skills.rest
black-yt
black-ytCOMMUNITY
Explore all skills by this author →
Website𝕏@wanghan_xuGitHub

lab-cluster-1-web-portal

Log into the lab-cluster-1 portal and query Prometheus and DCGM metrics.

7|1|Updated May 9, 2026
Search Tags:#prometheus#oidc#gpu monitoring#dcgm#rjob#lab-cluster-1#h.pjlab.org.cn
One-click install
npx skills add https://github.com/black-yt/skills --skill lab-cluster-1-web-portal
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: lab-cluster-1-web-portal
Source: https://github.com/black-yt/skills/tree/main/lab-cluster-1-web-portal
Command: npx skills add https://github.com/black-yt/skills --skill lab-cluster-1-web-portal

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill helps you safely access the lab-cluster-1 web portal, log in with OIDC, and query monitoring data without confusing it with task-level job information.

Core Features & Use Cases

  • Secure Web Login: Handles browser-style OIDC authorization code flow, token refresh, cookie/session handling, and credential safety for h.pjlab.org.cn.
  • Read-Only Monitoring Queries: Retrieves Prometheus and DCGM metrics such as GPU utilization, memory used/free, power, and temperature through the portal API.
  • Job-to-GPU Attribution: Maps monitoring labels like namespace and pod back to rjob replicas so reports can separate real job usage from other cluster occupancy.
  • Use Case: Use this Skill when you need to inspect which jobs are occupying GPUs, generate a usage report, or troubleshoot login and query failures on the lab-cluster-1 portal.

Quick Start

Ask the assistant to log into the lab-cluster-1 web portal, query GPU utilization for the target namespace, and summarize the results with rjob attribution while keeping credentials and tokens private.

Frequently Asked Questions about lab-cluster-1-web-portal

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I query Prometheus and DCGM metrics from the lab-cluster-1 web portal?▼

To query Prometheus and DCGM metrics, you log into the lab-cluster-1 web portal using OIDC authentication and retrieve read-only monitoring data like GPU utilization through the portal API.

How does OIDC login work for h.pjlab.org.cn cluster monitoring access?▼

OIDC login for h.pjlab.org.cn uses a browser-style authorization code flow that handles token refresh and cookie sessions while ensuring credential safety for cluster monitoring access.

Can I map Prometheus monitoring labels back to rjob replicas?▼

Yes, you can map Prometheus monitoring labels like namespace and pod back to rjob replicas, allowing you to separate real job usage from other cluster occupancy in reports.

What GPU metrics can I inspect through the DCGM and Prometheus portal APIs?▼

Through the DCGM and Prometheus portal APIs, you can inspect GPU metrics including utilization, memory used and free, power consumption, and temperature for the lab-cluster-1 environment.

How do I attribute GPU utilization to specific jobs on lab-cluster-1?▼

You attribute GPU utilization by querying read-only monitoring APIs and mapping labels like namespace and pod back to rjob replicas to generate reports separating task data from monitoring data.

What are the limitations of querying read-only monitoring APIs for lab-cluster-1 metrics?▼

The read-only monitoring APIs limit you to querying metrics without modifying cluster state, requiring secure token management and proper label mapping to ensure accurate rjob-to-pod attribution workflows.

Related Skills

Explore More Data & Analytics Skills→
cegit27cegit27
3

setup-browser-cookies

Import authenticated Chromium session cookies into a headless browser session.

Official
Intermediate
coreweavecoreweave
7

setup-sunk-gpu-monitoring

Set up GPU and Slurm metrics collection on a SUNK cluster.

Official
Advanced
Ikram-AlamIkram-Alam

deploying-postgres-k8s

Automate PostgreSQL cluster deployment on Kubernetes with CloudNativePG.

Community
Advanced
yaacovyaacov

metrics-tool-guide

Guide Prometheus/Thanos cluster monitoring with kubectl-metrics queries and PromQL.

Community
Basic