server-management

PM2 is the better fit for Node.js applications because it provides clustering and zero-downtime reloads out of the box. systemd works for any application type as a Linux-native solution, while containers and orchestration needs point toward Docker or Kubernetes.

Monitor four categories: availability (uptime, health checks), performance (response time, throughput), errors (rate and types), and resources (CPU, memory, disk). Assign alert severity levels so critical issues trigger immediate action while informational items are reviewed daily.

Vertical scaling works as a quick fix for a single instance, such as adding RAM for high memory usage. Horizontal scaling by adding instances is the sustainable approach for high CPU load, and auto-scaling fits variable traffic patterns.

Follow this priority: first check if the process is running, then check logs for error messages, then check resources like disk, memory, and CPU, then check network ports and DNS, and finally check dependencies such as databases and external APIs.

Avoid running services as root, ignoring log rotation, skipping monitoring, relying on manual restarts, and having no backup schedule. Instead use a non-root user, configure auto-restart, monitor from day one, and maintain regular backups.