Skills
.Work. You
Rest

sue-nm5-inode-status

Community

Instantly check NM5 storage and inode pressure status.

Education & Research#hpc#read-only diagnostics#scale-up experiments#nm5#inode monitoring#storage status#gpu clusters
Authordongzhuoyao
Version1.0.0
Installs0

System Documentation

What problem does it solve?

Running scale-up experiments on NM5 HPC clusters requires constant monitoring of storage and inode usage to avoid job failures from full disks or exhausted file counts, but manually running and parsing multiple df and du commands across different account namespaces is time-consuming and error-prone.

Core Features & Use Cases

  • Read-only status snapshots: Generates standardized OK/WARNING/BLOCKER verdicts for NM5 project/scratch storage and inode usage across compute and storage account namespaces.
  • Top consumer identification: Pinpoints the largest inode consumers in DeepResearch workspaces and mount-level storage, with private paths and credentials automatically redacted.
  • Use case: When preparing to launch a large-scale training run on NM5, use this skill to quickly confirm you have sufficient inode headroom, or diagnose unexpected job failures related to storage pressure.

Quick Start

Invoke the sue-nm5-inode-status skill to generate a full read-only report of NM5 storage and inode usage for your DeepResearch workspace.

Dependency Matrix

Required Modules

None required

Components

Standard package

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: sue-nm5-inode-status
Download link: https://github.com/dongzhuoyao/deepresearch/archive/main.zip#sue-nm5-inode-status

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 536,000+ vetted skills library on demand.