Kubernetes Monitoring

Your Kubernetes cluster.
Fully visible. Always.

HostAtlas automatically detects Kubernetes clusters on any server where kubectl is available. No SDK, no Helm chart, no extra configuration — just install the agent and get instant visibility into nodes, pods, deployments, and services.

How It Works

Zero extra setup required.

The HostAtlas agent runs on your server as a standard Linux service. At each discovery interval, it checks whether kubectl is available and reachable. If it is, the agent calls kubectl get nodes/pods/deployments/services and ships the structured data to the HostAtlas platform.

1

Install the agent

One curl command installs and starts the HostAtlas agent on any Linux server. No K8s-specific configuration needed.

2

kubectl auto-detected

The agent checks for kubectl availability. If found and cluster-info responds, K8s monitoring activates automatically.

3

Dashboard appears

Within minutes, the Kubernetes dashboard populates with your cluster data. Updates on every discovery cycle.

What's Monitored

Complete cluster visibility.

Every layer of your cluster is tracked — from node hardware to individual pod restarts.

dns

Node Health

Status (Ready/NotReady), roles (master/worker), CPU cores, memory, Kubernetes version, OS image, and architecture. Optional CPU and RAM utilization percentages via kubectl top nodes.

view_in_ar

Pod Status

All pods across all namespaces. Status (Running, Pending, Failed, CrashLoopBackOff, ImagePullBackOff), ready state, restart count, node assignment, and pod IP.

layers

Deployments

Ready vs desired replica counts across all namespaces. Instantly see which deployments are degraded or rolling out. Age and namespace for context.

swap_horiz

Services

Service type (ClusterIP, NodePort, LoadBalancer), namespace, cluster IP, external IP, and port mappings. Know which services are externally exposed at a glance.

category

Namespaces

All cluster namespaces listed in the summary KPI. Filter pods and deployments by namespace in the detail view.

emergency_home

Problem Pod Surfacing

Pods that are not Running or Succeeded are automatically flagged in a "Problem Pods" panel on the dashboard. Restart counts and failure statuses highlighted in red — no scrolling through hundreds of running pods.

Dashboard Overview

One view. Everything you need.

The Kubernetes index page shows all clusters across your infrastructure. Each cluster displays KPI cards, node status, and problem pods — without having to navigate per-namespace views.

Nodes

3/3

Pods

47/49

Deployments

12/12

Services

18

Namespaces

5

Start monitoring your K8s cluster in minutes.

Install the agent on any node where kubectl is configured. No additional setup required.

Start free — no credit card