System statistics

Plugin: proc.plugin Module: /proc/stat

Overview

CPU utilization, states and frequencies and key Linux system performance metrics.

The /proc/stat file provides various types of system statistics:

• The overall system CPU usage statistics
• Per CPU core statistics
• The total context switching of the system
• The total number of processes running
• The total CPU interrupts
• The total CPU softirqs

The collector also reads:

• /proc/schedstat for statistics about the process scheduler in the Linux kernel.
• /sys/devices/system/cpu/[X]/thermal_throttle/core_throttle_count to get the count of thermal throttling events for a specific CPU core on Linux systems.
• /sys/devices/system/cpu/[X]/thermal_throttle/package_throttle_count to get the count of thermal throttling events for a specific CPU package on a Linux system.
• /sys/devices/system/cpu/[X]/cpufreq/scaling_cur_freq to get the current operating frequency of a specific CPU core.
• /sys/devices/system/cpu/[X]/cpufreq/stats/time_in_state to get the amount of time the CPU has spent in each of its available frequency states.
• /sys/devices/system/cpu/[X]/cpuidle/state[X]/name to get the names of the idle states for each CPU core in a Linux system.
• /sys/devices/system/cpu/[X]/cpuidle/state[X]/time to get the total time each specific CPU core has spent in each idle state since the system was started.

This collector is only supported on the following platforms:

• linux

This collector only supports collecting metrics from a single instance of this integration.

Default Behavior

Auto-Detection

The collector auto-detects all metrics. No configuration is needed.

Limits

The default configuration for this integration does not impose any limits on data collection.

Performance Impact

The collector disables cpu frequency and idle state monitoring when there are more than 128 CPU cores available.

Metrics

Metrics grouped by scope.

The scope defines the instance that the metric belongs to. An instance is uniquely identified by a set of labels.

Per System statistics instance

This scope has no labels.

Metrics:

Metric	Dimensions	Unit
system.cpu	guest_nice, guest, steal, softirq, irq, user, system, nice, iowait, idle	percentage
system.intr	interrupts	interrupts/s
system.ctxt	switches	context switches/s
system.forks	started	processes/s
system.processes	running, blocked	processes
cpu.core_throttling	a dimension per cpu core	events/s
cpu.package_throttling	a dimension per package	events/s
cpu.cpufreq	a dimension per cpu core	MHz

Per cpu core

Labels:

Label	Description
cpu	TBD

Metrics:

Metric	Dimensions	Unit
cpu.cpu	guest_nice, guest, steal, softirq, irq, user, system, nice, iowait, idle	percentage
cpuidle.cpu_cstate_residency_time	a dimension per c-state	percentage

Alerts

The following alerts are available:

Alert name	On metric	Description
10min_cpu_usage	system.cpu	average CPU utilization over the last 10 minutes (excluding iowait, nice and steal)
10min_cpu_iowait	system.cpu	average CPU iowait time over the last 10 minutes
20min_steal_cpu	system.cpu	average CPU steal time over the last 20 minutes

Setup

Prerequisites

No action required.

Configuration

File

The configuration file name for this integration is netdata.conf. Configuration for this specific integration is located in the plugin:proc:/proc/stat section within that file.

The file format is a modified INI syntax. The general structure is:

[section1]
    option1 = some value
    option2 = some other value

[section2]
    option3 = some third value

You can edit the configuration file using the edit-config script from the Netdata config directory.

cd /etc/netdata 2>/dev/null || cd /opt/netdata/etc/netdata
sudo ./edit-config netdata.conf

Options

There are no configuration options.

Examples

There are no configuration examples.