A Linux server is easier to understand when you read it in order.

Newer engineers often start by running whichever command they remember first. That can work, but it also creates noise. A calmer pattern is to ask the host a few basic questions before making changes.

Start with identity

hostnamectl
ip addr
ip route

You are answering:

• what does this machine think it is?
• which networks can it see?
• where does default traffic go?

Then check storage

lsblk
df -h
mount

You are looking for the shape of the system. Is it booting from the disk you expected? Is anything full? Are important mounts present?

Then check services

systemctl --failed
systemctl status ssh
journalctl -p warning -n 80

Do not read every log line. Look for the first useful clue, then follow it.

The important habit

Before changing anything, say what you think is true:

This host is on the right network, booted from the expected disk, has no full filesystems, and SSH is healthy.

That sentence is not ceremony. It is how you keep troubleshooting from becoming guessing with a keyboard.