Add per-collector metric documentation

[willianpaixao]

Summary

• Add per-collector documentation under docs/collectors/
• Document metrics, labels, configuration flags, and data sources for first 5 collectors
• Update README.md to reference new documentation

Fixes #1170

Motivation

Users have requested detailed explanations of node_exporter metrics beyond the HELP strings. This PR establishes a documentation structure and provides initial documentation for commonly used collectors.

Structure

docs/collectors/
├── README.md        # Index and contributing guidelines
├── _TEMPLATE.md     # Template for new collector docs
├── cpu.md
├── cpufreq.md
├── diskstats.md
├── meminfo.md
└── netstat.md

Test plan

• Verify markdown renders correctly on GitHub
• Verify all internal links work
• Cross-reference metrics against actual collector output

[Copilot]

Pull request overview

This PR establishes a documentation framework for node_exporter collectors and provides detailed metric documentation for five commonly-used collectors (cpu, cpufreq, diskstats, meminfo, netstat). The documentation includes metric names, types, labels, configuration flags, and data sources.

Changes:

• Created standardized documentation structure under docs/collectors/ with a template and index
• Added comprehensive documentation for cpu, cpufreq, diskstats, meminfo, and netstat collectors
• Updated main README.md to reference the new collector documentation

Reviewed changes

Copilot reviewed 8 out of 8 changed files in this pull request and generated 3 comments.

Show a summary per file

File	Description
docs/collectors/README.md	Index page listing available collector documentation and contributing guidelines
docs/collectors/_TEMPLATE.md	Standard template for documenting collectors to ensure consistency
docs/collectors/cpu.md	Documents CPU time statistics, metadata, and thermal throttling metrics
docs/collectors/cpufreq.md	Documents CPU frequency scaling metrics from sysfs
docs/collectors/diskstats.md	Documents disk I/O statistics, device info, and ATA attributes
docs/collectors/meminfo.md	Documents memory statistics from /proc/meminfo across multiple platforms
docs/collectors/netstat.md	Documents network statistics from /proc/net files
README.md	Adds reference to new collector documentation with links to documented collectors

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[discordianfish]

I don't think anyone here has the resources to maintain this.. @SuperQ wdyt?

[SuperQ]

If we had a way to generate this, similar to how Go doc comments work, it would be feasible to maintain. We would have all the docs in the collector code that generates markdown.

[willianpaixao]

Totally understand your position, @discordianfish. Let me make my case.

1. We do not need to document all collectors, specially the less used, niche one. I can personally document about ten of the most used/important and I think it's enough.
2. Of those most used/important collectors, their code base is stable, rarely changed. Once the documentation is written, not much would need update, maybe not ever. If a new metric is added, old removed, that's a one liner in the Markdown. The bulk of the work is not in the maintenance of ~10 doc files.

My argument is on the line of "a bird in the hand is worth two in the bush".