Practical Linux System Administration

Chapter 9. Monitoring Your System

Monitoring is not optional. It’s essential. Every system administrator must monitor their system’s performance, health, and security with a vigilant eye. It’s impossible to manually monitor your systems and also perform your other administrative tasks, so you must rely on automated software, specific system configurations, and a reporting system to update you on each system’s status.

This chapter explores native monitoring tools and guides you in gathering performance and health statistics for your systems. 

Maintaining Vigilance on CPU, Memory, and Disk Performance and Capacity

The three primary monitoring focal points in Linux systems are CPU, memory, and disk performance.

Note

Network capacity on local area networks (LANs) isn’t usually much of an issue, especially with gigabit network speeds. Many data centers/server rooms use 10-gigabit connections so that system-to-system communications are practically instantaneous and backups, even huge ones, are of no consequence. That said, checking network performance will only require a few minutes of your day, so why not include it? The data exists and is freely available to you, so display it, study it, and act on it if you see a negative trend.

This chapter focuses on CPU, memory, and disk performance and capacity.

Tracking CPU Usage

The CPU is often at the top of the list of system components to monitor. It’s at the top of the list because it’s the system’s powerhouse. Suppose your system’s ...