Kubernetes monitoring

WhaTap Kubernetes monitoring is a unified solution that analyzes containers, pods, nodes, entire clusters, and the applications running inside containers from a single screen. It supports both multi-cloud and hybrid cloud environments, and lets infrastructure and application teams collaborate on the same data.

Kubernetes monitoring structure

The smallest deployment unit in Kubernetes is a Pod, not a container, and a Pod can contain one or more containers. Kubernetes monitoring therefore requires both a container view and a Pod view. WhaTap collects metrics for containers and Pods into separate categories and provides both views on the Container Map.

A category is a name that identifies the metrics collected from your project. Each category has its own Tags and Fields.

Key features

• Unified monitoring — Monitor Kubernetes systems across multi-cloud and hybrid cloud environments from a single screen.

• MSA call relationship analysis — Trace API call relationships of transactions in Pod-level distributed environments.

• Root cause analysis — Monitor hosts, containers, and applications together to quickly identify the root cause of incidents.

• In-container application analysis — Provides hitmaps, traces, and multi-transaction tracking for Java, Node.js, and Python applications.

• Unified visibility through Container Map — Provides Kubernetes resources, metrics, events, and logs on a single screen. Infrastructure and application teams can collaborate on the same view.

• Resource usage trend analysis — Provides usage trend dashboards for containers and nodes. You can drill down from the full container/node list within a WhaTap project to the details of each target.

• Instant meta information lookup — View meta information directly on the Container Map without accessing the master node. A separate master meta dashboard shows the meta information of all components at a glance.

• Focused log analysis — Stores container log files and standard output in real time, and automates classification and indexing. You can focus on log analysis without accessing each container directly.

Try WhaTap Kubernetes monitoring through the tutorial video below.

Monitoring targets

WhaTap Kubernetes monitoring targets Kubernetes clusters and the applications running on top of them.

• Kubernetes cluster

  • CPU, memory, disk (volume), and network for containers
  • Cluster components such as namespaces, deployments, ReplicaSets, and pods
  • CPU, memory, disk, and network for nodes (hosts)

• In-container applications — Java, Node.js, Python, PHP, Go

The following monitoring procedure is divided into cluster monitoring and application monitoring.

Monitoring procedure

Cluster monitoring procedure

1. Create a cluster project in WhaTap Monitoring Service. The first Kubernetes project you create automatically becomes the cluster project. For details, see Getting started.

2. Install the master agent (whatap-master-agent) and node agent (whatap-node-agent) in your Kubernetes environment. Both agents run as pods inside your cluster. For details, see Agent installation.

3. (Optional) If your cluster has many namespaces, you can split them into separate namespace projects. For details, see Creating a namespace project.

Tip

A Kubernetes namespace is a mechanism for isolating resource groups within a single cluster. You can list the namespaces in your cluster with the kubectl get namespace command.

Application monitoring procedure

To monitor applications running inside containers, install application agents separately. The supported languages are Java, Node.js, Python, PHP, and Go. For details, see Application agent installation.