10 Essential Linux Sysadmin Hacks to Boost Efficiency

Tip 1: Unmount an Unresponsive DVD Drive

When a DVD drive cannot be ejected because a process is holding it, use # fuser /media/cdrom to identify the offending process and then # fuser -k /media/cdrom to terminate it, after which # eject will succeed.

Tip 2: Recover a Frozen Screen

If the terminal becomes garbled, simply run # reset; the screen will return to normal without rebooting the machine.

Tip 3: Screen Collaboration

Use GNU screen to share a session: on one host run # screen -S foo, then have the colleague attach with # screen -x foo. Both users can see each other's input, facilitating joint debugging. Detach with Ctrl‑A D and reattach later.

Tip 4: Recover a Lost Root Password

Reboot, interrupt the GRUB menu, edit the kernel line, append 1 to the parameters, and boot into single‑user mode. Then change the password with # passwd and reboot.

Tip 5: Create an SSH Backdoor

Establish a reverse SSH tunnel from an internal machine ( ginger) to an external host ( blackbox.example.com) using # ssh -R 2222:localhost:22 [email protected]. Remote support can then connect to the internal host via the tunnel.

Tip 6: Remote VNC Over SSH

Start a VNC server on the internal host ( # vncserver -geometry 1024x768 -depth 24 :99), forward the VNC port through the same reverse tunnel, and connect from the remote machine with # vncviewer localhost:99. Windows users can achieve the same using PuTTY port forwarding.

Tip 7: Bandwidth Testing

Install iperf on both ends, run # iperf -s -f M on the server and # iperf -c ginger -P 4 -f M -w 256k -t 60 on the client to measure throughput, noting typical 1 GbE results around 112 Mbit/s and higher when bonding NICs.

Tip 8: Command‑Line Scripting and Utilities

Generate bulk /etc/hosts entries with a loop, or query memory across many nodes using SSH and free, grep, and awk. These one‑liners reduce manual errors and speed up cluster administration.

Tip 9: Console Inspection

Read virtual console devices (e.g., # cat /dev/vcs1) to see what users typed on a remote system, useful when logs omit console output.

Tip 10: Random System Information Gathering

Collect CPU details with # cat /proc/cpuinfo, disk usage with # df -h, BIOS data with # dmidecode, and NIC firmware via # ethtool -i eth0. These commands provide quick diagnostics without rebooting.

Conclusion

Effective sysadmins continuously learn from peers, share screen sessions, read manual pages, and solve problems hands‑on; this habit yields faster, more reliable operations and frees up personal time.