secrets Management Kubernetes](https://kubernetes.recipes/recipes/security/kubernetes-sealed-secrets-management/): Manage secrets securely with Bitnami Sealed Secrets on Kubernetes. Encrypt secrets for Git storage, cluster-scoped and namespace-scoped sealing.
- [Automate Secret and Key Rotation in Kubernetes](https://kubernetes.recipes/recipes/security/kubernetes-secret-key-rotation-automation/): Automate TLS certificate and secret key rotation in Kubernetes. CronJob-based rotation, external-secrets-operator, cert-manager auto-renewal.
- [K8s Secrets: Types and Usage Guide](https://kubernetes.recipes/recipes/security/kubernetes-secret-types-guide/): Create and manage Kubernetes Secrets: Opaque, docker-registry, TLS, and basic-auth types. Mount as volumes, inject as env vars, and encrypt at rest.
- [K8s Security Checklist 2026 Guide](https://kubernetes.recipes/recipes/security/kubernetes-security-checklist-2026/): Complete Kubernetes security checklist for 2026. RBAC audit, network policies, pod security standards, image scanning, and compliance hardening steps.
- [K8s SecurityContext: Container Hardening](https://kubernetes.recipes/recipes/security/kubernetes-security-context-guide/): Configure Kubernetes SecurityContext for pods and containers. runAsNonRoot, readOnlyRootFilesystem, capabilities, seccomp profiles, and privilege escalation.
- [Harden Kubernetes Security Posture](https://kubernetes.recipes/recipes/security/kubernetes-security-posture-hardening/): Kubernetes security hardening: Pod Security Standards, RBAC least-privilege, network policies, secret encryption, and audit logging.
- [Kubernetes Service Account Tokens: kubectl create token](https://kubernetes.recipes/recipes/security/kubernetes-service-account-token/): Create Kubernetes service account tokens after 1.24: kubectl create token, projected volumes, custom audiences, long-lived Secret tokens, and expiry limits.
- [Service Accounts and Workload Identity](https://kubernetes.recipes/recipes/security/kubernetes-service-accounts-workload-identity/): Configure Kubernetes service accounts with cloud workload identity for AWS IRSA, GCP Workload Identity, and Azure AD pod federation.
- [Service DNS Discovery Kubernetes](https://kubernetes.recipes/recipes/networking/kubernetes-service-dns-discovery/): How Kubernetes DNS service discovery works. Service FQDN format, headless services, SRV records, and cross-namespace DNS resolution patterns.
- [Kubernetes Service DNS Resolution](https://kubernetes.recipes/recipes/networking/kubernetes-service-dns-resolution/): How Kubernetes Service DNS works: naming conventions, FQDN format, headless services, cross-namespace resolution, and DNS debugging with nslookup.
- [Service Mesh Comparison Kubernetes](https://kubernetes.recipes/recipes/networking/kubernetes-service-mesh-comparison/): Compare Istio, Linkerd, and Cilium service mesh for Kubernetes. mTLS, observability, traffic management, resource overhead.
- [Istio Service Mesh: Traffic Management](https://kubernetes.recipes/recipes/networking/kubernetes-service-mesh-istio-guide/): Deploy Istio service mesh in Kubernetes for traffic management, mTLS, observability, and canary deployments. VirtualService, DestinationRule.
- [Topology-Aware Service Routing](https://kubernetes.recipes/recipes/networking/kubernetes-service-topology-routing/): Enable zone-aware traffic routing in Kubernetes to reduce cross-zone latency and egress costs. Configure topology hints and traffic distribution.
- [K8s Service Types: ClusterIP, NodePort, LoadBalancer](https://kubernetes.recipes/recipes/networking/kubernetes-service-types-clusterip-nodeport-loadbalancer/): Kubernetes Service types explained: ClusterIP, NodePort, LoadBalancer, ExternalName and headless. When to use each, with YAML and traffic flow examples.
- [Kubernetes Service Types LoadBalancer ClusterIP NodePort](https://kubernetes.recipes/recipes/networking/kubernetes-service-types-loadbalancer-guide/): Understand Kubernetes Service types: ClusterIP, NodePort, LoadBalancer, and ExternalName. When to use each type, configuration examples, and traffic routing
- [K8s ServiceAccount: Pod Identity Guide](https://kubernetes.recipes/recipes/security/kubernetes-serviceaccount-guide/): Create Kubernetes ServiceAccounts for pod authentication. Token projection, RBAC binding, workload identity, automountServiceAccountToken, and OIDC federation.
- [ServiceAccount for Running Pods](https://kubernetes.recipes/recipes/security/kubernetes-serviceaccount-running-pod/): Configure Kubernetes ServiceAccounts for Pods: token mounting, RBAC permissions, workload identity, automountServiceAccountToken control, and least-privilege
- [Kubernetes Sidecar Containers: Native Sidecars & Patterns](https://kubernetes.recipes/recipes/configuration/kubernetes-sidecar-patterns/): Kubernetes sidecar containers: native sidecars (initContainers + restartPolicy Always), startup/shutdown order, logging, Envoy, TLS proxy, Jobs, resources.
- [SPIFFE/SPIRE: Workload Identity for K8s](https://kubernetes.recipes/recipes/security/kubernetes-spiffe-spire-identity/): Deploy SPIRE for Kubernetes workload identity using SPIFFE standards. Automatic mTLS certificate issuance, cross-cluster identity federation.
- [Kubernetes Startup Probes for Slow Containers](https://kubernetes.recipes/recipes/configuration/kubernetes-startup-probe-slow-containers/): Configure Kubernetes startup probes for containers with long initialization. Separate startup from liveness checks, failureThreshold tuning.
- [StatefulSet Headless Service: serviceName and Pod DNS](https://kubernetes.recipes/recipes/deployments/kubernetes-statefulset-headless-service-guide/): Why a StatefulSet needs a headless Service: serviceName, per-pod DNS records, SRV lookups, publishNotReadyAddresses, and fixing pod names that won't resolve.
- [K8s Storage Best Practices for Production](https://kubernetes.recipes/recipes/storage/kubernetes-storage-best-practices/): Production storage best practices for Kubernetes. StorageClass selection, backup strategies, volume expansion, data migration, and performance tuning.
- [Kubernetes StorageClass Complete Guide](https://kubernetes.recipes/recipes/storage/kubernetes-storage-class-guide/): Configure StorageClasses for dynamic provisioning on Kubernetes. Covers reclaim policies, volume binding modes, and cloud provider examples.
- [Storage Classes and Provisioners](https://kubernetes.recipes/recipes/storage/kubernetes-storage-classes-provisioners/): Configure Kubernetes StorageClasses for dynamic volume provisioning. CSI drivers, reclaim policies, volume expansion, topology-aware provisioning.
- [Kubernetes Taints and Tolerations Guide](https://kubernetes.recipes/recipes/configuration/kubernetes-taint-toleration-guide/): Kubernetes taints and tolerations: kubectl taint syntax, NoSchedule vs NoExecute, tolerationSeconds, built-in taints, and dedicating GPU nodes.
- [Tekton Pipelines on Kubernetes: CI/CD Guide](https://kubernetes.recipes/recipes/deployments/kubernetes-tekton-pipelines-guide/): Build CI/CD with Tekton on Kubernetes: Tasks, Pipelines, PipelineRuns, workspaces, resolvers for catalog tasks, Triggers on git push, and OpenShift Pipelines.
- [Grafana Tempo Tracing Kubernetes](https://kubernetes.recipes/recipes/observability/kubernetes-tempo-tracing-guide/): Deploy Grafana Tempo for cost-effective distributed tracing on Kubernetes. Object storage backend, TraceQL queries, and Grafana integration.
- [Thanos HA Prometheus Kubernetes](https://kubernetes.recipes/recipes/observability/kubernetes-thanos-ha-prometheus/): Scale Prometheus with Thanos for high availability and long-term storage on Kubernetes. Sidecar, Store, Compactor, and Query frontend for multi-cluster metrics.
- [Topology-Aware Routing Kubernetes](https://kubernetes.recipes/recipes/networking/kubernetes-topology-aware-routing/): Enable topology-aware routing for cost optimization on Kubernetes. Zone-local traffic, EndpointSlice hints, and reducing cross-zone data transfer costs.
- [Kubernetes Topology Manager for GPU and NUMA Alignment](https://kubernetes.recipes/recipes/ai/kubernetes-topology-manager-gpu-numa/): Configure Kubernetes Topology Manager to align CPU, GPU, and NIC allocations on the same NUMA node. Covers policies, kubelet config, and GPU performance tuning.
- [Trivy: K8s Security Scanning and SBOM](https://kubernetes.recipes/recipes/security/kubernetes-trivy-security-scanning/): Scan Kubernetes clusters with Trivy for vulnerabilities, misconfigurations, and secrets. Trivy Operator for continuous scanning, SBOM generation.
- [Kubernetes Troubleshooting Flowchart](https://kubernetes.recipes/recipes/troubleshooting/kubernetes-troubleshooting-guide/): Systematic Kubernetes troubleshooting guide with flowcharts. Debug pods, services, networking, storage, and node issues step by step with kubectl commands.
- [Automate User Onboarding & Offboarding in K8s](https://kubernetes.recipes/recipes/security/kubernetes-user-onboarding-offboarding-automation/): Automate Kubernetes user onboarding and offboarding. RBAC provisioning, namespace creation, quota assignment, OIDC group sync, and access revocation scripts.
- [ValidatingAdmissionPolicy with CEL](https://kubernetes.recipes/recipes/security/kubernetes-validating-admission-policy/): Replace admission webhooks with ValidatingAdmissionPolicy and CEL expressions for in-process, low-latency Kubernetes policy enforcement.
- [Velero Backup and Restore Kubernetes](https://kubernetes.recipes/recipes/storage/kubernetes-velero-backup-restore/): Back up and restore Kubernetes applications with Velero. Scheduled backups, cross-cluster migration, selective restore, and disaster recovery workflows.
- [Velero Snapshot Locations on Kubernetes](https://kubernetes.recipes/recipes/storage/kubernetes-velero-snapshot-locations/): Configure Velero snapshot locations for Kubernetes backup. Volume snapshots, file system backup, cross-region copies, and backup verification.
- [Kubernetes Vertical Pod Autoscaler (VPA) Guide](https://kubernetes.recipes/recipes/autoscaling/kubernetes-vertical-pod-autoscaler-vpa/): Install and configure Kubernetes VPA: Off, Initial, Recreate and in-place modes, min/max bounds, reading recommendations, and VPA with HPA.
- [Kubernetes Volume Types Explained](https://kubernetes.recipes/recipes/storage/kubernetes-volume-types/): Compare all Kubernetes volume types: emptyDir, hostPath, PVC, ConfigMap, Secret, NFS, CSI, and projected volumes. When to use each type with examples.
- [VPA Recommender Setup Kubernetes](https://kubernetes.recipes/recipes/autoscaling/kubernetes-vpa-recommender-guide/): Configure the VPA Recommender for Kubernetes resource right-sizing. Off mode recommendations, memory-only mode, and interpreting VPA suggestions.
- [VPA Resource Right-Sizing Kubernetes](https://kubernetes.recipes/recipes/autoscaling/kubernetes-vpa-resource-rightsizing/): Use Vertical Pod Autoscaler to right-size Kubernetes resource requests and limits. Off mode for recommendations, Auto mode for live adjustment.
- [Run Windows Containers on Kubernetes](https://kubernetes.recipes/recipes/deployments/kubernetes-windows-containers/): Deploy Windows workloads on Kubernetes with mixed Linux and Windows node pools. Covers taints, node selectors, and Windows-specific networking.
- [KubeVirt: Run VMs on Kubernetes](https://kubernetes.recipes/recipes/deployments/kubevirt-virtual-machines/): Run virtual machines alongside containers on Kubernetes with KubeVirt. Covers VM creation, live migration, GPU passthrough, and VM-to-container networking.
- [Kueue for Batch Jobs and GPU Queues](https://kubernetes.recipes/recipes/ai/kueue-batch-gpu-jobs-kubernetes/): Use Kueue to manage batch job queues on Kubernetes. GPU quota, fair sharing, priority queues, ML training workloads, and multi-tenant cluster scheduling.
- [Kueue Job Queuing Fair Sharing Kubernetes](https://kubernetes.recipes/recipes/ai/kueue-job-queuing-kubernetes/): Implement fair-share GPU job queuing with Kueue on Kubernetes. ClusterQueues, LocalQueues, ResourceFlavors, and cohort-based borrowing for multi-team AI cl.
- [When to Use Kustomize vs Helm](https://kubernetes.recipes/recipes/configuration/kustomize-vs-helm-comparison/): Choose between Kustomize and Helm for Kubernetes. Learn when to use each tool, key trade-offs, and how to combine both for optimal configuration management.
- [Kyverno AI Workload Provenance Verification](https://kubernetes.recipes/recipes/security/kyverno-ai-software-provenance/): Use Kyverno to verify software and content provenance for AI workloads: SBOM validation, model signing with Sigstore, dataset integrity, and supply chain
- [Kyverno CEL Policy Model Migration](https://kubernetes.recipes/recipes/security/kyverno-cel-policy-model/): Migrate Kyverno policies from YAML-based rules to CEL expressions for type-safe, performant validation. Covers CEL syntax, migration patterns, and comparison
- [Kyverno Drift Prevention for GitOps](https://kubernetes.recipes/recipes/security/kyverno-drift-prevention-gitops/): Prevent configuration drift in GitOps workflows using Kyverno: block manual kubectl edits, enforce ArgoCD/Flux ownership, and detect out-of-band changes
- [Kyverno ISO 27001 Compliance Policies](https://kubernetes.recipes/recipes/security/kyverno-iso27001-compliance/): Implement ISO 27001 and BSI IT-Grundschutz security controls in Kubernetes using Kyverno policies: access control, cryptography, operations security, and audit
- [Kyverno LLM Inference Cost and Security Guardrails](https://kubernetes.recipes/recipes/security/kyverno-llm-inference-guardrails/): Implement policy-as-code guardrails for LLM inference workloads with Kyverno: GPU quota enforcement, model size limits, cost controls, prompt injection
- [Kyverno ReBAC Multi-Tenant RBAC Automation](https://kubernetes.recipes/recipes/security/kyverno-rebac-multi-tenant-rbac/): Implement Relationship-Based Access Control (ReBAC) with Kyverno to automate multi-tenant RBAC at scale: dynamic RoleBindings, namespace
- [Kyverno Webhook Topology and Admission Latency](https://kubernetes.recipes/recipes/security/kyverno-webhook-admission-latency/): Optimize Kyverno webhook topology for minimal admission latency: webhook configuration tuning, failure policies, timeout settings, and lessons from migrating
- [LACP Storage Switch Kubernetes Guide](https://kubernetes.recipes/recipes/networking/lacp-storage-switch-kubernetes/): Configure LACP bond aggregation for NFS and iSCSI storage switches in Kubernetes clusters. 802.3ad setup, hash policies, switch config, and failure handling.
- [LeaderWorkerSet (LWS) on Kubernetes and OpenShift](https://kubernetes.recipes/recipes/ai/leaderworkerset-operator/): Deploy LeaderWorkerSet on Kubernetes or OpenShift for multi-node LLM inference and distributed training: install, vLLM example, env vars, restart policy.
- [Kubernetes Linkerd Service Mesh mTLS Guide](https://kubernetes.recipes/recipes/networking/linkerd-service-mesh-mtls-kubernetes/): Deploy Linkerd service mesh on Kubernetes for automatic mTLS, traffic observability, and reliability features. Zero-config encryption, per-route
- [LitmusChaos: Chaos Engineering on Kubernetes](https://kubernetes.recipes/recipes/troubleshooting/litmus-chaos-engineering-kubernetes/): Run LitmusChaos experiments on Kubernetes: install ChaosCenter, pod-delete ChaosEngine, HTTP/cmd/Prometheus probes, ChaosResult verdicts vs Chaos Mesh.
- [Llama 2 70B Parameters & Memory: FP16 140GB](https://kubernetes.recipes/recipes/ai/llama-70b-model-size-gpu-requirements/): Llama 2 70B requires 140GB VRAM in FP16 (70B params × 2 bytes). FP8 uses 70GB, INT4 uses 35GB. Complete GPU options and Kubernetes deployment guide.
- [Llama Stack on Kubernetes with NVIDIA NIM](https://kubernetes.recipes/recipes/ai/llama-stack-kubernetes-nim/): Deploy Meta Llama Stack on Kubernetes for unified inference, RAG, agents, and safety APIs using NVIDIA NIM as the inference backend.
- [Autoscale LLM Inference on Kubernetes](https://kubernetes.recipes/recipes/ai/llm-autoscaling-kubernetes/): Configure Horizontal Pod Autoscaling and KEDA for LLM workloads using GPU utilization, request queue depth, and custom metrics.
- [LLM Deployment Challenges Kubernetes](https://kubernetes.recipes/recipes/ai/llm-deployment-challenges-kubernetes/): The five common LLM deployment failure modes on Kubernetes: GPU memory, cold starts, latency, autoscaling, and multi-node inference, with worked YAML.
- [Quantize LLMs for Efficient GPU Inference](https://kubernetes.recipes/recipes/ai/llm-quantization-kubernetes/): Run quantized LLM models (GPTQ, AWQ, GGUF) on Kubernetes to reduce GPU memory requirements and serve models on smaller GPUs.
- [Kubernetes LLM Serving Frameworks Compared](https://kubernetes.recipes/recipes/ai/llm-serving-frameworks-compared/): Compare vLLM, NVIDIA NIM, Triton, Ollama, and llama.cpp for serving LLMs on Kubernetes — features, performance, and when to use each.
- [Push a Podman-Saved Image to Local Quay](https://kubernetes.recipes/recipes/deployments/local-quay-push-podman-image/): Load a Podman image tar archive, tag it for your Local Quay registry, authenticate with robot accounts, and push it safely to your private repo.
- [Retag and Push an Image in Local Quay](https://kubernetes.recipes/recipes/deployments/local-quay-retag-and-push/): Pull an existing container image from Local Quay, retag it for a new repository path or version, and push the updated tag back to the registry.
- [Longhorn Distributed Storage on Kubernetes](https://kubernetes.recipes/recipes/storage/longhorn-distributed-storage/): Install Longhorn distributed block storage on Kubernetes: prerequisites, Helm install, replicated PVCs, RWX, snapshots, recurring S3 backups and DR restore.
- [LoRA Adapter Serving vLLM on K8s](https://kubernetes.recipes/recipes/ai/lora-adapter-serving-vllm-kubernetes/): Serve multiple LoRA adapters with a single vLLM base model on Kubernetes. Dynamic loading, per-request routing, and multi-tenant fine-tuned models.
- [Inspect MachineConfig Annotations on Nodes](https://kubernetes.recipes/recipes/configuration/machineconfig-annotation-inspection/): Read and interpret MachineConfig annotations on OpenShift nodes. Check desired vs current config, node state, and rendered config hashes to diagnose MCP issues.
- [Configure NTP Chrony via MachineConfig](https://kubernetes.recipes/recipes/configuration/machineconfig-chrony-ntp/): Set custom NTP servers on OpenShift RHCOS nodes using MachineConfig. Fix time drift, configure chrony, and verify time synchronization across your cluster.
- [Set Kernel Parameters via MachineConfig](https://kubernetes.recipes/recipes/configuration/machineconfig-kernel-parameters/): Tune kernel sysctl parameters on OpenShift nodes using MachineConfig. Set networking, memory, and performance sysctls on RHCOS.
- [MachineConfig NFS Mount on OpenShift Nodes](https://kubernetes.recipes/recipes/storage/machineconfig-nfs-mount-openshift/): Mount NFS shares on OpenShift worker nodes using MachineConfig systemd mount units. The only production-safe way to persist NFS mounts on RHCOS nodes.
- [Configure Container Registries via MachineConfig](https://kubernetes.recipes/recipes/configuration/machineconfig-registries-conf/): Set up mirror registries and blocked registries on OpenShift nodes using MachineConfig to control CRI-O image pull on RHCOS.
- [MariaDB Operator on Kubernetes](https://kubernetes.recipes/recipes/deployments/mariadb-operator-kubernetes/): Deploy highly available MariaDB clusters on Kubernetes using MariaDB Operator with Galera replication, automated backups, and connection pooling.
- [Deploy MariaDB on OpenShift with SCC](https://kubernetes.recipes/recipes/deployments/mariadb-scc-openshift-deployment/): Deploy MariaDB on OpenShift with proper Security Context Constraints. Configure anyuid SCC, persistent storage, custom my.
- [Fix Stale MachineConfigPool Updates](https://kubernetes.recipes/recipes/troubleshooting/mcp-blocked-stale-update/): Debug and resolve stale OpenShift MachineConfigPool updates. Identify blocked nodes, check MachineConfigDaemon logs, and unblock stuck MCP rollouts.
- [MCP Drain Blocked by PDB: Workaround](https://kubernetes.recipes/recipes/troubleshooting/mcp-drain-pdb-workaround/): Resolve OpenShift MachineConfigPool drain failures caused by PodDisruptionBudget violations. Scale down and restore after update.
- [Configure MCP maxUnavailable for Rollouts](https://kubernetes.recipes/recipes/configuration/mcp-maxunavailable-configuration/): Control how many nodes the MachineConfig Operator updates simultaneously. Set maxUnavailable for faster rollouts or safer one-at-a-time updates in production.
- [Pause and Unpause MCP Rollouts](https://kubernetes.recipes/recipes/configuration/mcp-pause-unpause-rollout/): Temporarily pause MachineConfigPool rollouts to batch multiple MachineConfig changes or coordinate with maintenance windows. Unpause to resume node updates.
- [Automate MCP Updates with Drain Script](https://kubernetes.recipes/recipes/configuration/mcp-update-automation-script/): Bash script to automate OpenShift MachineConfigPool updates when drains are blocked by PDB violations. Auto-detects blockers, scales down, drains, and restores.
- [Separate Worker and Infra MachineConfigPools](https://kubernetes.recipes/recipes/configuration/mcp-worker-infra-separation/): Create dedicated MachineConfigPools for infrastructure and GPU nodes. Isolate MCP rollout blast radius and control update order for different node types.
- [Mellanox RoCE DSCP QoS DaemonSet](https://kubernetes.recipes/recipes/networking/mellanox-roce-dscp-qos-daemonset/): Deploy a DaemonSet that configures DSCP trust, PFC priority 3, and RoCE ToS 106 on all Mellanox PFs. Uses DOCA driver image with ibdev2netdev, mlnx_qos.
- [Mirror OpenShift Releases to Disconnected Registry](https://kubernetes.recipes/recipes/configuration/mirror-openshift-releases-disconnected-registry/): Mirror OCP release images to an air-gapped Quay registry using oc adm release mirror. Auth setup, proxy config, ImageDigestMirrorSet, and disconnected updates.
- [Mirror Registry Disconnected OpenShift](https://kubernetes.recipes/recipes/deployments/mirror-registry-disconnected-openshift/): Set up a mirror registry for disconnected OpenShift installations. Deploy mirror-registry for Red Hat OpenShift, configure storage, TLS, and credentials.
- [Mistral FSDP LoRA Complete Accelerate Config](https://kubernetes.recipes/recipes/ai/mistral-fsdp-lora-accelerate-config/): Complete accelerate FSDP configuration for fine-tuning Mistral-Small-4 119B with LoRA on multi-GPU H200 clusters. Covers every FSDP2 setting with explanations.
- [ML Pipeline Automation Kubernetes](https://kubernetes.recipes/recipes/ai/ml-pipeline-automation-kubernetes/): Automate ML pipelines on Kubernetes with Kubeflow Pipelines, Argo Workflows, and Tekton. Data preprocessing, training, evaluation, model registration.
- [mlnx_qos QoS on MOFED Containers](https://kubernetes.recipes/recipes/networking/mlnx-qos-mofed-container-kubernetes/): Configure RDMA QoS with mlnx_qos from MOFED containers on Kubernetes. Set PFC, ETS, DSCP trust mode, and validate lossless RoCE traffic classes on ConnectX.
- [MLPerf Benchmarking on Kubernetes](https://kubernetes.recipes/recipes/ai/mlperf-benchmark-kubernetes/): Run MLPerf inference and training benchmarks on Kubernetes GPU clusters to validate AI workload performance and compare hardware configurations.
- [Shared Model Caching Across Pods on Kubernetes](https://kubernetes.recipes/recipes/ai/model-caching-shared-memory/): Optimize LLM inference startup and reduce storage costs by sharing model weights across pods using emptyDir, hostPath, ReadWriteMany PVCs, and init.
- [AI Model Storage: hostPath vs PVC Inference](https://kubernetes.recipes/recipes/ai/model-storage-hostpath-pvc/): Deploy AI models on Kubernetes using hostPath and PVC storage. Compare performance, security trade-offs, and production patterns for model serving.
- [ModelMesh Multi-Model Serving Kubernetes](https://kubernetes.recipes/recipes/ai/modelmesh-multi-model-serving-kubernetes/): Deploy hundreds of ML models on shared GPU infrastructure with ModelMesh. Intelligent model loading and unloading, memory management, routing.
- [MOFED and DOCA Driver Building for OpenShift](https://kubernetes.recipes/recipes/configuration/mofed-doca-driver-building-openshift/): Build NVIDIA MOFED and DOCA drivers for OpenShift using DriverToolKit, Buildah, and MachineConfig for RDMA and GPU networking.
- [MOFED Driver for Kubernetes: Setup Guide](https://kubernetes.recipes/recipes/networking/mofed-driver-kubernetes-guide/): Install and manage MOFED drivers in Kubernetes. Network Operator integration, NicClusterPolicy, driver versions, and RDMA troubleshooting.
- [MOFED Driver Operator Build Kubernetes](https://kubernetes.recipes/recipes/networking/mofed-driver-operator-build-kubernetes/): Let the NVIDIA Network Operator build MOFED drivers on-node via DKMS. Kernel header detection, compile flags, and DTK integration for OpenShift.
- [Monitor NCCL Benchmarks with Prometheus & Grafana](https://kubernetes.recipes/recipes/observability/monitor-nccl-performance-prometheus/): Track NCCL benchmark outcomes and GPU telemetry over time with Prometheus and Grafana dashboards to detect communication regressions early.
- [MPI DNS Resolution and Hostfile for Kubernetes GPU Jobs](https://kubernetes.recipes/recipes/ai/mpi-dns-resolution-hostfile-kubernetes/): Troubleshoot MPI hostfile DNS resolution in Kubeflow MPIJob on Kubernetes. Covers headless Service creation, subdomain configuration, DNS wait loops, FQDN
- [MPI Operator for Distributed Training](https://kubernetes.recipes/recipes/ai/mpi-operator-kubernetes/): Deploy MPI Operator on Kubernetes for distributed GPU training with Horovod and NCCL. Run multi-node MPI jobs natively in Kubernetes pods.
- [Multi-Architecture Container Images for Kubernetes](https://kubernetes.recipes/recipes/deployments/multi-architecture-container-images-kubernetes/): Build and deploy multi-architecture container images for mixed Kubernetes clusters. Docker buildx, manifest lists, image indexes, platform-aware
- [Multi-Cloud AI Workloads Kubernetes](https://kubernetes.recipes/recipes/ai/multi-cloud-ai-kubernetes/): Run AI workloads across multiple cloud providers with Kubernetes. GPU instance availability, spot pricing arbitrage, model portability.
- [Kubernetes Multi-Cluster Management: Fleet Patterns](https://kubernetes.recipes/recipes/deployments/multi-cluster-management-kubernetes/): Manage Kubernetes fleets: kubectl contexts/kubectx, Argo CD ApplicationSets, Rancher Fleet, OpenShift ACM, Cluster API, and cross-cluster service mesh.
- [Multi-GPU and Tensor Parallel LLM Inference](https://kubernetes.recipes/recipes/ai/multi-gpu-llm-inference/): Deploy large language models across multiple GPUs using tensor parallelism with vLLM and NVIDIA NIM on Kubernetes for high-throughput inference serving.
- [Multi-GPU PyTorch DDP on Kubernetes](https://kubernetes.recipes/recipes/ai/multi-gpu-pytorch-ddp-kubernetes/): Run PyTorch DistributedDataParallel across multiple GPUs on Kubernetes. torchrun, NCCL backend, pod topology, and scaling to multi-node training.
- [Multi-Node Distributed Training on Kubernetes](https://kubernetes.recipes/recipes/ai/multi-node-training-kubernetes/): Run distributed deep learning training across multiple GPU nodes on Kubernetes. Covers PyTorch DDP, DeepSpeed, Horovod, and MPI jobs with NCCL optimization.
- [Multi-Tenant GPU Namespace Isolation](https://kubernetes.recipes/recipes/security/multi-tenant-gpu-namespace-isolation/): Isolate GPU workloads across tenants using namespaces, RBAC, NetworkPolicy, and ResourceQuotas on OpenShift and Kubernetes.
- [Mutagen Kubernetes File Sync Guide](https://kubernetes.recipes/recipes/deployments/mutagen-kubernetes-file-sync/): Sync files between local machine and Kubernetes pods with Mutagen. Real-time bidirectional sync for development, hot-reload workflows.
- [Fix Kubernetes Namespace Stuck in Terminating](https://kubernetes.recipes/recipes/troubleshooting/namespace-stuck-terminating/): Fix a Kubernetes namespace stuck in Terminating: find leftover resources, finalizers and broken APIServices, then force-remove the namespace finalizer safely.
- [Run NCCL AllGather Benchmarks for Model Parallel](https://kubernetes.recipes/recipes/ai/nccl-allgather-benchmark-profile/): Use all-gather NCCL tests to evaluate GPU communication behavior and throughput for tensor-parallel and model-parallel distributed AI workloads on Kubernetes.
- [NCCL Channel Routing and Transport Path Analysis](https://kubernetes.recipes/recipes/ai/nccl-channel-routing-transport-analysis/): Interpret NCCL channel logs to understand GPU communication paths on Kubernetes. Decode P2P/CUMEM, SHM/direct, NET/IB/GDRDMA transport
- [NCCL Debug Subsystems for GPU Network Troubleshooting](https://kubernetes.recipes/recipes/troubleshooting/nccl-debug-subsystems-gpu-troubleshooting/): Configure NCCL_DEBUG and NCCL_DEBUG_SUBSYS for targeted logging during multi-node GPU training. Covers INIT, NET, GRAPH subsystems, log
- [NCCL DMABUF Enable for GPUDirect RDMA on Kubernetes](https://kubernetes.recipes/recipes/ai/nccl-dmabuf-gpudirect-rdma-kubernetes/): Enable NCCL DMA-BUF support for GPUDirect RDMA in Kubernetes GPU clusters. Covers NCCL_DMABUF_ENABLE=1, kernel requirements, nvidia-peermem vs dmabuf, GPU
- [NCCL Environment Variables Complete Reference](https://kubernetes.recipes/recipes/ai/nccl-environment-variables-reference-kubernetes/): NCCL environment variables reference for Kubernetes: NCCL_SOCKET_IFNAME, NCCL_IB_HCA, GID index, GDR level, socket tuning, algorithms, and NCCL_DEBUG.
- [NCCL_NET_GDR_LEVEL Environment Variable Guide](https://kubernetes.recipes/recipes/ai/nccl-gdr-level-tuning-pix-pxb-phb-sys/): NCCL_NET_GDR_LEVEL sets GPUDirect RDMA's max PCIe distance. Compare PIX, PXB, PHB, and SYS thresholds and pick the fastest safe level.
- [NCCL GPUDirect RDMA Distance Levels and PIX vs SYS](https://kubernetes.recipes/recipes/ai/nccl-gpudirect-rdma-distance-pix-sys/): Understand NCCL GPU Direct RDMA distance-based enablement. When PIX mode disables GDRDMA for distant GPU-HCA pairs (distance 9 > 4) and when SYS mode enables
- [NCCL_IB_DISABLE Environment Variable](https://kubernetes.recipes/recipes/ai/nccl-ib-disable-environment-variable/): NCCL_IB_DISABLE environment variable explained. Set NCCL_IB_DISABLE=1 for Ethernet-only clusters, debug InfiniBand errors, and tune GPU communication.
- [NCCL IB HCA Selection and QPS Tuning for RoCE](https://kubernetes.recipes/recipes/ai/nccl-ib-hca-qps-tuning-roce/): Configure NCCL_IB_HCA, NCCL_IB_GID_INDEX, NCCL_IB_QPS_PER_CONNECTION, and NCCL_IB_SPLIT_DATA_ON_QPS for optimal RoCE performance on Kubernetes GPU clusters.
- [NCCL Network Validation Script for OpenShift GPU Clusters](https://kubernetes.recipes/recipes/ai/nccl-network-validation-script-openshift/): Build a comprehensive NCCL network validation script for OpenShift GPU clusters with SR-IOV. Configure NCCL_IB_GID_INDEX, NCCL_NET_GDR_LEVEL=SYS, per-rank HCA
- [NCCL Network Validation Troubleshooting Checklist](https://kubernetes.recipes/recipes/troubleshooting/nccl-network-validation-troubleshooting-checklist/): Complete troubleshooting checklist for NCCL multi-node GPU bandwidth validation. Covers SR-IOV VF allocation, /dev/infiniband visibility, RoCE GID
- [Production NCCL Network Validator for Kubeflow MPIJob](https://kubernetes.recipes/recipes/ai/nccl-network-validator-production-mpijob/): Deploy a production-ready NCCL network validation framework using Kubeflow MPIJob on OpenShift. Complete validate_network.sh script
- [Diagnose GPU Peer-to-Peer Latency NCCL Tests](https://kubernetes.recipes/recipes/troubleshooting/nccl-p2p-latency-diagnostics/): Use NCCL point-to-point and collective tests to isolate GPU peer-to-peer latency issues between GPU pairs in multi-node Kubernetes clusters.
- [NCCL PXN Cross-NIC Communication via NVLink](https://kubernetes.recipes/recipes/ai/nccl-pxn-cross-nic-nvlink-topology/): Configure NCCL PXN (PCIe cross-NIC via NVLink) for multi-node GPU training where not every GPU has a direct RDMA NIC. Covers topology
- [NCCL and RCCL Networking Performance on Kubernetes](https://kubernetes.recipes/recipes/ai/nccl-rccl-networking-performance-kubernetes/): Optimize NCCL (NVIDIA) and RCCL (AMD) collective communication performance on Kubernetes GPU clusters. Network transport selection, bandwidth tuning, latency
- [NCCL RoCE Validation with Kubeflow MPIJob on Kubernetes](https://kubernetes.recipes/recipes/ai/nccl-roce-validation-mpijob-kubernetes/): Validate NCCL over RoCE with a Kubeflow MPIJob: launcher/worker YAML, OpenMPI vs NCCL network planes, RDMA device requests, IB logs, and expected busbw.
- [Shared Memory Transport for NCCL Intra-Node GPU](https://kubernetes.recipes/recipes/ai/nccl-shared-memory-transport-kubernetes/): Configure NCCL shared memory (SHM) transport for intra-node GPU communication on Kubernetes. Covers /dev/shm sizing with emptyDir and NVLink/PCIe P2P paths.
- [NCCL_SOCKET_IFNAME Environment Variable Guide](https://kubernetes.recipes/recipes/networking/nccl-socket-ifname-environment-variable/): NCCL_SOCKET_IFNAME documentation for GPU training on Kubernetes. Network interface selection, bonding, InfiniBand, and NCCL timeout troubleshooting.
- [NCCL SR-IOV GDS PyTorch Configuration](https://kubernetes.recipes/recipes/ai/nccl-sriov-gds-pytorch-kubernetes/): Configure NCCL with SR-IOV RDMA and GPUDirect Storage on Kubernetes. PyTorch 25.11 container with NCCL 2.28, CUDA 13, MOFED 5.4, GDRCopy 2.
- [NCCL Topology Dump File for GPU Debugging](https://kubernetes.recipes/recipes/ai/nccl-topo-dump-file-gpu-topology/): Use NCCL_TOPO_DUMP_FILE to capture and analyze GPU interconnect topology in Kubernetes. Debug NVLink, NVSwitch, and PCIe connection paths.
- [NCCL Topology Dump and Tuning on Kubernetes](https://kubernetes.recipes/recipes/ai/nccl-topology-dump-tuning-kubernetes/): Use NCCL_TOPO_DUMP_FILE to export and inject GPU topology on Kubernetes for reproducible distributed training performance. Topology XML caching, environment
- [How to Implement Advanced NetworkPolicies](https://kubernetes.recipes/recipes/security/network-policies-advanced/): Master advanced Kubernetes NetworkPolicies for fine-grained traffic control. Learn egress rules, CIDR blocks, namespace isolation, and common security.
- [Debug NetworkPolicy Connectivity Issues](https://kubernetes.recipes/recipes/troubleshooting/network-policy-debug-connectivity/): Troubleshoot pods unable to communicate despite correct Services. Verify NetworkPolicy rules, label selectors, and default deny.
- [K8s NetworkPolicy: Default Deny All Traffic](https://kubernetes.recipes/recipes/networking/networkpolicy-deny-all/): Implement zero-trust network security in Kubernetes with default deny-all NetworkPolicy. Block all ingress and egress traffic with allow-list rules.
- [NetworkPolicy Deny-Default for GPU Tenants](https://kubernetes.recipes/recipes/security/networkpolicy-deny-default-gpu/): Implement deny-by-default NetworkPolicy for GPU tenant namespaces with NCCL port exceptions and DNS egress on Kubernetes.
- [NFS Tenant Segregation Kubernetes](https://kubernetes.recipes/recipes/storage/nfs-tenant-segregation-kubernetes/): Implement NFS tenant segregation in Kubernetes with six-layer defense-in-depth. Exports, StorageClass, quotas, and admission policies.
- [NFSoRDMA Bond with Access Mode Switch](https://kubernetes.recipes/recipes/networking/nfsordma-bond-access-mode/): Configure bonded NICs for NFS over RDMA using switch access mode for VLAN assignment. Aggregation on untagged interfaces for RDMA redundancy.
- [NFSoRDMA Dedicated NIC Configuration](https://kubernetes.recipes/recipes/networking/nfsordma-dedicated-nic/): Configure dedicated NICs for NFS over RDMA on Kubernetes worker nodes. NFSoRDMA requires untagged interfaces — no VLAN tagging supported.
- [NFSoRDMA Jumbo Frames MTU Configuration](https://kubernetes.recipes/recipes/networking/nfsordma-jumbo-frames-mtu/): Configure 9000 MTU jumbo frames for NFSoRDMA interfaces using NNCP to maximize RDMA throughput on Kubernetes worker nodes.
- [NFSoRDMA Multi-VLAN Switch Access Mode](https://kubernetes.recipes/recipes/networking/nfsordma-multi-vlan-switch-access/): Design multi-VLAN NFSoRDMA networks using switch access mode ports. Separate storage, replication, and backup traffic with dedicated NICs per VLAN.
- [NFSoRDMA Persistent Volume for Kubernetes](https://kubernetes.recipes/recipes/storage/nfsordma-persistent-volume/): Create PersistentVolumes and StorageClasses for NFSoRDMA storage with RDMA transport, optimized mount options, and ReadWriteMany access.
- [NFSoRDMA Troubleshooting and Performance](https://kubernetes.recipes/recipes/networking/nfsordma-troubleshooting-performance/): Troubleshoot NFS over RDMA connectivity issues, diagnose TCP fallback, tune performance, and benchmark RDMA throughput on Kubernetes workers.
- [NFSoRDMA Worker Node Setup Guide](https://kubernetes.recipes/recipes/networking/nfsordma-worker-node-setup/): Complete worker node setup for NFS over RDMA including kernel modules, NFS client configuration, PersistentVolume mounts, and RDMA transport verification.
- [NGINX Ingress limit-burst-multiplier](https://kubernetes.recipes/recipes/networking/nginx-ingress-limit-burst-multiplier/): Configure nginx.ingress.kubernetes.io/limit-burst-multiplier for rate limiting burst control. Tune burst size, rate limits, and 429 response handling.
- [NicClusterPolicy MOFED Affinity & Node Selection](https://kubernetes.recipes/recipes/configuration/nicclusterpolicy-node-selection/): Configure NicClusterPolicy node selectors and affinity rules to deploy MOFED drivers only on RDMA-capable nodes in Kubernetes clusters.
- [NIM Model Profiles and Selection on Kubernetes](https://kubernetes.recipes/recipes/ai/nim-model-profiles-selection-kubernetes/): Configure NIM_MODEL_PROFILE for NVIDIA NIM deployments on Kubernetes. List profiles, select by ID or name, tune VRAM, and override with vLLM CLI args.
- [NIM Multi-Node Deployment with Helm on K8s](https://kubernetes.recipes/recipes/ai/nim-multinode-deployment-helm-kubernetes/): Deploy NVIDIA NIM across multiple Kubernetes nodes using Helm, LeaderWorkerSet, Ray, and vLLM. Run Llama 405B and DeepSeek-R1 on 16+ GPUs.
- [NIM LLM Support Matrix and GPU Compatibility](https://kubernetes.recipes/recipes/ai/nim-support-matrix-gpu-compatibility/): Complete NVIDIA NIM support matrix for Kubernetes. Supported models, profiles, precision formats, GPU compatibility, and hardware requirements per model.
- [NMState Bond LACP Configuration OpenShift](https://kubernetes.recipes/recipes/networking/nmstate-bond-lacp-openshift/): Configure LACP bonding with NMState on OpenShift. NodeNetworkConfigurationPolicy for 802.3ad bonds, VLAN tagging, and storage network bonds.
- [NMState Network Config for GPU Worker Nodes](https://kubernetes.recipes/recipes/networking/nmstate-ethernet-bond-vlan-gpu-nodes/): Declaratively configure Ethernet bonding, VLANs, MTU, and static routes on GPU worker nodes using NMState on OpenShift. Covers bonding modes, LACP
- [NMState & nmstatectl: Node Network Management](https://kubernetes.recipes/recipes/networking/nmstate-nmstatectl-node-network-guide/): Manage node networking with NMState declarative API and nmstatectl CLI. Create NodeNetworkConfigurationPolicy manifests, verify with nmstatectl.
- [NMState Operator Install OpenShift K8s](https://kubernetes.recipes/recipes/networking/nmstate-operator-install-openshift/): Install and configure the NMState operator on OpenShift and Kubernetes. Enable declarative node networking with NNCP, NodeNetworkState, and enactments.
- [NNCP Bond Interfaces on Worker Nodes](https://kubernetes.recipes/recipes/networking/nncp-bond-interfaces-workers/): Create bonded network interfaces on Kubernetes worker nodes using NodeNetworkConfigurationPolicy for NIC redundancy and link aggregation.
- [NNCP DNS and Static Routes on Workers](https://kubernetes.recipes/recipes/networking/nncp-dns-routes-workers/): Configure static routes, DNS servers, and policy-based routing on worker nodes using NodeNetworkConfigurationPolicy for multi-network setups.
- [NNCP Linux Bridge on Worker Nodes](https://kubernetes.recipes/recipes/networking/nncp-linux-bridge-workers/): Create Linux bridges on Kubernetes worker nodes using NodeNetworkConfigurationPolicy for KubeVirt VM networking and pod bridging.
- [NNCP MTU and Jumbo Frames on Workers](https://kubernetes.recipes/recipes/networking/nncp-mtu-jumbo-frames-workers/): Set MTU and enable jumbo frames on worker node interfaces using NodeNetworkConfigurationPolicy for high-throughput storage and AI networking.
- [NNCP Multi-NIC Architecture for Workers](https://kubernetes.recipes/recipes/networking/nncp-multi-nic-workers/): Design a complete multi-NIC worker node architecture with NNCP for separated management, storage, tenant, and GPU traffic using bonds, VLANs, and bridges.
- [NNCP NodeNetworkConfigurationPolicy](https://kubernetes.recipes/recipes/networking/nncp-nodenetworkconfigurationpolicy-guide/): Master NodeNetworkConfigurationPolicy (NNCP) on OpenShift and Kubernetes. Configure VLANs, bonds, bridges, SR-IOV, MTU, static IPs, and DNS with NMState.
- [NNCP OVS Bridge on Worker Nodes](https://kubernetes.recipes/recipes/networking/nncp-ovs-bridge-workers/): Configure Open vSwitch bridges on Kubernetes worker nodes using NodeNetworkConfigurationPolicy for advanced SDN and DPDK networking.
- [NNCP Rollback and Troubleshooting](https://kubernetes.recipes/recipes/networking/nncp-rollback-troubleshooting/): Troubleshoot NodeNetworkConfigurationPolicy failures, monitor enactments, configure rollback timeouts, and recover from bad network configurations.
- [NNCP SR-IOV and Macvlan on Workers](https://kubernetes.recipes/recipes/networking/nncp-sriov-macvlan-workers/): Configure SR-IOV virtual functions and macvlan interfaces on worker nodes using NodeNetworkConfigurationPolicy for high-performance networking.
- [NNCP Static IP Assignment on Worker Nodes](https://kubernetes.recipes/recipes/networking/nncp-static-ip-workers/): Use NodeNetworkConfigurationPolicy to assign static IPv4 and IPv6 addresses to worker node interfaces with nodeSelector targeting.
- [NNCP VLAN Tagging on Worker Nodes](https://kubernetes.recipes/recipes/networking/nncp-vlan-tagging-workers/): Configure VLAN interfaces on Kubernetes worker nodes using NodeNetworkConfigurationPolicy for network segmentation and traffic isolation.
- [Node Drain Blocked by hostNetwork Port Conflict](https://kubernetes.recipes/recipes/troubleshooting/node-drain-hostnetwork-ports/): Debug and fix OpenShift node drains that fail because hostNetwork pods cannot schedule replacements due to port exhaustion across the cluster.
- [Node Feature Discovery Operator for Kubernetes](https://kubernetes.recipes/recipes/ai/node-feature-discovery-operator/): Install and configure Node Feature Discovery (NFD) Operator to auto-detect hardware features like GPUs, NICs, CPU flags, and USB devices on Kubernetes nodes.
- [Debug Node NotReady Status in Kubernetes](https://kubernetes.recipes/recipes/troubleshooting/node-not-ready-troubleshooting/): Diagnose Kubernetes nodes stuck in NotReady state. Check kubelet logs, container runtime, network, disk pressure, and certificates.
- [How to Set Up Node Problem Detector](https://kubernetes.recipes/recipes/observability/node-problem-detector/): Detect and report node-level issues automatically with Node Problem Detector. Learn to identify kernel problems, hardware failures, and container.
- [NodePort Raw Traffic vs HTTPS Ingress](https://kubernetes.recipes/recipes/networking/nodeport-raw-vs-https-ingress/): Route raw GPU inference traffic via NodePort for low-latency gRPC and HTTPS model serving via OpenShift ingress controller.
- [NVIDIA IPAM for GPU Fabric IP Address Allocation](https://kubernetes.recipes/recipes/networking/nv-ipam-gpu-fabric-ip-allocation/): Configure nv-ipam (NVIDIA IPAM) to assign IP addresses on GPU fabric SR-IOV networks in Kubernetes. Covers IPPool CRDs, per-node allocation, InfiniBand IPoIB
- [Deploy NVIDIA Clara on Kubernetes](https://kubernetes.recipes/recipes/ai/nvidia-clara-medical-ai-kubernetes/): Deploy NVIDIA Clara medical AI and drug discovery platform on Kubernetes. Run digital biology and medtech inference workloads with GPU acceleration.
- [NVIDIA CNS with Insight Operator for Network Diagnostics](https://kubernetes.recipes/recipes/observability/nvidia-cns-insight-operator-network-diagnostics/): Deploy NVIDIA Cloud-Native Stack (CNS) with the Insight Operator and NVIDIA Insight tools for deep GPU fabric diagnostics. Collect NIC firmware health, link
- [NVIDIA DCGM Exporter GPU Monitoring](https://kubernetes.recipes/recipes/observability/nvidia-dcgm-exporter-monitoring/): Monitor GPU metrics with DCGM Exporter on K8s. Prometheus integration, Grafana dashboards, and alerting on utilization and temperature.
- [NVIDIA DOCA Bench for DPU Performance Testing on Kubernetes](https://kubernetes.recipes/recipes/networking/nvidia-doca-bench-dpu-performance-kubernetes/): Benchmark NVIDIA BlueField DPU accelerators in Kubernetes with DOCA Bench: throughput/latency modes, RDMA, compression offload, and multi-core scaling.
- [NVIDIA DOCA Telemetry for Network Monitoring on Kubernetes](https://kubernetes.recipes/recipes/observability/nvidia-doca-telemetry-network-monitoring-kubernetes/): Deploy NVIDIA DOCA Telemetry Service (DTS) to collect real-time network metrics from BlueField DPUs and ConnectX NICs. Export RoCE counters, port
- [NVIDIA Driver Update K8s Nodes Guide](https://kubernetes.recipes/recipes/ai/nvidia-driver-update-kubernetes/): Safely update NVIDIA GPU drivers on Kubernetes nodes. Rolling updates, drain strategy, driver compatibility matrix, and GPU Operator upgrades.
- [NVIDIA Dynamo Distributed Inference](https://kubernetes.recipes/recipes/ai/nvidia-dynamo-distributed-inference-kubernetes/): Deploy NVIDIA Dynamo on Kubernetes for disaggregated LLM inference. KV-aware routing, prefill/decode splitting, Grove operator, and zero-config deployment.
- [NVIDIA Dynamo Production Tuning on Kubernetes](https://kubernetes.recipes/recipes/ai/nvidia-dynamo-production-tuning-kubernetes/): Tune NVIDIA Dynamo for production LLM inference: prefill/decode pool sizing, KV cache transfer optimization, NCCL backend selection, SLA-driven autoscaling
- [NVIDIA GPUDirect Storage Benchmark on K8s](https://kubernetes.recipes/recipes/ai/nvidia-gds-benchmark-kubernetes/): Benchmark NVIDIA GPUDirect Storage (GDS) on Kubernetes for direct NVMe-to-GPU data transfers. Covers gdsio, gds_stats, performance validation, and comparison
- [NVIDIA GPU Feature Discovery for Kubernetes](https://kubernetes.recipes/recipes/ai/nvidia-gpu-feature-discovery-kubernetes/): Deploy GPU Feature Discovery (GFD) to auto-label Kubernetes nodes with GPU model, MIG capability, CUDA version, and driver info for intelligent scheduling.
- [NVIDIA GPU Operator GitOps on OpenShift](https://kubernetes.recipes/recipes/ai/nvidia-gpu-operator-gitops-openshift/): Deploy NVIDIA GPU Operator on OpenShift via GitOps with ArgoCD. Covers ClusterPolicy configuration, DCGM exporter, drain settings, tolerations, and rolling
- [Install NVIDIA GPU Operator on Kubernetes](https://kubernetes.recipes/recipes/ai/nvidia-gpu-operator-install/): Deploy the NVIDIA GPU Operator to automate GPU driver, container toolkit, and device plugin management across your Kubernetes cluster.
- [NVIDIA GPU Operator Troubleshooting](https://kubernetes.recipes/recipes/troubleshooting/nvidia-gpu-operator-troubleshooting/): Fix common NVIDIA GPU Operator issues on Kubernetes. Driver pod crashes, toolkit failures, device plugin not ready, and validation pod errors.
- [NVIDIA GPU Topology Matrix Interpretation on Kubernetes](https://kubernetes.recipes/recipes/ai/nvidia-gpu-topology-matrix-kubernetes/): Read and interpret nvidia-smi topo and nvidia-device-plugin topology matrices on Kubernetes GPU nodes. Understand X, NV, SYS, NODE, PIX, PXB, PHB connection
- [NVIDIA H200 GPU Workloads on Kubernetes](https://kubernetes.recipes/recipes/ai/nvidia-h200-gpu-kubernetes/): Deploy and optimize AI workloads on NVIDIA H200 GPUs with 141GB HBM3e memory for large model inference and training on Kubernetes.
- [NVIDIA H300 vs H100 vs H200 GPU Setup on Kubernetes](https://kubernetes.recipes/recipes/ai/nvidia-h300-gpu-kubernetes/): Compare NVIDIA H300 (upcoming), H100, and H200 GPUs for Kubernetes. Specs, memory bandwidth, deployment with GPU Operator, and inference optimization.
- [NVIDIA NeMo Training on Kubernetes](https://kubernetes.recipes/recipes/ai/nvidia-nemo-training-kubernetes/): Deploy NVIDIA NeMo framework on Kubernetes for large language model pre-training, fine-tuning, and RLHF with multi-node GPU clusters.
- [NVIDIA Network Operator NicClusterPolicy](https://kubernetes.recipes/recipes/networking/nvidia-network-operator-nicclusterpolicy/): Deploy NVIDIA Network Operator on OpenShift with NicClusterPolicy for DOCA telemetry, NIC feature discovery, RDMA IPAM, and OFED drivers. GitOps-managed
- [RDMA Configuration with NVIDIA Network Operator](https://kubernetes.recipes/recipes/networking/nvidia-network-operator-rdma-kubernetes/): Deploy and configure RDMA for GPU clusters using the NVIDIA Network Operator. NicClusterPolicy setup, MLNX_OFED driver container, shared and SR-IOV RDMA device
- [NVIDIA NIC Driver Container Entrypoint](https://kubernetes.recipes/recipes/networking/nvidia-nic-driver-container-entrypoint/): Understand and customize the NVIDIA NIC driver container entrypoint for MOFED and DOCA driver lifecycle on Kubernetes and OpenShift.
- [NVIDIA Nsight Operator for GPU Profiling on Kubernetes](https://kubernetes.recipes/recipes/observability/nvidia-nsight-operator-gpu-profiling-kubernetes/): Deploy NVIDIA Nsight Systems and Nsight Compute on Kubernetes for GPU workload profiling. Capture kernel traces, memory bandwidth, SM occupancy, and NCCL
- [NVIDIA Open GPU GPUDirect RDMA with DOCA-OFED](https://kubernetes.recipes/recipes/ai/nvidia-open-gpu-dma-buf-gpudirect-rdma-doca-sriov/): Deploy NVIDIA AI networking on Kubernetes: Open GPU driver with DMA-BUF, GPUDirect RDMA, DOCA-OFED, and SR-IOV VF isolation.
- [NVIDIA OpenShell Sandboxed AI Agent Runtime on Kubernetes](https://kubernetes.recipes/recipes/ai/nvidia-openshell-sandboxed-ai-agents-kubernetes/): Deploy NVIDIA OpenShell on Kubernetes for safe, private autonomous AI agent execution. Declarative YAML network policies, sandboxed containers
- [NVIDIA PeerMem for GPU-Direct RDMA](https://kubernetes.recipes/recipes/ai/nvidia-peermem-gpu-direct-rdma-module/): Install and configure nvidia_peermem kernel module to enable GPU-Direct RDMA between NVIDIA GPUs and Mellanox RDMA NICs. Covers module
- [Fix modprobe nvidia_peermem Invalid Argument](https://kubernetes.recipes/recipes/ai/nvidia-peermem-gpudirect-rdma-k8s/): nvidia-peermem won't load or ib_register_peer_memory_client is missing? Fix GPUDirect RDMA on Kubernetes: MOFED load order, GPU Operator settings, or DMA-BUF.
- [NVIDIA PyTorch Container on Kubernetes](https://kubernetes.recipes/recipes/ai/nvidia-pytorch-container-kubernetes/): Deploy nvcr.io/nvidia/pytorch containers on Kubernetes for GPU training. Version selection, CUDA compatibility, multi-node DDP, and NCCL configuration.
- [NVIDIA Pyxis and Enroot for SLURM](https://kubernetes.recipes/recipes/ai/nvidia-pyxis-slurm-containers/): Use NVIDIA Pyxis and Enroot to run GPU containers in SLURM jobs. Bridge SLURM HPC scheduling with container-native AI workloads and NGC images.
- [nvidia-smi Monitoring in K8s Pods](https://kubernetes.recipes/recipes/ai/nvidia-smi-kubernetes-monitoring/): Run nvidia-smi inside Kubernetes pods for GPU monitoring. Memory usage, temperature, utilization, and automated health checks with liveness probes.
- [NVLink Bridge Architecture for GPU Kubernetes Nodes](https://kubernetes.recipes/recipes/ai/nvlink-bridge-architecture-gpu-kubernetes/): Understand NVLink Bridge logical architecture in GPU servers for Kubernetes. Dual-socket PCIe Gen5 topology, NVL4 groups, GPU-NIC-NVMe placement, PCIe switch
- [NXDOMAIN DNS Troubleshooting Kubernetes](https://kubernetes.recipes/recipes/troubleshooting/nxdomain-dns-troubleshooting-kubernetes/): Fix NXDOMAIN errors in Kubernetes. Debug CoreDNS failures, ndots configuration, search domain issues, and external DNS lookup problems.
- [Use oc adm drain Dry-Run for Diagnostics](https://kubernetes.recipes/recipes/configuration/oc-adm-drain-dry-run-diagnostics/): Preview node drain impact without evicting pods. Identify PDB violations, unmanaged pods, and local storage blockers before maintenance.
- [oc-mirror Plugin Disconnected OpenShift](https://kubernetes.recipes/recipes/deployments/oc-mirror-disconnected-openshift/): Use oc-mirror to mirror OpenShift content for disconnected installations. ImageSetConfiguration, incremental mirrors, and operator catalog mirroring.
- [oc-mirror Troubleshooting: Fix Common Errors](https://kubernetes.recipes/recipes/troubleshooting/oc-mirror-troubleshooting-disconnected/): Fix oc-mirror errors in disconnected OpenShift: unexpected EOF writing layers, registry 401/403, disk space, v1/v2 mismatch, graph image and delta issues.
- [OCI Container Image Internals on Kubernetes](https://kubernetes.recipes/recipes/configuration/oci-container-image-internals-kubernetes/): Understand OCI container image internals: layers as tar archive diffs, image configuration JSON, content-addressable storage with SHA-256, multi-platform image
- [Octopus Deploy 2025.4 on Kubernetes](https://kubernetes.recipes/recipes/deployments/octopus-deploy-kubernetes-agent/): Deploy Octopus Deploy 2025.4 with MSSQL and Kubernetes agent on k3s. Release orchestration with ephemeral preview environments, approval gates.
- [OLM Disconnected OpenShift Operators](https://kubernetes.recipes/recipes/deployments/olm-disconnected-openshift/): Use Operator Lifecycle Manager in disconnected OpenShift clusters. Mirror catalogs, create CatalogSources, and manage Operators without internet access.
- [Fix OLM ResolutionFailed in Disconnected OpenShift](https://kubernetes.recipes/recipes/troubleshooting/olm-subscription-catalogsource-disconnected/): Fix ConstraintsNotSatisfiable and UnhealthyCatalogSourceFound errors when an OLM Subscription references a catalog that doesn't exist in your mirrored cluster.
- [OpenClaw GitOps Deployment with ArgoCD](https://kubernetes.recipes/recipes/deployments/openclaw-argocd-gitops/): Deploy OpenClaw on Kubernetes using ArgoCD for GitOps automation. Application definition, sync policies, drift detection, and secrets.
- [OpenClaw Auto-Scaling with KEDA](https://kubernetes.recipes/recipes/autoscaling/openclaw-autoscaling-keda/): Scale OpenClaw agents based on message queue depth using KEDA event-driven autoscaling for Discord, Telegram, and Slack.
- [OpenClaw Backup Restore Command Guide](https://kubernetes.recipes/recipes/storage/openclaw-backup-restore/): OpenClaw backup and restore command guide. VolumeSnapshots, CronJobs to S3, disaster recovery procedures, and session state management on Kubernetes.
- [OpenClaw Blue-Green Deployment](https://kubernetes.recipes/recipes/deployments/openclaw-blue-green-deployment/): Implement zero-downtime OpenClaw upgrades using blue-green deployments with traffic switching and rollback in Kubernetes.
- [OpenClaw Cron Jobs and Heartbeats](https://kubernetes.recipes/recipes/configuration/openclaw-cron-heartbeat-kubernetes/): Configure OpenClaw's built-in cron scheduling and heartbeat system on Kubernetes for proactive notifications, periodic checks, and automated background.
- [Build a Custom OpenClaw Docker Image for K8s](https://kubernetes.recipes/recipes/deployments/openclaw-custom-docker-image/): Create an optimized Docker image for OpenClaw with pre-installed dependencies, custom skills, and workspace files for faster Kubernetes deployments.
- [Run an OpenClaw Discord Bot on Kubernetes](https://kubernetes.recipes/recipes/deployments/openclaw-discord-bot-kubernetes/): Deploy OpenClaw as a Discord bot on Kubernetes with channel routing, mention handling, group chat rules, and persistent conversation memory.
- [OpenClaw API Keys External Secrets Operator](https://kubernetes.recipes/recipes/security/openclaw-external-secrets-kubernetes/): Manage OpenClaw API keys and gateway tokens using External Secrets Operator with AWS Secrets Manager, Vault, or GCP Secret Manager on Kubernetes.
- [High Availability OpenClaw with Kubernetes](https://kubernetes.recipes/recipes/deployments/openclaw-ha-kubernetes/): Run OpenClaw in a high-availability configuration on Kubernetes with health checks, automatic restarts, backup strategies, and monitoring for.
- [OpenClaw Helm Chart with Chromium Sidecar](https://kubernetes.recipes/recipes/helm/openclaw-helm-chart-chromium/): Deploy OpenClaw using the community Helm chart with Chromium browser sidecar for web automation, declarative skill installation, and custom values overlays.
- [Expose OpenClaw via K8s Ingress with TLS](https://kubernetes.recipes/recipes/networking/openclaw-ingress-tls-kubernetes/): Configure Kubernetes Ingress with TLS to expose OpenClaw gateway securely. Covers cert-manager, NGINX Ingress, and allowed origins.
- [OpenClaw Local Development with Kind](https://kubernetes.recipes/recipes/deployments/openclaw-kind-local-development/): Set up a local Kind cluster for OpenClaw development and testing. Auto-detect Docker or Podman, create a single-node cluster, and deploy OpenClaw in minutes.
- [Deploy OpenClaw AI Gateway on Kubernetes](https://kubernetes.recipes/recipes/deployments/openclaw-kubernetes-deployment/): Deploy the OpenClaw multi-channel AI gateway on Kubernetes with persistent storage, TLS ingress, and high availability for WhatsApp, Telegram, Discord.
- [OpenClaw Multi-Env Deploy with Kustomize](https://kubernetes.recipes/recipes/deployments/openclaw-kustomize-overlays/): Deploy OpenClaw across dev, staging, and production Kubernetes environments using Kustomize overlays for configs and secrets.
- [OpenClaw Health Probes on Kubernetes](https://kubernetes.recipes/recipes/observability/openclaw-liveness-readiness-probes/): Configure liveness and readiness probes for OpenClaw on Kubernetes. Custom Node.js health checks against /healthz and /readyz endpoints with proper timing.
- [OpenClaw Logging with EFK Stack](https://kubernetes.recipes/recipes/observability/openclaw-logging-efk/): Collect and analyze OpenClaw agent logs using Elasticsearch, Fluent Bit, and Kibana (EFK stack) for debugging and audit trails.
- [Monitor OpenClaw with Prometheus and Grafana](https://kubernetes.recipes/recipes/observability/openclaw-monitoring-prometheus/): Set up monitoring for OpenClaw AI gateway on Kubernetes with Prometheus metrics, Grafana dashboards, and alerting for uptime, message throughput, and.
- [Multi-Agent Routing with OpenClaw](https://kubernetes.recipes/recipes/deployments/openclaw-multi-agent-kubernetes/): Configure multiple isolated AI agents in a single OpenClaw gateway on Kubernetes with per-agent workspaces, channel bindings, and session isolation.
- [OpenClaw Multi-Agent Team Deployment](https://kubernetes.recipes/recipes/deployments/openclaw-multi-agent-team-kubernetes/): Deploy multiple specialized OpenClaw agents as Kubernetes pods. Dedicated DevOps, security, and writing agents with shared workspace.
- [OpenClaw Multi-Model Provider Setup](https://kubernetes.recipes/recipes/configuration/openclaw-multi-model-providers-kubernetes/): Configure OpenClaw with multiple AI providers on Kubernetes. Anthropic, OpenAI, Gemini, OpenRouter with fallback chains and cost control.
- [OpenClaw Network Policies: Default-Deny & Gradual Rollout](https://kubernetes.recipes/recipes/security/openclaw-networkpolicy-security/): Default-deny NetworkPolicies for OpenClaw: restrict egress to APIs, test with audit mode, and handle edge cases before enforcement.
- [OpenClaw Node Pairing for IoT and Edge Devices](https://kubernetes.recipes/recipes/configuration/openclaw-node-pairing-kubernetes/): Pair phones, Raspberry Pi, and edge devices with OpenClaw on Kubernetes. Camera, location, screen control, and remote command execution.
- [OpenClaw on OpenShift with SCCs and Routes](https://kubernetes.recipes/recipes/deployments/openclaw-openshift-deployment/): Deploy OpenClaw on OpenShift with Security Context Constraints, Routes for TLS termination, and OpenShift-specific considerations for non-root containers.
- [OpenClaw Operator for Kubernetes](https://kubernetes.recipes/recipes/deployments/openclaw-operator-kubernetes/): Deploy OpenClaw AI agents on Kubernetes using the official operator. CRD-based lifecycle, Chromium sidecar, auto-update, and backup.
- [OpenClaw with Persistent Storage](https://kubernetes.recipes/recipes/storage/openclaw-persistent-storage/): Configure persistent storage for OpenClaw workspaces using PVCs, StorageClasses, and backup strategies in Kubernetes clusters.
- [OpenClaw Persistent State Management](https://kubernetes.recipes/recipes/storage/openclaw-pvc-state-management/): Manage OpenClaw agent state and workspace data with Kubernetes PVCs. Init container config seeding, backups, and storage classes.
- [OpenClaw RBAC and Multi-Tenant Isolation](https://kubernetes.recipes/recipes/security/openclaw-rbac-multi-tenant/): Configure OpenClaw RBAC policies and namespace isolation for multi-tenant Kubernetes clusters with per-team agent access controls.
- [OpenClaw Resource Limits and Tuning](https://kubernetes.recipes/recipes/autoscaling/openclaw-resource-limits-tuning/): Size CPU, memory, and storage for OpenClaw on Kubernetes. Tuning profiles for light usage, browser automation, and production deployments.
- [Secure Secrets Management for OpenClaw](https://kubernetes.recipes/recipes/security/openclaw-secrets-management/): Manage API keys, bot tokens, and credentials for OpenClaw on Kubernetes using Kubernetes Secrets, External Secrets Operator, and Sealed Secrets.
- [OpenClaw Pod Security Hardening on Kubernetes](https://kubernetes.recipes/recipes/security/openclaw-security-hardening-pod/): Harden OpenClaw pods with read-only filesystem, dropped capabilities, non-root user, seccomp profiles, and resource limits.
- [Deploy an OpenClaw Signal Messenger Bot](https://kubernetes.recipes/recipes/deployments/openclaw-signal-kubernetes/): Run OpenClaw as a Signal messenger AI assistant on Kubernetes with linked device pairing, end-to-end encryption, and persistent sessions.
- [Manage OpenClaw Skills on Kubernetes](https://kubernetes.recipes/recipes/configuration/openclaw-skills-kubernetes/): Deploy and manage OpenClaw agent skills (tools, automations, integrations) on Kubernetes using ConfigMaps, PVCs, and git-sync for dynamic capability.
- [Deploy an OpenClaw Telegram Bot on Kubernetes](https://kubernetes.recipes/recipes/deployments/openclaw-telegram-bot-kubernetes/): Run OpenClaw as a Telegram bot on Kubernetes with BotFather setup, webhook configuration, inline commands, and persistent conversation history.
- [OpenClaw Webhook Automation on Kubernetes](https://kubernetes.recipes/recipes/deployments/openclaw-webhook-automation-kubernetes/): Configure OpenClaw webhooks on Kubernetes for GitHub, Jira, and PagerDuty event-driven automation. Ingress routing, HMAC validation, and hook handler patterns.
- [Self-Host an OpenClaw WhatsApp AI Assistant](https://kubernetes.recipes/recipes/deployments/openclaw-whatsapp-kubernetes/): Deploy OpenClaw on Kubernetes to run a personal WhatsApp AI assistant with QR code pairing, persistent sessions, media support, and allow-list security.
- [GitOps for OpenClaw Workspaces on Kubernetes](https://kubernetes.recipes/recipes/configuration/openclaw-workspace-gitops/): Manage OpenClaw agent workspaces (SOUL.md, skills, memory) with GitOps using Flux or ArgoCD, enabling version-controlled AI persona management on.
- [OpenMPI Control Plane Separation for NCCL RDMA](https://kubernetes.recipes/recipes/ai/openmpi-control-plane-separation-nccl-rdma/): Configure OpenMPI to use eth0 for MPI control traffic while NCCL uses net1 SR-IOV for data. Covers btl_tcp_if_include, pml, routed direct, plm_rsh_agent SSH
- [OpenShift 4.20: New Features and Upgrade Guide](https://kubernetes.recipes/recipes/deployments/openshift-4-20-new-features-guide/): OpenShift 4.20 (EUS) new features, Kubernetes 1.33 alignment, the upgrade path from 4.18, and what administrators need to know before upgrading.
- [OpenShift 4.21: New Features and Upgrade Guide](https://kubernetes.recipes/recipes/deployments/openshift-4-21-new-features-guide/): OpenShift 4.21 new features, K8s 1.34 alignment, upgrade from 4.20. Non-EUS release with latest innovations: in-place pod resize GA, DRA improvements.
- [OpenShift ACS Security for Kubernetes](https://kubernetes.recipes/recipes/security/openshift-acs-kubernetes/): Deploy and configure Red Hat Advanced Cluster Security (ACS/RHACS) for vulnerability scanning, compliance, network policies, and runtime threat detection.
- [OpenShift ACS RHACS Security Guide](https://kubernetes.recipes/recipes/security/openshift-acs-rhacs-security-guide/): Deploy Red Hat Advanced Cluster Security (RHACS/ACS) on OpenShift. Vulnerability scanning, compliance, runtime threat detection, and policy enforcement.
- [OpenShift BuildConfig with ImageStream](https://kubernetes.recipes/recipes/deployments/openshift-buildconfig-imagestream/): Build container images on OpenShift using BuildConfig with ImageStream triggers, pushing to internal registry or local Quay.
- [OpenShift BuildConfig with Local Quay Registry](https://kubernetes.recipes/recipes/deployments/openshift-buildconfig-quay/): Build container images on OpenShift and push to a local Quay registry using BuildConfig, ImageStream, and robot account credentials.
- [Create Custom CatalogSources for OLM Operators](https://kubernetes.recipes/recipes/configuration/openshift-catalogsource-custom-operator/): Configure CatalogSource in OpenShift to serve custom operator catalogs from private registries or air-gapped environments.
- [Filter CatalogSource Operators by Package](https://kubernetes.recipes/recipes/security/openshift-catalogsource-filtering/): Curate a minimal CatalogSource with only approved operators using opm index pruning and file-based catalog filtering for security and compliance.
- [Troubleshoot CatalogSource and OLM Issues](https://kubernetes.recipes/recipes/troubleshooting/openshift-catalogsource-troubleshooting/): Debug CatalogSource failures including pod crashes, gRPC errors, stale caches, and operator install problems in OpenShift OLM environments.
- [OpenShift Cluster Operator Upgrade Debug](https://kubernetes.recipes/recipes/troubleshooting/openshift-cluster-operator-upgrade-debug/): Debug degraded cluster operators during OpenShift upgrades. Identify stuck operators, decode status conditions, and unblock stalled rollouts.
- [OpenShift Cluster Update Process Explained](https://kubernetes.recipes/recipes/configuration/openshift-cluster-update-process-explained/): Complete guide to OpenShift Container Platform cluster updates. CVO workflow, Runlevels, Machine Config Operator node updates, update channels
- [OpenShift Cluster-Wide Pull Secret Setup](https://kubernetes.recipes/recipes/security/openshift-cluster-wide-pull-secret/): Replace admin credentials in the OpenShift cluster-wide pull secret with a Quay robot account for secure, auditable container image pulls across all namespaces.
- [crun vs runc: Container Runtimes in OpenShift](https://kubernetes.recipes/recipes/troubleshooting/openshift-crun-runc-runtime-differences/): Understand why pods behave differently on GPU vs CPU nodes in OpenShift. Compare crun and runc container runtimes, seccomp profiles, and syscall filtering.
- [OpenShift Custom CA for Private Registries](https://kubernetes.recipes/recipes/security/openshift-custom-ca-registry/): Configure OpenShift to trust a custom Certificate Authority for private container registries using additionalTrustedCA and image.config.openshift.io settings.
- [OpenShift oc debug Mount Limitation](https://kubernetes.recipes/recipes/troubleshooting/openshift-debug-mount-limitation/): Why NFS and filesystem mounts via oc debug node disappear after the debug pod exits. Understand the container namespace isolation and use MachineConfig instead.
- [Deploy a New Certificate Each OpenShift Tenant](https://kubernetes.recipes/recipes/security/openshift-deploy-new-certificate-per-tenant/): Replace and activate new TLS certificates tenant by tenant in OpenShift IngressController deployments with verification steps and rollback guidance.
- [OpenShift EgressFirewall with Helm ApplicationSet](https://kubernetes.recipes/recipes/networking/openshift-egressfirewall-gitops-helm-applicationset/): Generate tenant-specific OVN-Kubernetes EgressFirewall objects from inventory files using Helm and OpenShift GitOps ApplicationSet.
- [Kustomize Deployments with OpenShift GitOps](https://kubernetes.recipes/recipes/deployments/openshift-gitops-kustomize/): Use Kustomize overlays with the OpenShift GitOps Operator (ArgoCD) to manage environment-specific configurations across dev, staging, and production clusters.
- [OpenShift GPU Node Resource Planning](https://kubernetes.recipes/recipes/ai/openshift-gpu-node-resource-planning/): Plan CPU, memory, and overhead budgets for GPU nodes running NVIDIA GPU Operator, Network Operator, Run:ai, and OpenShift infrastructure Pods. Understand what
- [OpenShift IDMS & install-config.yaml Mirroring](https://kubernetes.recipes/recipes/deployments/openshift-idms-install-config/): Configure ImageDigestMirrorSet and install-config.yaml imageContentSources for OpenShift disconnected installations with mirror registries.
- [OpenShift IDMS ITMS Mirror Rules Guide](https://kubernetes.recipes/recipes/configuration/openshift-idms-itms-mirror-rules/): Configure IDMS and ITMS mirror rules in OpenShift for disconnected registries. NeverContactSource vs AllowContactingSource and ICSP migration.
- [OpenShift Ingress Router Troubleshooting](https://kubernetes.recipes/recipes/troubleshooting/openshift-ingress-router-troubleshooting/): Debug OpenShift HAProxy router issues: pods stuck Pending, hostPort conflicts, PDB violations during maintenance, and custom router deployment scaling problems.
- [OpenShift ITMS ImageTagMirrorSet](https://kubernetes.recipes/recipes/deployments/openshift-itms-image-tag-mirror/): Configure ImageTagMirrorSet in OpenShift 4.13+ for tag-based image mirroring. Mirror container images by tag instead of digest for disconnected clusters.
- [OpenShift MachineConfig and MCP Deep Dive](https://kubernetes.recipes/recipes/configuration/openshift-machineconfig-mcp-guide/): Master MachineConfig and MachineConfigPool on OpenShift. Configure kernel args, files, systemd units, and manage rolling node updates with MCP strategies.
- [Debug MachineConfigDaemon Logs](https://kubernetes.recipes/recipes/troubleshooting/openshift-mcd-logs-debugging/): Read and interpret OpenShift MachineConfigDaemon logs to diagnose node update failures. Common error patterns, drain issues, and config application problems.
- [OpenShift MachineConfigPool After ITMS](https://kubernetes.recipes/recipes/deployments/openshift-mcp-itms-rollout/): Monitor and manage MachineConfigPool rollouts after applying ImageTagMirrorSet in OpenShift. Handle node restarts, paused pools, and degraded states.
- [OpenShift MCP Validation Broken Rules](https://kubernetes.recipes/recipes/troubleshooting/openshift-mcp-validation-broken-rules/): Validate MachineConfigPool rules before applying in OpenShift. Detect broken MachineConfigs, degraded MCPs, and implement pre-flight checks.
- [OpenShift Multi-Tenant TLS per IngressController](https://kubernetes.recipes/recipes/security/openshift-multi-tenant-certificates/): Set up tenant-isolated TLS in OpenShift by assigning a dedicated certificate Secret to each IngressController for multi-tenant routing security.
- [OpenShift Multus CNI Multiple Network Interfaces](https://kubernetes.recipes/recipes/networking/openshift-multus-cni-multiple-networks/): Attach multiple network interfaces to Pods using Multus CNI on OpenShift. Covers NetworkAttachmentDefinitions, SR-IOV, macvlan, IPVLAN, traffic separation
- [OpenShift NVIDIA MIG Reconfiguration Without Reboot](https://kubernetes.recipes/recipes/ai/openshift-nvidia-mig-reconfiguration/): Reconfigure NVIDIA MIG geometry on OpenShift without rebooting nodes. Use nvidia-mig-manager with node labels to dynamically switch GPU partitions.
- [Debug OpenShift OAuth Login Failures](https://kubernetes.recipes/recipes/troubleshooting/openshift-oauth-login-failures/): Troubleshoot OpenShift console and CLI login failures. Check OAuth server pods, identity provider config, and expired tokens.
- [OpenShift OAuth Proxy Sidecar Guide](https://kubernetes.recipes/recipes/security/openshift-oauth-proxy-guide/): Protect K8s services with OpenShift OAuth proxy sidecar. Authentication, RBAC delegation, and SSO for internal dashboards.
- [OpenShift oc cp File Copy Guide](https://kubernetes.recipes/recipes/troubleshooting/openshift-oc-cp-file-copy/): Use oc cp to copy files and directories between local machine and Pods. Covers tar-based transfer, container selection, large file handling, and comparison
- [OpenShift oc rsync File Transfer](https://kubernetes.recipes/recipes/troubleshooting/openshift-oc-rsync-file-transfer/): Use oc rsync to copy files between local machine and Pods in OpenShift. Covers upload, download, live sync, filtering, and common patterns for debugging
- [OpenShift OIDC Claims Mapping Troubleshooting](https://kubernetes.recipes/recipes/troubleshooting/openshift-oidc-claims-mapping-troubleshooting/): Fix OpenShift OIDC login failures caused by claims mapping mismatches and wrong issuer URLs. Debug the OAuth CR openID identityProvider config.
- [Disable OperatorHub Default Sources](https://kubernetes.recipes/recipes/configuration/openshift-operatorhub-disable-sources/): Disable default OperatorHub catalog sources in OpenShift for air-gapped clusters. Use OperatorHub CR to disable individual or all sources with Ansible auto.
- [OpenShift Project Request Template Pull Secret](https://kubernetes.recipes/recipes/configuration/openshift-project-request-template/): Configure an OpenShift Project Request Template so every new namespace automatically gets a ServiceAccount with imagePullSecrets for your private Quay registry.
- [Configure PDBs for OpenShift Routers](https://kubernetes.recipes/recipes/configuration/openshift-router-pdb-configuration/): Set PodDisruptionBudgets for OpenShift IngressController routers. Balance availability during maintenance with node drain ability.
- [OpenShift Routes vs Ingress Guide](https://kubernetes.recipes/recipes/networking/openshift-routes-vs-ingress/): Compare OpenShift Routes and Kubernetes Ingress. Covers edge, passthrough, and re-encrypt TLS termination, and when to use each option.
- [OpenShift SCC Security Context Guide](https://kubernetes.recipes/recipes/security/openshift-scc-guide-2026/): Configure OpenShift Security Context Constraints for pods. Restricted, anyuid, privileged SCCs, custom SCC, and migration to PSA.
- [OpenShift SCC: Security Context Constraints](https://kubernetes.recipes/recipes/security/openshift-scc-security-context-constraints/): Configure Security Context Constraints on OpenShift. Manage SCCs for pods requiring privileged access, host networking, custom UID/GID, and volume types.
- [OpenShift Serverless KnativeServing](https://kubernetes.recipes/recipes/deployments/openshift-serverless-knativeserving/): Deploy and configure OpenShift Serverless Operator with KnativeServing for autoscaling, scale-to-zero, and traffic splitting on Kubernetes.
- [Fix SR-IOV 'Not Enough MMIO Resources' Error](https://kubernetes.recipes/recipes/networking/openshift-sriov-mmio-resources-fix/): Resolve the mlx5_core 'not enough MMIO resources for SR-IOV' error on OpenShift nodes with Mellanox ConnectX NICs. Covers BIOS settings, PCIe BAR
- [OpenShift SR-IOV Network with NVIDIA IPAM for GPU Fabric](https://kubernetes.recipes/recipes/networking/openshift-sriov-network-nv-ipam-gpu-fabric/): Configure SriovNetwork resources on OpenShift with nv-ipam for GPU fabric IP allocation. SR-IOV Network Operator setup, Mellanox NIC resource targeting, IPAM
- [OpenShift SR-IOV RDMA InfiniBand Device Plugin](https://kubernetes.recipes/recipes/networking/openshift-sriov-rdma-infiniband-device-plugin/): Configure and troubleshoot SR-IOV Network Operator with Mellanox ConnectX RDMA InfiniBand devices on OpenShift. Covers SriovNetworkNodePolicy, device
- [Create SR-IOV VFs on OpenShift Operator](https://kubernetes.recipes/recipes/networking/openshift-sriov-vf-creation/): Use the OpenShift SR-IOV Network Operator to create and manage Virtual Functions from selected Physical Functions on GPU worker nodes.
- [OpenShift Versions, EOL & Kubernetes Mapping](https://kubernetes.recipes/recipes/configuration/openshift-support-lifecycle-version-matrix/): OpenShift lifecycle: EOL dates for OCP 4.12–4.22, OpenShift to Kubernetes version mapping, EUS vs standard support, release cycle and upgrade paths.
- [Rotate OpenShift Tenant Secrets Safely](https://kubernetes.recipes/recipes/security/openshift-tenant-secret-rotation/): Implement low-risk secret rotation in OpenShift multi-tenant environments using versioned Secrets and controlled rollouts.
- [OpenShift Upgrade Disconnected Cluster](https://kubernetes.recipes/recipes/deployments/openshift-upgrade-disconnected-environment/): Step-by-step guide to upgrading OpenShift in a disconnected air-gapped environment. Mirror releases, configure ICSP/IDMS, validate, and execute the upgrade.
- [OpenShift Upgrade Planning for 2026](https://kubernetes.recipes/recipes/configuration/openshift-upgrade-planning-2026/): Plan OpenShift upgrades for 2026. EUS-to-EUS paths, operator compatibility, pre-upgrade checks, canary node pools, and rollback strategy for OCP 4.14 to 4.18.
- [OpenShift Upgrade Service Graph Guide](https://kubernetes.recipes/recipes/deployments/openshift-upgrade-service-graph/): Use the OpenShift Upgrade Service (OSUS) and Cincinnati graph to plan safe upgrade paths. Channel selection, conditional edges, and air-gapped graph data.
- [OpenShift User Account Management](https://kubernetes.recipes/recipes/security/openshift-user-account-management/): Manage user accounts in OpenShift: create users, assign roles, configure identity providers, manage groups, and implement RBAC for multi-tenant clusters.
- [Enable User Workload Monitoring OpenShift](https://kubernetes.recipes/recipes/observability/openshift-user-workload-monitoring/): Enable user workload monitoring on OpenShift. Deploy ServiceMonitor, PodMonitor, alerting rules, and Grafana dashboards.
- [OpenTelemetry Complete Setup on Kubernetes](https://kubernetes.recipes/recipes/observability/opentelemetry-kubernetes-observability/): Deploy OpenTelemetry Collector, auto-instrumentation, and exporters on Kubernetes. Unified traces, metrics, and logs pipeline to Jaeger, Prometheus, and Loki.
- [Fix Stuck OLM Operator Subscriptions](https://kubernetes.recipes/recipes/troubleshooting/operator-subscription-stuck/): Debug Operator Lifecycle Manager subscriptions stuck in pending or failed state. Resolve catalog source issues, approval policies, and CSV dependency conflicts.
- [OSUS Direct vs Replicated OpenShift](https://kubernetes.recipes/recipes/configuration/osus-direct-vs-replicated-openshift/): Choose between direct and replicated OSUS graph data modes in OpenShift. Configure UpdateService for connected and disconnected environments.
- [OSUS Operator Disconnected OpenShift](https://kubernetes.recipes/recipes/deployments/osus-operator-disconnected-openshift/): Deploy the OpenShift Update Service (OSUS) operator for disconnected clusters. Local Cincinnati graph, graph-data image mirroring, and upgrade path serving.
- [PDB Allowed Disruptions Zero: Debugging](https://kubernetes.recipes/recipes/troubleshooting/pdb-allowed-disruptions-zero/): Debug PodDisruptionBudgets stuck at zero allowed disruptions. Understand minAvailable vs maxUnavailable, fix eviction failures, and plan for maintenance.
- [Fix PVC Resize Stuck or Failed](https://kubernetes.recipes/recipes/storage/persistent-volume-resize-troubleshooting/): Debug PVC expansion failures in Kubernetes. Covers allowVolumeExpansion, filesystem resize, and offline vs online expansion.
- [Fix PV Stuck in Terminating State](https://kubernetes.recipes/recipes/troubleshooting/persistent-volume-stuck-terminating/): Resolve PVs and PVCs stuck in Terminating status. Remove finalizers safely, check volume detachment, and handle storage issues.
- [PFC Configuration on Mellanox ConnectX NICs](https://kubernetes.recipes/recipes/networking/pfc-mellanox-connectx-configuration/): Enable Priority Flow Control on Mellanox ConnectX-6/7 NICs for lossless RoCE. mlnx_qos, cma_roce_mode, DSCP trust, ECN, and firmware-level PFC verification.
- [Configure PFC with NMState on Kubernetes](https://kubernetes.recipes/recipes/networking/pfc-nmstate-roce-lossless-kubernetes/): Enable Priority Flow Control (PFC) for lossless RDMA using NMState and NodeNetworkConfigurationPolicy. Configure DSCP-to-priority mapping, ECN, and RoCEv2 QoS.
- [Physical AI and Robotics Orchestration](https://kubernetes.recipes/recipes/ai/physical-ai-robotics-kubernetes/): Orchestrate physical AI and robotics fleets with Kubernetes. ROS 2 on K8s, robot fleet management, edge-cloud hybrid, NVIDIA Isaac.
- [Platform Engineering on K8s: Build an IDP](https://kubernetes.recipes/recipes/deployments/platform-engineering-internal-developer-platform-kubernetes/): Build an internal developer platform on Kubernetes. Backstage, Crossplane, ArgoCD, self-service templates, golden paths.
- [Kubernetes Pod Disruption Budget (PDB) Guide](https://kubernetes.recipes/recipes/deployments/pod-disruption-budget-config/): Configure PodDisruptionBudgets with minAvailable or maxUnavailable to keep apps up during node drains, upgrades and autoscaler scale-down.
- [Manage hostNetwork Pod Port Allocation](https://kubernetes.recipes/recipes/networking/pod-hostnetwork-port-mapping/): Plan and manage host port usage for hostNetwork pods. Prevent port conflicts, track allocations, and handle port exhaustion.
- [How to Configure Pod Lifecycle Hooks](https://kubernetes.recipes/recipes/deployments/pod-lifecycle-hooks/): Execute custom actions during pod startup and shutdown with lifecycle hooks. Implement graceful shutdown, initialization tasks, and cleanup operations.
- [How to Use Pod Presets and Mutations](https://kubernetes.recipes/recipes/configuration/pod-mutation-injection/): Automatically inject configurations into pods using admission controllers. Configure environment variables, volumes, and annotations at deployment time.
- [Fix Pod Stuck in Pending State](https://kubernetes.recipes/recipes/troubleshooting/pod-pending-troubleshooting/): Debug pods stuck in Pending status. Covers insufficient resources, node affinity mismatches, taint/toleration issues, and PVC binding failures.
- [Pod Readiness Gates for Custom Conditions](https://kubernetes.recipes/recipes/deployments/pod-readiness-gates/): Implement Pod Readiness Gates to add custom conditions that must be satisfied before a pod is considered ready for traffic, enabling integration with.
- [Kubernetes Pod Security Standards (PSS/PSA)](https://kubernetes.recipes/recipes/security/pod-security-standards/): Enforce Kubernetes Pod Security Standards with Pod Security Admission: privileged, baseline and restricted levels, namespace labels, rollout and exemptions.
- [topologySpreadConstraints: maxSkew, minDomains, Examples](https://kubernetes.recipes/recipes/deployments/pod-topology-constraints/): Kubernetes pod topology spread constraints with YAML: maxSkew, whenUnsatisfiable, minDomains, matchLabelKeys, nodeTaintsPolicy, and cluster defaults.
- [Fix Podman TLS x509 Behind Corporate Proxy](https://kubernetes.recipes/recipes/troubleshooting/podman-tls-x509-corporate-proxy/): Resolve podman pull x509 certificate signed by unknown authority errors caused by corporate TLS-intercepting proxies. Extract and install the proxy CA.
- [Poolside AI Foundation Models on Kubernetes](https://kubernetes.recipes/recipes/ai/poolside-ai-foundation-models-kubernetes/): Deploy Poolside AI foundation models for enterprise software agents on Kubernetes. On-prem and VPC deployment, multi-agent orchestration, sandboxed
- [Post-Quantum Cryptography on Kubernetes](https://kubernetes.recipes/recipes/security/post-quantum-cryptography-kubernetes/): Prepare Kubernetes clusters for post-quantum cryptography. NIST PQC standards, hybrid TLS certificates, quantum-safe mTLS, Istio/Cilium integration.
- [Dell PowerScale NFS Access Zones for Kubernetes AI Storage](https://kubernetes.recipes/recipes/storage/powerscale-nfs-access-zones-kubernetes-ai/): Configure Dell PowerScale (Isilon) access zones and SmartConnect pools for Kubernetes AI storage with per-environment NFS isolation and IP pool sizing.
- [Preemptive Cybersecurity on Kubernetes](https://kubernetes.recipes/recipes/security/preemptive-cybersecurity-kubernetes/): Implement preemptive cybersecurity on Kubernetes. Threat prediction, automated vulnerability patching, runtime behavior analysis, CNAPP.
- [Prefix Caching vLLM KV Cache K8s](https://kubernetes.recipes/recipes/ai/prefix-caching-vllm-kubernetes/): Enable automatic prefix caching in vLLM on Kubernetes for shared-prompt workloads. KV cache reuse, memory savings, and chatbot latency optimization.
- [PriorityClasses for GPU Workloads](https://kubernetes.recipes/recipes/configuration/priorityclasses-gpu-workloads/): Configure Kubernetes PriorityClasses for GPU workloads with training, serving, batch, and interactive tiers and preemption policies.
- [Private Container Registry on Kubernetes](https://kubernetes.recipes/recipes/deployments/private-container-registry-kubernetes/): Deploy a private OCI container registry on Kubernetes with persistent storage, TLS, authentication, garbage collection, and high availability. Self-hosted
- [Fix PVC Stuck in Pending State](https://kubernetes.recipes/recipes/troubleshooting/pvc-pending-troubleshooting/): Debug PersistentVolumeClaims stuck in Pending status. Covers storage class issues, provisioner failures, capacity problems, and access mode mismatches.
- [PersistentVolumeClaims with StorageClasses](https://kubernetes.recipes/recipes/storage/pvc-storageclass-examples/): Learn how to provision persistent storage for your Kubernetes workloads using PersistentVolumeClaims and StorageClasses. Includes examples for dynamic.
- [Quantize LLMs AWQ GPTQ for K8s Deploy](https://kubernetes.recipes/recipes/ai/quantize-llm-awq-gptq-kubernetes/): Deploy AWQ and GPTQ quantized LLMs on Kubernetes. 4-bit inference with vLLM, model conversion, accuracy trade-offs, and GPU memory savings guide.
- [Quantum Computing on K8s: Hybrid Workflows](https://kubernetes.recipes/recipes/ai/quantum-computing-kubernetes-integration/): Run quantum computing workloads on Kubernetes. Qiskit, Cirq, PennyLane hybrid classical-quantum pipelines, quantum job scheduling, and QPU integration patterns.
- [Quay Default Permissions for Robot Accounts](https://kubernetes.recipes/recipes/security/quay-default-permissions-robot/): Configure Quay Registry default permissions to auto-grant read access to robot accounts on every new repository. API and team patterns.
- [Red Hat Quay Registry Kubernetes](https://kubernetes.recipes/recipes/deployments/quay-registry-kubernetes-guide/): Deploy and manage Quay container registry on Kubernetes. Mirror policies, robot accounts, security scanning, and integration with OpenShift.
- [Quay Robot Accounts for Kubernetes Image Pulls](https://kubernetes.recipes/recipes/deployments/quay-robot-account-kubernetes/): Create Quay robot accounts and configure Kubernetes imagePullSecrets for automated container image pulls from private registries.
- [Build a RAG Pipeline on Kubernetes](https://kubernetes.recipes/recipes/ai/rag-pipeline-kubernetes/): Deploy a Retrieval-Augmented Generation pipeline on Kubernetes using a vector database, embedding model, and LLM inference server.
- [Fix Ray Data PyTorch CPU Thread Oversubscription](https://kubernetes.recipes/recipes/troubleshooting/ray-data-pytorch-cpu-thread-oversubscription/): Fix GPU starvation from Ray Data CPU preprocessing on Kubernetes. Pin torch.set_num_threads(1) per worker to stop thread oversubscription.
- [Fix RBAC Permission Denied Errors](https://kubernetes.recipes/recipes/security/rbac-permission-denied-troubleshooting/): Debug RBAC forbidden and unauthorized errors in Kubernetes. Covers ClusterRole vs Role scope and service account permissions.
- [RDMA Network QoS Traffic Classes DCQCN](https://kubernetes.recipes/recipes/networking/rdma-network-qos-traffic-classes-dcqcn/): Complete RDMA network QoS architecture with traffic classes TC0-TC6, DSCP and dot1p mappings, PFC, ECN, WRED, and DCQCN congestion control for lossless RoC.
- [Rebuild NIM with Custom Model on Kubernetes](https://kubernetes.recipes/recipes/ai/rebuild-nim-custom-model-kubernetes/): Step-by-step guide to deploy custom, fine-tuned, or self-hosted models with NVIDIA NIM on Kubernetes. Model-free NIM from HuggingFace, S3, NGC, or local path.
- [Record kubectl Sessions for Kubernetes](https://kubernetes.recipes/recipes/configuration/record-kubectl-sessions-kubernetes/): Record and replay kubectl sessions for auditing, documentation, and training. Terminal recording with asciinema, script, and kubectl plugins for OpenShift.
- [Red Hat AI Studio on OpenShift](https://kubernetes.recipes/recipes/ai/red-hat-ai-studio-openshift-model-development/): Deploy Red Hat AI Studio on OpenShift for end-to-end LLM development. Model catalog, InstructLab fine-tuning, experiment tracking, model
- [Fix ResourceQuota Exceeded Errors](https://kubernetes.recipes/recipes/troubleshooting/resource-quota-exceeded-error/): Debug resource quota violations preventing pod scheduling. Understand LimitRange defaults, ResourceQuota, and namespace management.
- [ResourceQuota and LimitRange for GPUs](https://kubernetes.recipes/recipes/configuration/resourcequota-limitrange-gpu/): Configure ResourceQuota and LimitRange for GPU workloads with per-tenant caps on GPU, CPU, memory, and object counts in Kubernetes.
- [Restore Scaled Deployments After Node Drain](https://kubernetes.recipes/recipes/configuration/restore-scaled-deployments-after-drain/): Restore deployments scaled down for maintenance. Verify node health, check pod scheduling, and confirm service availability.
- [RetinaNet GPU Training on Kubernetes](https://kubernetes.recipes/recipes/ai/retinanet-gpu-training-kubernetes/): Train RetinaNet object detection models on Kubernetes with unlimited memlock for RDMA, CRI-O ulimit configuration, and multi-GPU distributed training.
- [RHACS CI/CD Pipeline Integration](https://kubernetes.recipes/recipes/security/rhacs-cicd-pipeline-integration/): Integrate Red Hat Advanced Cluster Security into CI/CD pipelines with roxctl for image scanning, policy checks, and deployment validation.
- [RHACS Compliance Scanning in OpenShift](https://kubernetes.recipes/recipes/security/rhacs-compliance-scanning/): Run CIS, NIST, PCI DSS, and HIPAA compliance scans with Red Hat Advanced Cluster Security and automate reporting for audits.
- [RHACS Custom Security Policies Guide](https://kubernetes.recipes/recipes/security/rhacs-custom-security-policies/): Create and manage custom security policies in Red Hat Advanced Cluster Security for image scanning, deployment config, and runtime enforcement.
- [RHACS Multi-Cluster Management](https://kubernetes.recipes/recipes/security/rhacs-multi-cluster-management/): Manage security across multiple Kubernetes clusters with RHACS Central hub, secured cluster registration, and unified policy enforcement.
- [RHACS Network Segmentation Policies](https://kubernetes.recipes/recipes/security/rhacs-network-segmentation/): Use Red Hat Advanced Cluster Security network graph to discover traffic flows, generate NetworkPolicies, and enforce micro-segmentation.
- [RHACS NFS Tenant Security Kubernetes](https://kubernetes.recipes/recipes/security/rhacs-nfs-tenant-security-kubernetes/): Enforce NFS tenant isolation with RHACS policies. Detect direct NFS mounts, wrong StorageClass usage, privileged escalation, and cross-tenant violations.
- [RHCOS Node Management for OpenShift](https://kubernetes.recipes/recipes/configuration/rhcos-openshift-node-management/): Understand and manage Red Hat Enterprise Linux CoreOS (RHCOS) for OpenShift nodes including MachineConfig, ignition, OS updates, and node customization.
- [RoCE PFC and ECN Lossless Ethernet for GPU Clusters](https://kubernetes.recipes/recipes/networking/roce-pfc-ecn-lossless-ethernet-gpu/): Configure RoCE v2 with Priority Flow Control (PFC) and ECN for lossless Ethernet RDMA on GPU clusters. Covers DSCP mapping, switch configuration, NIC
- [RoCEv2 End-to-End Lossless Stack](https://kubernetes.recipes/recipes/networking/rocev2-end-to-end-lossless-stack/): Complete RoCEv2 lossless fabric configuration from GPU node to switch and back. Dell OS10 switches, Mellanox NICs, OpenShift MachineConfig, PFC, ECN.
- [Fix Deploy Rollout Stuck at Partial Progress](https://kubernetes.recipes/recipes/deployments/rollout-stuck-troubleshooting/): Debug deployments stuck with unavailable replicas during rollout. Covers readiness probes, resource constraints, and rollback.
- [Rook Ceph on Kubernetes: Block, File and S3 Storage](https://kubernetes.recipes/recipes/storage/rook-ceph-storage-kubernetes/): Deploy Rook Ceph on Kubernetes: operator Helm install, CephCluster, RBD and CephFS StorageClasses, S3 object store, toolbox health checks and sizing.
- [Rotate Quay Robot Tokens in Kubernetes](https://kubernetes.recipes/recipes/security/rotate-quay-robot-tokens/): Automate Quay robot account token rotation across Kubernetes namespaces with zero-downtime credential updates and validation scripts.
- [Run NCCL Tests on Kubernetes (all_reduce_perf)](https://kubernetes.recipes/recipes/ai/run-nccl-tests-kubernetes/): Run nccl-tests all_reduce_perf on Kubernetes or OpenShift, single-node and multi-node with MPIJob, and read algbw/busbw to validate GPU interconnect bandwidth.
- [Run:ai Backend Architecture on OpenShift](https://kubernetes.recipes/recipes/ai/runai-backend-architecture-openshift/): Understand the full Run:ai backend deployment on OpenShift with 40+ microservices including Keycloak, PostgreSQL, NATS, Thanos, Traefik, and workload
- [Run:ai Distributed Inference with SR-IOV RDMA](https://kubernetes.recipes/recipes/ai/runai-distributed-inference-sriov-rdma/): Deploy distributed vLLM inference on Run:ai using SR-IOV RDMA for NCCL inter-node communication. Covers extended-resource for Mellanox VFs, network annotation
- [Run:ai Distributed Inference with vLLM and NCCL](https://kubernetes.recipes/recipes/ai/runai-distributed-inference-vllm-nccl/): Deploy distributed LLM inference on Run:ai with vLLM tensor parallelism across multiple workers. Covers multi-node GPU splitting, NCCL configuration, PVC model
- [Run:ai Distributed PyTorch Training on OpenShift](https://kubernetes.recipes/recipes/ai/runai-distributed-training-openshift/): Submit multi-node distributed PyTorch training jobs on OpenShift using Run:ai CLI. Covers DDP, FSDP, RDMA networking, and GPU scheduling.
- [Run:ai Distributed vLLM with NCCL](https://kubernetes.recipes/recipes/ai/runai-distributed-vllm-nccl-rdma/): Deploy distributed vLLM inference on Run:ai with NCCL over NVLink and RDMA. Tensor parallelism across GPUs with NCCL debug logging, SR-IOV networking.
- [Run:ai + Dynamo Multi-Node Scheduling on K8s](https://kubernetes.recipes/recipes/ai/runai-dynamo-multinode-scheduling-kubernetes/): Deploy NVIDIA Dynamo with Run:ai v2.23 for gang scheduling and topology-aware placement. Atomic pod launches, zone co-location, and disaggregated inference.
- [FSDP Distributed Training on Run:ai](https://kubernetes.recipes/recipes/ai/runai-fsdp-training-mistral-gpu/): Run PyTorch FSDP distributed training workloads on Run:ai with GPU scheduling, event tracking, and GPU memory monitoring. Covers Mistral-class model
- [Run:ai GPU Metrics Pipeline with DCGM and Thanos](https://kubernetes.recipes/recipes/ai/runai-gpu-metrics-dcgm-thanos-grafana/): End-to-end GPU metrics pipeline on Run:ai: DCGM exporter collects GPU utilization, Prometheus scrapes, remote-writes to Thanos Receive, and Grafana dashboards
- [Run:AI GPU Quotas on OpenShift](https://kubernetes.recipes/recipes/ai/runai-gpu-quotas-openshift/): Configure Run:AI scheduler quotas for fair GPU sharing with guaranteed, over-quota borrowing, and per-tenant GPU allocation policies.
- [Run:ai GPU Scheduling with Kubeflow MPIJob](https://kubernetes.recipes/recipes/ai/runai-gpu-scheduling-kubeflow-mpijob/): Integrate Run:ai GPU scheduler with Kubeflow MPIJob for multi-node NCCL workloads. Covers Run:ai project namespaces, GPU quota annotations, pod group
- [Run:ai Keycloak SSO Authentication Setup](https://kubernetes.recipes/recipes/security/runai-keycloak-sso-authentication/): Configure Run:ai SSO authentication with Keycloak on OpenShift: OIDC integration, user federation, role mapping, and troubleshooting login failures.
- [Run:ai NIM Distributed Inference Tutorial](https://kubernetes.recipes/recipes/ai/runai-nim-distributed-inference-tutorial/): Step-by-step guide to deploy DeepSeek-R1 distributed inference on Run:ai with LeaderWorkerSet, SGLang, PVC caching, and OpenShift security.
- [Run:ai Observability with OpenTelemetry](https://kubernetes.recipes/recipes/observability/runai-observability-opentelemetry-openshift/): Configure Run:ai observability on OpenShift with OpenTelemetry Collector, Prometheus receivers, metrics enrichment, OAuth2 export, and GPU metric collection
- [Run:ai Platform Backend Components](https://kubernetes.recipes/recipes/ai/runai-platform-components-architecture/): Overview of Run:ai backend StatefulSets and components on OpenShift: Thanos receive/query, Keycloak, NATS, Redis, PostgreSQL, workload controllers, and their
- [Run:ai Topology-Aware Scheduling Deep Dive](https://kubernetes.recipes/recipes/ai/runai-topology-aware-scheduling-kubernetes/): Configure Run:ai topology-aware scheduling for distributed AI workloads. Multi-level hierarchies, required vs preferred placement, LeaderWorkerSet.
- [Run:ai Training Job Submit Script Pattern](https://kubernetes.recipes/recipes/ai/runai-training-submit-script-pattern/): Production pattern for submitting Run:ai training jobs via shell scripts with GPU fractional allocation, NFS mounts, custom Python environments, and private
- [Run:ai Distrib. vLLM Inference Multimodal LLMs](https://kubernetes.recipes/recipes/ai/runai-vllm-distributed-inference-multimodal/): Deploy multimodal LLMs with Run:ai distributed inference and vLLM on Kubernetes. Tensor parallelism, NCCL over NVLink, GPUDirect RDMA.
- [Run:ai Workload Controllers on OpenShift](https://kubernetes.recipes/recipes/ai/runai-workload-controllers-openshift/): Understand Run:ai cluster-level workload controllers on OpenShift: workload-controller, workload-overseer, workload-exporter, and status-updater components.
- [Configure S3 Storage Permissions for ML Models](https://kubernetes.recipes/recipes/storage/s3-model-storage-permissions/): Set up S3 bucket ACLs, IAM roles, and PVC permissions so Kubernetes inference pods can securely read large ML model weights from object storage.
- [Scale Deployments to Unblock Node Drains](https://kubernetes.recipes/recipes/configuration/scale-deploy-unblock-drain/): Safely scale down deployments that block node drains due to PDB violations. Record original replicas, scale to zero, drain, then restore after the node returns.
- [Kubernetes Scheduler Configuration and Tuning](https://kubernetes.recipes/recipes/configuration/scheduler-configuration-tuning/): Customize the Kubernetes scheduler with scheduling profiles, plugins, and advanced placement strategies for optimal pod placement and resource utilization
- [How to Encrypt Secrets at Rest with KMS](https://kubernetes.recipes/recipes/security/secrets-encryption-kms/): Configure Kubernetes secrets encryption at rest using external KMS providers. Learn to set up AWS KMS, GCP KMS, and Azure Key Vault encryption.
- [Kubernetes Secrets Management Best Practices](https://kubernetes.recipes/recipes/security/secrets-management-best-practices/): Kubernetes Secrets management best practices: create and decode Secrets, encrypt etcd at rest, lock down RBAC, rotate safely, and sync from Vault or AWS.
- [SELinux and SCC Config for GPU Operator](https://kubernetes.recipes/recipes/security/selinux-scc-gpu-operator-openshift/): Understand SELinux device relabeling and Security Context Constraints (SCC) requirements for the NVIDIA GPU Operator driver pods on OpenShift.
- [SELinux SSH Login Failure Troubleshoot](https://kubernetes.recipes/recipes/troubleshooting/selinux-ssh-login-failure-troubleshooting/): Fix SSH login failures caused by SELinux enforcement. Diagnose AVC denials, restore file labels, fix custom SSH ports, and resolve PAM denials.
- [Kubernetes Service Account: YAML, RBAC, Tokens](https://kubernetes.recipes/recipes/security/service-accounts-rbac/): Create a Kubernetes ServiceAccount, attach it to a Deployment, grant least-privilege RBAC with Roles and bindings, and test it with kubectl auth can-i.
- [Debug Service with No Ready Endpoints](https://kubernetes.recipes/recipes/troubleshooting/service-endpoints-not-ready/): Troubleshoot Services showing zero endpoints. Verify label selectors, readiness probes, pod status, and port configuration.
- [Expose Services with LoadBalancer and NodePort](https://kubernetes.recipes/recipes/networking/service-loadbalancer-nodeport/): Learn different ways to expose Kubernetes services externally using LoadBalancer, NodePort, and ExternalIPs. Compare options for various environments.
- [Fix Service Mesh Sidecar Injection Failures](https://kubernetes.recipes/recipes/troubleshooting/service-mesh-sidecar-troubleshooting/): Debug Istio and Envoy sidecar injection issues. Covers missing sidecars, port conflicts, init container failures, and mTLS connection errors.
- [Shared RDMA Device Plugin for Kubernetes GPU Pods](https://kubernetes.recipes/recipes/networking/shared-rdma-device-plugin-kubernetes/): Configure the RDMA shared device plugin to allow multiple pods to share RDMA-capable NICs on Kubernetes. K8s-rdma-shared-dev-plugin setup, resource
- [Skopeo Container Image Operations](https://kubernetes.recipes/recipes/deployments/skopeo-container-image-operations/): Use skopeo to inspect, copy, sync, and delete container images across registries. Essential tool for disconnected Kubernetes and OpenShift environments.
- [SLURM and Kubernetes Integration](https://kubernetes.recipes/recipes/ai/slurm-kubernetes-integration/): Integrate SLURM HPC workload manager with Kubernetes for hybrid AI and scientific computing. Bridge HPC batch scheduling with container orchestration.
- [Sovereign Air-Gapped Kubernetes Clusters](https://kubernetes.recipes/recipes/security/sovereign-air-gapped-kubernetes-clusters/): Deploy sovereign and air-gapped Kubernetes clusters. Offline installation, private registry mirrors, disconnected GitOps, sovereign cloud.
- [Speculative Decoding with vLLM on Kubernetes](https://kubernetes.recipes/recipes/ai/speculative-decoding-vllm-kubernetes/): Enable speculative decoding in vLLM on Kubernetes for 2-3x faster LLM inference. Draft model selection, acceptance rates, and latency optimization.
- [SR-IOV Device Plugin PF Flag on Kubernetes](https://kubernetes.recipes/recipes/networking/sriov-device-plugin-pf-flag-kubernetes/): Configure SR-IOV device plugin PF flag in Kubernetes. Expose physical functions as allocatable resources for exclusive RDMA access.
- [SR-IOV Mixed NICs for GPU Nodes](https://kubernetes.recipes/recipes/networking/sriov-mixed-nic-gpu-nodes/): Configure SR-IOV with mixed ConnectX-7 and ConnectX-6 NICs for RDMA data plane and management traffic on GPU worker nodes.
- [SR-IOV Multus Network Attachment for GPU RDMA Pods](https://kubernetes.recipes/recipes/networking/sriov-multus-network-attachment-gpu-rdma/): Configure Multus CNI NetworkAttachmentDefinition for SR-IOV RDMA in Kubernetes GPU workloads. Covers k8s.v1.cni.cncf.io/networks annotation, IPAM
- [SR-IOV NetworkNodePolicy for RDMA](https://kubernetes.recipes/recipes/networking/sriov-network-node-policy-rdma-openshift/): Configure SriovNetworkNodePolicy on OpenShift to create RDMA-capable VFs on Mellanox ConnectX NICs for GPUDirect RDMA and high-performance AI networking.
- [SR-IOV NicClusterPolicy for VF Configuration](https://kubernetes.recipes/recipes/networking/sriov-nicclusterpolicy-vfs/): Configure SR-IOV Virtual Functions on Mellanox ConnectX NICs using the NVIDIA Network Operator NicClusterPolicy for high-performance Kubernetes networking.
- [SR-IOV VF Networking for AI Workloads](https://kubernetes.recipes/recipes/networking/sriov-vf-ai-workloads/): Deploy SR-IOV Virtual Functions with RDMA support for distributed AI training on Kubernetes, including multi-NIC pod configuration and NCCL tuning.
- [SR-IOV VF to Container Mapping and Lifecycle](https://kubernetes.recipes/recipes/networking/sriov-vf-container-mapping-lifecycle/): How SR-IOV Virtual Functions are mapped to containers in Kubernetes. Covers VF allocation flow, link state management (VFs are down when unassigned), device
- [SR-IOV VF Troubleshooting on Kubernetes](https://kubernetes.recipes/recipes/troubleshooting/sriov-vf-troubleshooting/): Diagnose and fix SR-IOV Virtual Function issues including VF creation failures, device plugin errors, RDMA problems, and network attachment failures.
- [How to Deploy Stateful Applications](https://kubernetes.recipes/recipes/storage/stateful-applications/): Run stateful workloads on Kubernetes with StatefulSets. Manage stable identities, persistent storage, and ordered deployment for databases and caches.
- [Kubernetes StatefulSet Guide with Examples](https://kubernetes.recipes/recipes/deployments/statefulset-management/): Kubernetes StatefulSet guide: stable pod names and DNS, per-pod PVCs, ordered scaling, partitioned rolling updates, PVC retention and troubleshooting.
- [How to Deploy MySQL with StatefulSet](https://kubernetes.recipes/recipes/storage/statefulset-mysql/): Deploy a production-ready MySQL database on Kubernetes using StatefulSet. Learn persistent storage, headless services, and backup strategies.
- [Strimzi Kafka Operator on Kubernetes](https://kubernetes.recipes/recipes/deployments/strimzi-kafka-operator-kubernetes/): Deploy Apache Kafka on Kubernetes with Strimzi operator. Covers Kafka CR, KafkaTopic, KafkaUser, KafkaConnect, KafkaBridge, rack awareness, storage
- [Kubernetes Finalizers: Remove Stuck Finalizers](https://kubernetes.recipes/recipes/troubleshooting/stuck-resources-finalizers/): What Kubernetes finalizers are and how to safely remove stuck ones with kubectl patch. Fix namespaces, PVCs, CRDs and Helm releases stuck in Terminating.
- [Switch to Open NVIDIA Kernel Modules on OpenShift](https://kubernetes.recipes/recipes/configuration/switch-proprietary-to-open-kernel-modules/): Step-by-step guide to migrate the NVIDIA GPU Operator from proprietary to open kernel modules on OpenShift, enabling DMA-BUF and GPUDirect Storage support.
- [Tabnine AI Code Assistant Self-Hosted on Kubernetes](https://kubernetes.recipes/recipes/ai/tabnine-enterprise-self-hosted-kubernetes/): Deploy Tabnine Enterprise self-hosted on Kubernetes for private AI code completion and chat. On-prem model serving, multi-model support (Tabnine
- [Fix Node Untolerated Taint Scheduling Errors](https://kubernetes.recipes/recipes/troubleshooting/taint-toleration-scheduling-issues/): Fix node untolerated taint errors causing pods stuck in Pending. NoSchedule, PreferNoSchedule, NoExecute effects, and toleration syntax guide.
- [Talos Linux MIG Configuration with GPU Operator](https://kubernetes.recipes/recipes/ai/talos-kubernetes-nvidia-mig-gpu-operator/): Configure NVIDIA MIG on Talos Linux Kubernetes clusters. Install GPU Operator, set MIG strategy, and dynamically partition A100 GPUs without node reboot.
- [TensorRT-LLM Kubernetes Deployment](https://kubernetes.recipes/recipes/ai/tensorrt-llm-kubernetes-guide/): Deploy TensorRT-LLM on K8s for optimized inference. Engine building, model conversion, and serving with Triton Inference Server.
- [TensorRT-LLM vs vLLM Benchmark 2026](https://kubernetes.recipes/recipes/ai/tensorrt-llm-vs-vllm-benchmark/): Compare TensorRT-LLM vs vLLM for LLM inference on Kubernetes. TTFT, throughput, GPU utilization benchmarks, and when to use each inference engine.
- [Test LLM Inference Endpoints with curl](https://kubernetes.recipes/recipes/ai/test-llm-inference-endpoints/): Validate Kubernetes-hosted LLM inference services using curl against OpenAI-compatible /v1/models, /v1/completions, and /v1/chat/completions endpoints.
- [Thanos Receive Memory Sizing Guide](https://kubernetes.recipes/recipes/observability/thanos-receive-memory-sizing/): Calculate correct memory limits for Thanos Receive based on WAL segments, active series, retention, and ingestion rate. Prevent OOMKill crash loops
- [Thanos Receive OOMKilled CrashLoopBackOff](https://kubernetes.recipes/recipes/troubleshooting/thanos-receive-oom-crashloop-statefulset/): Debug and fix Thanos Receive StatefulSet OOMKilled CrashLoopBackOff caused by WAL replay exceeding memory limits. Covers ArgoCD conflict resolution, liveness
- [Fix Thanos Receive OOMKilled in Run:ai](https://kubernetes.recipes/recipes/troubleshooting/thanos-receive-oomkilled-runai/): Troubleshoot and fix Thanos Receive OOMKilled (exit code 137) with 143+ restarts in Run:ai backend on OpenShift. Covers memory tuning, TSDB
- [Triton Autoscaling with GPU Metrics](https://kubernetes.recipes/recipes/ai/triton-autoscaling-gpu-metrics/): Autoscale Triton Inference Server on Kubernetes using GPU utilization, request queue depth, and inference latency metrics with KEDA and HPA.
- [Triton vs vLLM (and TensorRT-LLM) for LLM Serving](https://kubernetes.recipes/recipes/ai/triton-inference-server-vs-vllm-comparison/): Triton Inference Server vs vLLM for LLM serving on Kubernetes: architecture, TensorRT-LLM vs vLLM backends, APIs, multi-model support, and when to pick each.
- [Triton Multi-Model Serving on Kubernetes](https://kubernetes.recipes/recipes/ai/triton-multi-model-serving/): Serve multiple LLMs simultaneously on Triton Inference Server using TensorRT-LLM and vLLM backends with model routing and GPU scheduling.
- [Triton TensorRT-LLM on Kubernetes](https://kubernetes.recipes/recipes/ai/triton-tensorrt-llm-kubernetes/): Deploy NVIDIA Triton Inference Server with TensorRT-LLM backend on Kubernetes for optimized large language model serving with GPU acceleration.
- [Triton with vLLM Backend on Kubernetes](https://kubernetes.recipes/recipes/ai/triton-vllm-kubernetes/): Deploy NVIDIA Triton Inference Server with vLLM backend on Kubernetes for flexible LLM serving with PagedAttention and continuous batching.
- [Fix NVIDIA NIM TensorRT-LLM Initialization Errors](https://kubernetes.recipes/recipes/troubleshooting/troubleshoot-nim-tensorrt-llm/): Diagnose and fix common NIM TensorRT-LLM executor failures including DecoderState mismatch, version incompatibilities, and engine build errors.
- [Fix 'No Supported NIC Is Selected' in SR-IOV](https://kubernetes.recipes/recipes/troubleshooting/troubleshoot-no-supported-nic-selected/): Diagnose SR-IOV operator webhook rejections by validating node state, label selectors, PF eligibility, and SriovNetworkNodePolicy configuration.
- [Fix nv-ipam 'Pool Not Found' Errors in Multus](https://kubernetes.recipes/recipes/troubleshooting/troubleshoot-nv-ipam-pool-not-found/): Fix nv-ipam IPPool lookup failures in Multus by aligning SriovNetwork, NetworkAttachmentDefinition, and IPPool names and namespaces correctly.
- [Fix nvidia-fs Module Conflict on OpenShift](https://kubernetes.recipes/recipes/troubleshooting/troubleshoot-nvidia-fs-module-conflict/): Diagnose and fix the 'insmod: ERROR: could not insert module nvidia-fs.ko: File exists' error when enabling GPUDirect Storage with the NVIDIA GPU Operator.
- [Troubleshooting Pods with GPU Devices](https://kubernetes.recipes/recipes/troubleshooting/troubleshooting-pods-devices-gpu-kubernetes/): Fix GPU device issues in Kubernetes pods. Troubleshoot device plugin errors, DRA claims, CUDA failures, driver mismatches.
- [NCCL Env Vars: Force InfiniBand/RoCE and GDR](https://kubernetes.recipes/recipes/configuration/tune-nccl-env-rdma-ethernet/): Force NCCL onto InfiniBand or RoCE instead of TCP sockets: set NCCL_IB_HCA, NCCL_SOCKET_IFNAME, NCCL_IB_GID_INDEX and NCCL_NET_GDR_LEVEL on Kubernetes.
- [Ubuntu 26.04 LTS K8s Node Hardening](https://kubernetes.recipes/recipes/security/ubuntu-2604-kubernetes-sudo-rs/): Harden Kubernetes nodes with Ubuntu 26.04 LTS Resolute Raccoon. sudo-rs Rust rewrite, APT rollback, Kernel 7.0 TDX, ROCm GPU, and secure base images.
- [Update CA Certificates in Kubernetes](https://kubernetes.recipes/recipes/security/update-ca-certificates-kubernetes/): Rotate and update Certificate Authority (CA) certificates in Kubernetes clusters including kube-apiserver, etcd, kubelet, and custom CA bundles for TLS.
- [Validate GPU & NIC Topology Before NCCL](https://kubernetes.recipes/recipes/troubleshooting/validate-gpu-topology-nccl/): Inspect node-level GPU, NIC, and PCI topology on Kubernetes workers to predict and explain NCCL benchmark performance before running tests.
- [Validate GPUDirect RDMA Performance DMA-BUF](https://kubernetes.recipes/recipes/networking/validate-gpudirect-rdma-performance/): Run ib_write_bw with CUDA DMA-BUF to verify GPUDirect RDMA data transfer rates between GPU pods and validate network operator configuration.
- [Validate SR-IOV Operator Health Across Nodes](https://kubernetes.recipes/recipes/troubleshooting/validate-sriov-on-multiple-nodes/): Run a full checklist to confirm SR-IOV discovery, VF creation, scheduler resources, and pod attachment on multiple nodes.
- [Deploying Vector Databases on Kubernetes](https://kubernetes.recipes/recipes/ai/vector-database-kubernetes/): Deploy and operate vector databases (Milvus, Weaviate, Qdrant) on Kubernetes for RAG pipelines, semantic search, and AI applications with persistent.
- [Velero Kubernetes Backup, Restore and Schedules](https://kubernetes.recipes/recipes/storage/velero-kubernetes-backup-disaster-recovery/): Velero backup and disaster recovery for Kubernetes: install on S3/MinIO, velero backup create, schedules with TTL, PV snapshots vs Kopia, restore, migration.
- [Verify NCCL RDMA Traffic with Debug Logging](https://kubernetes.recipes/recipes/ai/verify-nccl-rdma-traffic-debug/): Prove NCCL uses RDMA for GPU communication on Kubernetes. Use NCCL_DEBUG and NCCL_DEBUG_SUBSYS=ALL to verify InfiniBand, RoCE.
- [Verify Which Interface Carries OVN Underlay](https://kubernetes.recipes/recipes/networking/verify-ovn-underlay-interface/): Confirm the actual OVN underlay network path by checking ovn-encap-ip, bridge port ownership, and physical route associations on Kubernetes nodes.
- [Virtual Kubelet for Serverless K8s Scaling](https://kubernetes.recipes/recipes/autoscaling/virtual-kubelet-serverless-scaling/): Deploy Virtual Kubelet to burst Kubernetes workloads to serverless backends like Azure ACI, AWS Fargate, and Hashicorp Nomad for infinite scaling.
- [VT-x vs VT-d vs SR-IOV Explained](https://kubernetes.recipes/recipes/networking/virtualization-vt-x-vs-vt-d-vs-sriov/): Understand the difference between CPU virtualization (VT-x/SVM), I/O virtualization (VT-d/AMD-Vi/IOMMU), and SR-IOV. Which to enable or disable for GPU
- [vLLM Alternatives LLM Inference K8s](https://kubernetes.recipes/recipes/ai/vllm-alternatives-llm-inference-k8s/): Compare vLLM alternatives for LLM inference on Kubernetes. TensorRT-LLM, SGLang, NVIDIA NIM, Ollama, and text-generation-inference feature comparison.
- [vLLM Inference Optimization on Ascend NPU: K8s](https://kubernetes.recipes/recipes/ai/vllm-ascend-npu-inference-kubernetes/): Optimize vLLM inference on Huawei Ascend NPUs in Kubernetes: tensor parallelism sizing, memory tuning, and OOM prevention for Atlas 300I/910B.
- [Debug Distributed vLLM Inference with NCCL Verbose Logging](https://kubernetes.recipes/recipes/ai/vllm-distributed-inference-debugging-nccl/): Debug distributed vLLM inference using NCCL_DEBUG=INFO and NCCL_DEBUG_SUBSYS=ALL. Covers air-gapped deployment with TRANSFORMERS_OFFLINE, interpreting NCCL
- [vLLM Deployment Kubernetes Guide](https://kubernetes.recipes/recipes/ai/vllm-kubernetes-deployment-guide/): Deploy vLLM inference engine on K8s. Model loading, tensor parallelism, continuous batching, and OpenAI-compatible API setup.
- [Deploy vLLM OpenAI Container on Kubernetes](https://kubernetes.recipes/recipes/ai/vllm-openai-container-kubernetes/): Deploy the vLLM OpenAI-compatible server container on Kubernetes. Pull ghcr.io/vllm-project/vllm-openai, configure GPU resources, model loading.
- [Install VPA with hack/vpa-up.sh Script](https://kubernetes.recipes/recipes/autoscaling/vpa-hack-vpa-up-sh-install-kubernetes/): Install Kubernetes Vertical Pod Autoscaler using hack/vpa-up.sh from the official repository. VPA components, prerequisites, and troubleshooting guide.
- [WebAssembly Runtime with Spin and SpinKube](https://kubernetes.recipes/recipes/deployments/wasm-runtime-kubernetes/): Deploy WebAssembly workloads on Kubernetes using SpinKube and the Spin Operator. Run Wasm components alongside containers with sub-millisecond cold starts.
- [Serverless Functions with WebAssembly](https://kubernetes.recipes/recipes/deployments/wasm-serverless-functions-kubernetes/): Build serverless functions using WebAssembly on Kubernetes with Fermyon Cloud, KEDA, and SpinKube. Achieve sub-millisecond scale-to-zero with Wasm cold starts.
- [WASI and containerd Wasm Shims on Kubernetes](https://kubernetes.recipes/recipes/deployments/wasm-wasi-containerd-kubernetes/): Run WebAssembly workloads using containerd Wasm shims with WASI support on Kubernetes. Configure runwasi, wasmtime, and WasmEdge as container runtimes.
- [Run WebAssembly Workloads on Kubernetes](https://kubernetes.recipes/recipes/deployments/wasm-workloads-kubernetes/): Deploy WASM workloads on Kubernetes using SpinKube and containerd-shim. Sub-millisecond cold starts, polyglot runtimes, and sandboxed edge computing.
- [Fix Admission Webhook Timeout Errors](https://kubernetes.recipes/recipes/troubleshooting/webhook-timeout-admission-errors/): Debug admission webhook failures blocking pod creation. Identify failing webhooks, check timeouts, and set failurePolicy.
- [Weights and Biases Experiment Tracking on Kubernetes](https://kubernetes.recipes/recipes/ai/weights-and-biases-experiment-tracking-kubernetes/): Deploy Weights & Biases (W&B) on Kubernetes for ML experiment tracking, model registry, and hyperparameter sweeps. Self-hosted W&B Server, agent-based
- [How to Use Workload Identity for Cloud Access](https://kubernetes.recipes/recipes/security/workload-identity-cloud-access/): Securely access cloud services from Kubernetes pods without static credentials. Configure Workload Identity for AWS, Azure, and GCP with IRSA, Workload.