DevOps Daily with Fexingo: CI/CD, Kubernetes, and Modern Software Operations

Lucas and Luna dig into a common but overlooked Kubernetes pain point: how ResourceQuotas designed to prevent resource abuse can accidentally deadlock your batch processing pipeline. They walk through a real incident at a mid-size fintech that ran 24,000 short-lived Pods per day, where namespace-level quotas caused cascading failures during a routine data-backfill job. Lucas explains the mechanics—how admission controllers evaluate quotas at Pod creation time, why burst concurrency matters more than total resource limits, and the counterintuitive fix: lowering quota ceilings to raise throughput. Luna pushes back on the conventional wisdom that more quota always means more capacity. The episode closes with a practical monitoring heuristic using the 'kubectl describe quota' command to spot the problem before it hits production.

#Kubernetes #ResourceQuotas #BatchJobs #DevOps #CloudNative #Fintech #PodLifecycle #AdmissionController #Concurrency #Backfill #Kubectl #ProductionIncident #CapacityPlanning #ClusterAutoscaling #Scheduling #FexingoBusiness #BusinessPodcast #Technology

Keep every episode free: buymeacoffee.com/fexingo

Lucas and Luna break down how container image layers silently consume node disk space in Kubernetes clusters. Using real numbers from a 500-node production cluster, Lucas explains the hidden cost of unused base images, intermediate layers, and stale tags that build up over time. Luna shares a horror story of a team that had to drain an entire node because garbage collection ran too late. The episode covers practical strategies to audit image usage, configure image garbage collection thresholds in kubelet, and set up a retention policy. No abstract theory: listeners get a concrete checklist they can apply to their clusters by end of day. This episode is for DevOps engineers who want to keep their nodes healthy without buying more storage.

#Kubernetes #ContainerImages #DevOps #NodeDisk #ImageGC #Kubelet #TechOperations #CloudNative #Docker #ContainerLayers #DiskPressure #ClusterHealth #ImageAudit #Storage #SRE #FexingoBusiness #BusinessPodcast #Technology

Keep every episode free: buymeacoffee.com/fexingo

Lucas and Luna unpack a specific Kubernetes failure pattern: cluster DNS caching. When one node experiences a DNS resolution error, cached negative responses can propagate across the cluster, causing cascading service discovery failures. They walk through how the NodeLocal DNSCache plugin can actually make this worse under certain conditions, and dig into a real incident at a mid-size fintech where a single misconfigured CoreDNS pod took down three microservices for 12 minutes. They also explain why the default 'ndots:5' setting causes unnecessary upstream lookups, and share concrete tuning strategies for reducing DNS latency in high-traffic deployments. If you run Kubernetes in production and your services talk to each other, this episode is a troubleshooting cheat sheet.

#Kubernetes #DNS #CoreDNS #NodeLocalDNSCache #ServiceDiscovery #ClusterNetworking #SiteReliabilityEngineering #Microservices #ProductionIncidents #KubernetesTroubleshooting #CloudNative #DevOps #Infrastructure #Technology #FexingoBusiness #BusinessPodcast #Podcast #SoftwareOperations

Keep every episode free: buymeacoffee.com/fexingo

In this episode of DevOps Daily, Lucas and Luna dive into the specific mechanics and benefits of Kubernetes Node Local DNS cache (NodeLocal DNSCache). They explain how it reduces DNS latency by up to 30 percent, prevents conntrack table exhaustion, and alleviates pressure on CoreDNS in large clusters. The discussion covers real-world deployment configurations, including the use of iptables rules to intercept DNS queries, the trade-offs with cluster DNS autoscaling, and a concrete migration strategy for production clusters running 500+ nodes. Tune in for actionable insights on optimizing DNS resolution in Kubernetes.

#Kubernetes #DNS #NodeLocalDNSCache #CoreDNS #Latency #Optimization #DevOps #CloudNative #CNCF #K8s #Performance #Scalability #Network #Technology #FexingoBusiness #BusinessPodcast #Podcast #SoftwareOperations

Keep every episode free: buymeacoffee.com/fexingo

Episode 55 of DevOps Daily dives into Kubernetes Quality of Service (QOS) classes — Guaranteed, Burstable, and BestEffort — and how they determine which pods get killed when memory runs out. Lucas explains why a Burstable pod with a 1GB limit but a 500MB request is more likely to be OOM-killed than a BestEffort pod using 100MB, using a real-world case from a fintech startup that lost payment-processing pods during a memory spike. Luna shares a counterintuitive example from her own Kubernetes cluster where setting requests equal to limits actually reduced total utilization. The hosts discuss the trade-offs between resource efficiency and reliability, and why many teams blindly set requests too low. The episode also covers practical strategies: using Vertical Pod Autoscaler to tune requests, setting memory limits with a 20% headroom buffer, and monitoring the oom_kill counter in Prometheus. A must-listen for any engineer running stateful workloads on Kubernetes.

#Kubernetes #QOS #OOMKill #DevOps #ContainerOrchestration #PodScheduling #MemoryManagement #CloudNative #Fintech #SiteReliabilityEngineering #Prometheus #VerticalPodAutoscaler #KubernetesBestPractices #ProductionIncidents #ResourceManagement #Technology #FexingoBusiness #BusinessPodcast

Keep every episode free: buymeacoffee.com/fexingo

Lucas and Luna dig into a specific Kubernetes failure pattern that most teams discover the hard way: network policies that block healthy traffic after a rollout. They walk through a real incident at a mid-size fintech company where a seemingly innocuous deny-all ingress policy took down payment processing for 47 minutes. Lucas explains how Kubernetes NetworkPolicy objects are evaluated, why default-deny patterns can backfire, and why teams should use audit-mode logging before enforcing. Luna shares a pragmatic checklist for testing network policies in staging. They close with the one label selector rule every DevOps engineer should memorize before their next production deploy. A solid episode for anyone running Kubernetes beyond a hobby cluster.

#Kubernetes #NetworkPolicy #DevOps #ProductionIncident #Fintech #PodNetworking #CNI #Calico #Cilium #DefaultDeny #LabelSelector #SilentOutage #InfrastructureAsCode #CloudNative #Technology #FexingoBusiness #BusinessPodcast #TechOps

Keep every episode free: buymeacoffee.com/fexingo

In this episode of DevOps Daily, Lucas and Luna tackle a subtle but costly problem in Kubernetes configuration management: Kustomize overlays that silently diverge from their base configurations. Using a real-world example from a mid-size e-commerce company, they break down how three overlays intended to customize deployments for dev, staging, and production environments ended up creating a cascade of misconfigurations. Lucas explains the root cause—a common anti-pattern where overlay patches override base fields without version control warnings—and walks through how a simple diff-and-lint pipeline could have caught the drift before it caused a production outage. Luna challenges the assumption that Kustomize alone is enough for configuration hygiene, and they discuss when teams should consider tools like Helm or Jsonnet for tighter validation. If you are using Kustomize in your Kubernetes workflows, this episode will save you from a painful debugging session.

#Kustomize #ConfigurationDrift #Kubernetes #DevOps #CI/CD #InfrastructureAsCode #GitOps #ConfigurationManagement #KubernetesDeployment #OverlayPatches #YAML #ProductionOutage #TechPodcast #SoftwareEngineering #FexingoBusiness #BusinessPodcast #CloudNative #DevOpsDaily

Keep every episode free: buymeacoffee.com/fexingo

In episode 52 of DevOps Daily with Fexingo, Lucas and Luna dig into a hidden performance gotcha in Kubernetes service discovery: the EndpointSlice controller's default sync interval. Lucas walks through a real incident at a mid-size e-commerce company where a deployment rollout caused five minutes of traffic blackholing because EndpointSlices lagged behind pod readiness. They compare EndpointSlice to the older Endpoints API, explain the kube-controller-manager flags that control sync timing, and discuss the trade-offs between faster propagation and API server load. Luna challenges Lucas on whether most teams should even tweak these defaults, and they land on a practical recommendation: monitor your service propagation latency before you tune anything. If you manage Kubernetes at scale and have ever seen 'no endpoints available' for longer than you expected, this episode explains why.

#Kubernetes #EndpointSlice #ServiceDiscovery #kubeControllerManager #PodLifecycle #TrafficBlackholing #CloudNative #DevOps #SiteReliabilityEngineering #KubernetesNetworking #EndpointsAPI #ContainerOrchestration #KubernetesPerformance #RollingUpdate #FexingoTechnology #TechnologyPodcast #FexingoBusiness #BusinessPodcast

Keep every episode free: buymeacoffee.com/fexingo

In this episode of DevOps Daily, Lucas and Luna explore a hidden cost of Kubernetes service meshes: the memory overhead of sidecar proxies. They break down how Istio's Envoy sidecars commonly consume 50-100 MB per pod, and how a 50-node cluster with 200 pods can waste over 10 GB of RAM just on proxy overhead. They compare sidecar-less alternatives like Cilium's eBPF-based mesh and Istio's ambient mode, and share real benchmarks from large-scale deployments that saved 30-40% on compute costs by switching. The hosts also discuss practical strategies—like tuning proxy resource limits and using selective sidecar injection—to avoid blowing your cluster's memory budget. A must-listen for any Kubernetes operator optimizing cloud costs.

#Kubernetes #ServiceMesh #Istio #Envoy #Sidecar #MemoryOverhead #Cilium #eBPF #AmbientMesh #CloudCosts #ResourceOptimization #DevOps #Technology #FexingoBusiness #BusinessPodcast #Podcast #CloudNative #SRE

Keep every episode free: buymeacoffee.com/fexingo

Episode 50 of DevOps Daily with Fexingo takes a hard look at Kubernetes PriorityClasses — the feature meant to ensure critical pods run first but often backfires into scheduling chaos. Lucas and Luna unpack a real production outage at a mid-sized fintech company where misconfigured priority classes caused a cascade of evictions and resource starvation. They walk through the mechanics of preemption, the hidden cost of default priorities, and why many teams set up their priority tiers wrong. The conversation also touches on the tension between reliability engineers and cost optimizers, and offers a practical heuristic for assigning priority classes that won't surprise you during a traffic spike. If you've ever wondered why your cluster autoscaler is scaling up like crazy while pods are still pending, this episode explains the silent culprit.

#Kubernetes #DevOps #PriorityClasses #Scheduling #CloudNative #Preemption #ClusterAutoscaler #PodLifecycle #Fintech #ProductionOutage #ResourceManagement #SRE #PlatformEngineering #K8sBestPractices #TechPodcast #DevOpsDaily #FexingoBusiness #BusinessPodcast

Keep every episode free: buymeacoffee.com/fexingo