Tag: sysadmin

Compare zram (compressed RAM) and traditional swap files. Learn when each works, setup, and real-world performance trade-offs for memory-constrained systems.

Use strace to trace system calls and debug mysterious permission errors, missing files, and network issues without ever reading source code.

lsof shows every open file, socket, and port on the box. Find what holds a port, trace deleted files still eating disk, and see what a service owns.

Decode Docker exit codes: 0 (success), 1 (app error), 125/126/127 (Docker errors), 137 (OOM), 143 (SIGTERM). Debug restarts in minutes.

Manage Linux process priority with nice (CPU) and ionice (I/O). Learn priority levels, renice running processes, and avoid tanking server performance.

The 1024 file descriptor limit stops servers under load. What fds are, soft vs hard limits, how to raise them properly, and how to diagnose exhaustion.

Docker's default JSON logging driver writes unbounded logs to disk. Learn how to set log rotation and reclaim your storage.

Schedule one-off Linux tasks with `at` instead of cron. Learn syntax, queue management, batch jobs, and when to use it for delayed reboots and reminders.

PID 1 in a container ignores signals, so Ctrl+C does nothing and stops take 10 seconds. Fix it with exec form CMD or tini, ranked best to worst approach.

btmp logs failed logins and grows forever on internet-facing servers. Set up logrotate to keep it under control before your disk fills up.

Something's squatting on port 8080 and you need to know what. ss, netstat, and lsof one-liners to find the PID behind a port fast.

nohup, disown, and & all keep Linux processes alive after logout, but they work differently. Here's which one to reach for, when, and why.