Server diagnostics: how would you act if you connected via SSH to an unfamiliar server that is slow?

When connecting via SSH to an unfamiliar server that is slow, I would act as follows:

1. Load check:

   • top or htop — check CPU load, memory, processes.
   • uptime — find out the average load over 1, 5, 15 minutes.

2. Memory and swap:

   • free -m — check the usage of RAM and swap.

3. Disk activity:

   • iostat or iotop — check if the disk is overloaded.
   • df -h — check free space on disks.

4. Network resources:

   • netstat -tulnp or ss -tulnp — view open ports and active connections.
   • iftop — monitor network traffic.

5. Logs:

   • Review system logs (/var/log/syslog, /var/log/messages) for errors.

6. Processes:

   • Identify processes with high load (ps aux --sort=-%cpu).

7. Check for locks:

   • dmesg — view system messages related to hardware.

8. Action plan:

   • If the problem is with a specific process — restart or optimize it.
   • If resources are exhausted — possibly increase resources or optimize configuration.

This systematic approach helps quickly localize the cause of slowdowns and take measures.