Kubernetes 2.0

Introduction

The OpsRamp Kubernetes 2.0 Agent provides advanced visibility, monitoring, and observability for Kubernetes environments. It extends the capabilities of Kubernetes 1.0 with broader resource discovery, richer telemetry data collection, and modern observability integrations to help you manage complex containerized workloads more effectively.

With Kubernetes 2.0, you can:

Discover and monitor a wide range of Kubernetes resources.
Access real-time insights through customizable dashboards.
Troubleshoot faster with topology-based visualization.
Collect pod logs, events, traces, and performance metrics for complete observability.
Define Prometheus-based alert conditions and integrate seamlessly into OpsRamp’s alerting workflows.

Note

Migration: Kubernetes 1.0 and Kubernetes 2.0 cannot co-exist. You must uninstall Kubernetes 1.0 before installing Kubernetes 2.0.
Compatibility: Kubernetes 2.0 provides all the functionality of Kubernetes 1.0, plus extended monitoring and observability.

Key Benefits

Comprehensive Monitoring: Gain visibility into Nodes, Pods, Namespaces, Services, Volumes, and workloads.
Faster Troubleshooting: Use the Topology Explorer to visualize dependencies and quickly isolate network or infrastructure issues.
Flexible Alerting: Define alert conditions using PromQL for precise monitoring across pods, deployments, and namespaces.
Complete Observability: Collect metrics, logs, traces, and Kubernetes events, powered by the OpenTelemetry framework and eBPF for low-overhead performance monitoring.
Future-Ready: Built on open standards and cloud-native observability practices, ensuring compatibility and scalability.

Features of Kubernetes 2.0 Agent

Resource & Metrics Visibility

Continuously monitors Kubernetes resources such as nodes, pods, services, and volumes to track CPU, memory, storage, and network usage. This ensures complete visibility into cluster performance and helps identify issues early.

Workload Monitoring

Tracks the health and performance of application workloads like Deployments, StatefulSets, and DaemonSets. This ensures applications scale properly, remain stable, and deliver the expected performance.

Pod Log Collection

Captures and analyzes logs from pods and containers to understand application behavior and diagnose errors. This makes troubleshooting failures and performance issues easier and faster.

Kubernetes Events

Monitors real-time Kubernetes events such as pod scheduling, scaling, or failures. This provides instant awareness of cluster changes and anomalies, enabling faster responses to operational issues.

Tracing

Collects distributed traces across workloads using OpenTelemetry. This delivers end-to-end visibility into how requests move through microservices, making it easier to detect bottlenecks and optimize performance.

eBPF-based Monitoring

Leverages eBPF (extended Berkeley Packet Filter) to capture low-level system metrics from the Linux kernel with minimal overhead. This provides deep insights into networking, processes, and I/O activities without impacting performance.

Topology Explorer

Offers a visual representation of Kubernetes resources and their relationships. This helps you quickly identify dependencies, analyze the impact of issues, and speed up root-cause troubleshooting.

Custom Alerts with PromQL

Enables you to create precise alerts using Prometheus queries, so you focus only on meaningful issues instead of being overwhelmed by alert noise.

K8s 1.0 vs K8s 2.0: What’s Changed

Kubernetes	Kubernetes 2.0
Only the following are discovered and monitored: Container Node	We are discovering and monitoring (collecting metrics) for more Kubernetes objects such: Nodes Pods Namespaces Services Persistent Volumes (PVs) Persistent Volume Claims (PVCs) Daemonsets Deployments
Metrics are collected via our agent only	We are using open-source OpenTelemetry framework which is an observability framework and toolkit designed to create and manage telemetry data such as metrics.
Kubernetes was having limited topology relationship view	Kubernetes 2.0 comes with enhanced topology explorer which provides a visual representation of resources, their relationships, and availability, facilitating effective troubleshooting issues.
Kubernetes event alert was generated with metric name as KubeEvents and it was associated to cluster resource.	Kubernetes event alert is generating using metric name as kubernetes_events and it is associated to resource involved in event if it is created by the time, else to cluster resource.

Resource Hierarchy

      • Kubernetes 2.0
              • Cluster
                      • Node
                      • Pod
                      • Namespaces
                      • Services
                      • Deployment, DaemonSet, ReplicaSet, StatefulSet
                      • PersistentVolume (PV)
                      • PersistentVolume Claim (PVC)