Node Resource Monitoring Page

[rugved099]

Node Resource Monitoring Page

Overview

The dashboard currently provides visibility into Jobs, Queues, Pods, and PodGroups, but there is no dedicated node-level resource monitoring view.

Kubernetes operators need a way to monitor node health and resource availability, including:

• CPU capacity and allocatable resources
• Memory capacity and allocatable resources
• Node conditions and health status
• Resource utilization across nodes

The Kubernetes client (k8s.ts) already has CoreV1Api configured, so retrieving node information should require only a listNode() API call.

Motivation

A node-level monitoring page would provide operators with a more complete view of cluster health and resource availability.

This would also complement the existing workload-level views and make it easier to identify resource bottlenecks or unhealthy nodes.

Proposed Scope

1. Navigation

Add a new Nodes navigation item in:

constants/index.ts

2. Backend

Create a new tRPC router under:

packages/trpc/server/router/nodes/

The router should expose the necessary node information from the Kubernetes API, including:

• Node name
• CPU capacity
• CPU allocatable
• Memory capacity
• Memory allocatable
• Node conditions
• Other useful node metadata where appropriate

The existing CoreV1Api configuration in k8s.ts should be reused.

3. Frontend

Create a new dashboard page at:

apps/web/src/app/[locale]/(dashboard)/nodes/

The page should include:

• Node list/table
• CPU resource information
• Memory resource information
• Resource utilization visualizations/bar charts
• Node health/condition status

The UI should follow the existing dashboard's design patterns and components wherever possible.

4. Internationalization

Add the required translation messages for:

• en
• zh-CN

All newly introduced user-facing text should use the existing i18n system.

Acceptance Criteria

• A Nodes item is available in the dashboard navigation.
• The Nodes page is accessible through the dashboard.
• Node data is fetched from the Kubernetes API through a dedicated tRPC router.
• The page displays node CPU and memory capacity/allocatable resources.
• Node conditions/health status are visible.
• Resource utilization is represented using clear visualizations.
• The implementation works with the existing CoreV1Api setup.
• English (en) translations are added.
• Chinese (zh-CN) translations are added.
• Existing dashboard functionality is not affected.
• The implementation follows the project's existing architecture and UI conventions.