Linux System Files: The Configs, the Logs, and the Live Stats

The handful of files every admin learns to read — configs, logs, and the live kernel state.

How to Read This Index

Linux's "everything is a file" philosophy means a huge amount of the system's configuration and live state lives in plain text you can open with cat or less. No special tool, no API — just a path. Once you know the map, half of "how do I check X" answers itself.

We've grouped the files by where they live, because that's how Linux itself organizes them: /etc for configuration, /var/log for logs, /proc for the kernel's live view of the running machine, /dev for device nodes. Start with the section that matches what you're trying to learn; the full alphabetical list at the bottom is for jumping straight to a known path.

1. The Live Machine State — /proc

The files under /proc are not on disk. They're a window the kernel opens into its own state, generated on the fly each time you read them. cat /proc/loadavg is the same number uptime prints — uptime just dresses it up. Once you realize this, half the monitoring tools you've ever used stop looking magical.

2. System Configuration — /etc

The "where everything is configured" directory. Every daemon, every system service, every login policy keeps its config here as plain text. If you've ever wondered "how does the system know to do that?" — the answer is almost always a file under /etc.

3. Logs — /var/log

Where text-file logs live. Even on a systemd box where most logs flow through journalctl, plenty of daemons still write here — auth, the kernel, login records, and anything that predates the journal.

4. Device Nodes — /dev

The files under /dev aren't files at all — they're handles to devices, exposed as paths so any program that can open() a file can talk to hardware. The most useful ones are the synthetic ones: a bottomless source of zeros, a stream of randomness, a black hole.

  • /dev/null — the bit bucket; anything written here vanishes
  • /dev/random — high-entropy random bytes, blocks when the pool is low
  • /dev/tty0 — the current virtual console
  • /dev/urandom — random bytes that never block; what you should almost always use
  • /dev/video0 — the first video-capture device
  • /dev/zero — endless stream of zero bytes; useful for wiping and benchmarks

5. Other Files Worth Knowing

  • robots.txt — the web crawler convention; not a Linux system file but every server admin meets it

Full Alphabetical List

Every file in the knowledge base, sorted by path. If you know exactly what you're looking for, jump straight to it.

The fastest way to learn a Linux server is to actually open these files.

CleverUptime reads the same /proc and /sys files in the background every minute — load, memory, swap, disk usage, temperatures — and tells you in plain language what's changing and why it matters.

Want to see your own server's health right now? One command, no signup, no install.

Check your server →