June 11, 2026 16 min to read

Linux Physical Server Hardware Monitoring Guide: Complete Toolkit

Monitoring CPU temperature, fan speed, voltage, power consumption, and throttling on bare-metal servers

Overview

When operating physical servers, it’s essential to monitor various hardware metrics such as CPU temperature, fan speed, voltage, power consumption, and throttling (performance degradation) status.

This article introduces various commands and tools for checking hardware sensor information and real-time management on Ubuntu-based Linux servers. We’ll cover tools applicable to servers from Supermicro, Dell, HP including ipmitool, lm-sensors, powerstat, and nvme-cli.

1. CPU Temperature Monitoring with lm-sensors

Installation and Detection

sudo apt install lm-sensors -y
sudo sensors-detect

Repeatedly enter YES to detect all available sensors:

coretemp (Intel CPU)
nct6775 (Fan, voltage sensors)

Example Detected Sensor Modules

coretemp
nct6775
it87
k10temp  # AMD CPUs

Load Kernel Modules Immediately

sudo modprobe coretemp
sudo modprobe nct6775

Display Sensor Values

sensors

Example Output:

coretemp-isa-0000
Adapter: ISA adapter
Package id 0:  +45.0°C  (high = +80.0°C, crit = +100.0°C)
Core 0:        +42.0°C  (high = +80.0°C, crit = +100.0°C)
Core 1:        +44.0°C  (high = +80.0°C, crit = +100.0°C)
Core 2:        +43.0°C  (high = +80.0°C, crit = +100.0°C)
Core 3:        +45.0°C  (high = +80.0°C, crit = +100.0°C)

nct6775-isa-0290
Adapter: ISA adapter
fan1:        1200 RPM  (min =  800 RPM)
fan2:        1150 RPM  (min =  800 RPM)

Tip: Add module names to /etc/modules for automatic loading after reboot:

echo "coretemp" | sudo tee -a /etc/modules
echo "nct6775" | sudo tee -a /etc/modules

2. System Resource Monitoring

Essential Monitoring Tools

Tool	Purpose
htop	Real-time CPU, memory, thread monitoring (GUI-based)
iotop	Real-time disk read/write tracking
iftop	Network traffic tracking (per interface)

htop - CPU and Memory

sudo apt install htop
htop

Key Features:

Color-coded CPU usage bars
Per-process CPU/memory consumption
Load averages
Thread view (F2 → Display options → Show custom thread names)

iotop - Disk I/O

sudo apt install iotop
sudo iotop

Useful Options:

sudo iotop -o  # Only show processes with I/O activity
sudo iotop -a  # Accumulated I/O instead of bandwidth

iftop - Network Traffic

sudo apt install iftop
sudo iftop -i eth0  # Specify interface

Key Shortcuts:

t - Toggle between one-line, two-line, three-line display
n - Toggle DNS resolution
p - Toggle port display

3. Fan Speed, Voltage, Power (IPMI-Supported Servers)

ipmitool Installation

sudo apt install ipmitool

View All Sensors

sudo ipmitool sensor

Example Output:

Temp             | 38.000     | degrees C  | ok    | na        | na        | na        | 85.000    | 90.000    | na        
CPU Temp         | 45.000     | degrees C  | ok    | na        | na        | na        | 95.000    | 100.000   | na        
FAN1             | 2400.000   | RPM        | ok    | 300.000   | 500.000   | 700.000   | 25300.000 | 25400.000 | 25500.000 
FAN2             | 2350.000   | RPM        | ok    | 300.000   | 500.000   | 700.000   | 25300.000 | 25400.000 | 25500.000 
Vcore            | 1.040      | Volts      | ok    | 0.640     | 0.720     | 0.800     | 1.520     | 1.600     | 1.680     
12V              | 12.192     | Volts      | ok    | 10.176    | 10.416    | 10.656    | 12.960    | 13.200    | 13.440    
PSU1 Status      | 0x1        | discrete   | 0x0100| na        | na        | na        | na        | na        | na        

Requirements

IPMI must be enabled in BIOS or BMC:

Supermicro: BMC configuration
Dell: iDRAC
HP: iLO (Integrated Lights-Out)

View Specific Sensors

# Temperature
sudo ipmitool sensor get "CPU Temp"

# Fan speed
sudo ipmitool sensor get "FAN1"

# Power supply status
sudo ipmitool sensor get "PSU1 Status"

4. Intel CPU Power Consumption and Throttling

powerstat

sudo apt install powerstat
sudo powerstat

Example Output:

Time    User  Nice   Sys  Idle    IO  Run Ctxt/s  IRQ/s   Watts
30:01  5.2   0.0   1.8  92.8   0.2    2   1234   2456   45.2
30:06  4.9   0.0   2.1  92.7   0.3    1   1198   2389   44.8
30:11  5.5   0.0   1.9  92.4   0.2    3   1267   2512   46.1

turbostat - Detailed CPU Metrics

sudo apt install linux-tools-common linux-tools-$(uname -r)
sudo turbostat

Key Metrics:

Package: Package power consumption (Watts)
Core: Core power consumption
Frequency: Current CPU frequency
C-state: Power-saving state depth
Throttling: Performance throttling status

Example Output:

Core  CPU  Avg_MHz  Busy%  Bzy_MHz  TSC_MHz  Package  Core   PkgTmp  PkgWatt  CorWatt
-     -    1234     24.68   5000     2400     45       22     56      45.2     22.1
0     0    1200     24.00   5000     2400     45       22     55      -        -
0     4    1180     23.60   5000     2400     -        -      -       -        -
1     1    1290     25.80   5000     2400     -        23     56      -        -
1     5    1245     24.90   5000     2400     -        -      -       -        -

5. NVMe SSD Status and Temperature

nvme-cli Installation

sudo apt install nvme-cli

SMART Log

sudo nvme smart-log /dev/nvme0

Example Output:

Smart Log for NVME device:nvme0 namespace-id:ffffffff
critical_warning                    : 0
temperature                         : 45 C
available_spare                     : 100%
available_spare_threshold           : 10%
percentage_used                     : 2%
data_units_read                     : 12,345,678
data_units_written                  : 8,765,432
host_read_commands                  : 234,567,890
host_write_commands                 : 123,456,789
controller_busy_time                : 1,234
power_cycles                        : 50
power_on_hours                      : 8,760
unsafe_shutdowns                    : 0
media_errors                        : 0
num_err_log_entries                 : 0

Critical Metrics

Metric	Description	Healthy Range
Temperature	Current SSD temperature	< 70°C
Available Spare	Remaining spare blocks	> 80%
Percentage Used	Wear indicator	< 80%
Media Errors	Bad sectors	0

List All NVMe Devices

sudo nvme list

6. System Hardware Information Summary

lshw - Hardware Lister

sudo lshw -short

Example Output:

H/W path        Device      Class          Description
======================================================
                            system         Computer
/0                          bus            Motherboard
/0/0                        memory         32GiB System Memory
/0/1                        processor      Intel Core i7-9700K
/0/100/1c.4/0   eth0        network        Ethernet Controller
/0/100/1f.2                 storage        SATA Controller

inxi - System Information Tool

sudo apt install inxi
sudo inxi -Fxz

Features:

CPU, GPU, Network card info
Memory slot details
Disk information
Sensor readings
Battery status (for laptops)

7. Disk Health (SMART) Check

smartmontools Installation

sudo apt install smartmontools

Full SMART Report

sudo smartctl -a /dev/sda

Key Attributes:

ID# ATTRIBUTE_NAME          FLAG     VALUE WORST THRESH TYPE      UPDATED  WHEN_FAILED RAW_VALUE
Reallocated_Sector_Ct   0x0033   100   100   010    Pre-fail  Always       -       0
Power_On_Hours          0x0032   099   099   000    Old_age   Always       -       8760
Power_Cycle_Count       0x0032   100   100   000    Old_age   Always       -       50
Temperature_Celsius     0x0022   042   055   000    Old_age   Always       -       42
Current_Pending_Sector  0x0012   100   100   000    Old_age   Always       -       0
Offline_Uncorrectable   0x0010   100   100   000    Old_age   Offline      -       0

Critical Attributes to Monitor

ID	Attribute	Critical?	Action
5	Reallocated Sectors	⚠️ Yes	Replace if > 0 and increasing
197	Pending Sectors	⚠️ Yes	Replace immediately if > 0
198	Uncorrectable Sectors	⚠️ Yes	Replace immediately if > 0
194	Temperature	⚠️ Monitor	Check cooling if > 55°C

Test Disk

# Short test (2 minutes)
sudo smartctl -t short /dev/sda

# Long test (hours)
sudo smartctl -t long /dev/sda

# View test results
sudo smartctl -l selftest /dev/sda

8. Thermal Throttling Manual Check

Thermal Zone Temperatures

cat /sys/class/thermal/thermal_zone*/temp

Output is in millidegrees Celsius (divide by 1000):

45000  # 45°C
52000  # 52°C

Intel Turbo Boost Status

cat /sys/devices/system/cpu/intel_pstate/no_turbo

Values:

0: Turbo Boost enabled (default)
1: Turbo Boost disabled

Disable Turbo Boost

echo 1 | sudo tee /sys/devices/system/cpu/intel_pstate/no_turbo

Use Cases:

Reduce heat during sustained high load
Lower power consumption
Stabilize performance (avoid frequency fluctuations)

Re-enable Turbo Boost

echo 0 | sudo tee /sys/devices/system/cpu/intel_pstate/no_turbo

9. IPMI Remote Access Configuration

Enable IPMI LAN in BIOS

If IPMI LAN is enabled in BIOS, you can access remotely:

ipmitool -I lanplus -H <IPMI_IP> -U ADMIN -P ADMIN sensor

Web UI Access

http://<IPMI_IP>

Vendor-Specific Names:

Supermicro: BMC (Baseboard Management Controller)
Dell: iDRAC (Integrated Dell Remote Access Controller)
HP: iLO (Integrated Lights-Out)

Common IPMI Commands

# System information
ipmitool -I lanplus -H <IP> -U <USER> -P <PASS> fru print

# SEL (System Event Log)
ipmitool -I lanplus -H <IP> -U <USER> -P <PASS> sel list

# Power status
ipmitool -I lanplus -H <IP> -U <USER> -P <PASS> chassis power status

10. System Power and Fan RPM Control

Power Management

# Check power status
ipmitool chassis power status

# Power on
ipmitool chassis power on

# Power off
ipmitool chassis power off

# Hard reset
ipmitool chassis power cycle

# Soft shutdown
ipmitool chassis power soft

Fan Control

# Get specific fan status
ipmitool sensor get "FAN1"

# Set fan to manual mode (Supermicro)
ipmitool raw 0x30 0x45 0x01 0x00

# Set fan speed (0-100%)
ipmitool raw 0x30 0x70 0x66 0x01 0x00 0x64  # 100%
ipmitool raw 0x30 0x70 0x66 0x01 0x00 0x32  # 50%

# Return to automatic mode
ipmitool raw 0x30 0x45 0x01 0x01

⚠️ Warning: Manual fan control can lead to overheating. Monitor temperatures closely.

11. Detailed CPU Throttling Diagnostics

turbostat Advanced Usage

sudo turbostat --Summary --interval 5

Key Fields

Field	Description
PkgWatt	Package power consumption (Watts)
CorWatt	Core power consumption (Watts)
PkgTmp	Package temperature (°C)
GFXMHz	Integrated GPU clock (MHz)
Busy%	CPU utilization percentage
Bzy_MHz	Average frequency when busy
CPU%c1	C1 state residency (light sleep)
CPU%c6	C6 state residency (deep sleep)

Interpreting Results

Normal Operation:

Bzy_MHz close to base frequency
Low C6 residency during load
Temperature < 80°C

Throttling Detected:

Bzy_MHz significantly below base frequency
High temperature (> 90°C)
Frequent C6 entries under load

12. Tracking Thermal Events

Kernel Messages

dmesg | grep -i thermal

Example Output:

[  567.890123] thermal thermal_zone0: critical temperature reached (100 C), shutting down
[  890.123456] CPU0: Package temperature above threshold, cpu clock throttled

System Logs

journalctl -k | grep -i throttle
journalctl -k | grep -i thermal

Continuous Monitoring

# Watch thermal events in real-time
watch -n 1 'dmesg | grep -i thermal | tail -20'

Management Tips and Checklist

Monitoring Checklist

Item	Recommended Range	Check Frequency
CPU Temperature	35~80°C	Regular (sensors)
FAN RPM	> 800 RPM	When noise increases
SSD Wear Level	> 80% remaining	Monthly
SMART Errors	0	Weekly
IPMI Sensor Anomalies	None	During load increases
Turbo Status	Enabled (unless thermal issues)	When throttling suspected

Alert Thresholds

# Example threshold script
#!/bin/bash

CPU_TEMP=$(sensors | grep 'Package id 0' | awk '{print $4}' | sed 's/+//;s/°C//')
FAN_RPM=$(sensors | grep 'fan1' | awk '{print $2}')
DISK_TEMP=$(smartctl -A /dev/sda | grep Temperature | awk '{print $10}')

if (( $(echo "$CPU_TEMP > 85" | bc -l) )); then
    echo "ALERT: CPU temperature critical: ${CPU_TEMP}°C"
fi

if (( FAN_RPM < 800 )); then
    echo "ALERT: Fan speed low: ${FAN_RPM} RPM"
fi

if (( DISK_TEMP > 55 )); then
    echo "ALERT: Disk temperature high: ${DISK_TEMP}°C"
fi

Integration with Monitoring Systems

Prometheus + Node Exporter

# Install node_exporter
wget https://github.com/prometheus/node_exporter/releases/download/v1.6.1/node_exporter-1.6.1.linux-amd64.tar.gz
tar xvfz node_exporter-1.6.1.linux-amd64.tar.gz
sudo mv node_exporter-1.6.1.linux-amd64/node_exporter /usr/local/bin/

Systemd Service (/etc/systemd/system/node_exporter.service):

[Unit]
Description=Node Exporter
After=network.target

[Service]
User=node_exporter
ExecStart=/usr/local/bin/node_exporter --collector.hwmon

[Install]
WantedBy=multi-user.target

Grafana Dashboard

Import dashboard ID: 1860 (Node Exporter Full)

Key Panels:

CPU Temperature
Fan Speeds
Disk I/O
Network Traffic
Power Consumption

SNMP Integration

sudo apt install snmpd snmp

# Configure /etc/snmp/snmpd.conf
rocommunity public
syslocation "Datacenter 1"
syscontact "admin@example.com"

# Restart service
sudo systemctl restart snmpd

Troubleshooting Common Issues

Issue 1: sensors-detect Finds No Sensors

Solutions:

# Update kernel
sudo apt update && sudo apt upgrade

# Load generic driver
sudo modprobe coretemp

# Check BIOS settings (enable HW monitoring)

Issue 2: IPMI Not Working

Diagnosis:

# Check if IPMI kernel module loaded
lsmod | grep ipmi

# Load manually
sudo modprobe ipmi_devintf
sudo modprobe ipmi_si

Solutions:

Enable IPMI in BIOS
Check BMC firmware version
Reset BMC from BIOS

Issue 3: High Temperature Despite Low Load

Diagnosis:

# Check thermal paste age
sensors

# Check fan operation
sensors | grep fan

# Check dust accumulation
sudo ipmitool sensor | grep -i fan

Solutions:

Clean dust from heatsinks
Replace thermal paste
Verify fan operation
Check ambient temperature

Issue 4: NVMe Not Detected

# Check if NVMe module loaded
lsmod | grep nvme

# Load module
sudo modprobe nvme

# List PCIe devices
lspci | grep -i nvme

Automated Monitoring Scripts

Daily Health Check Script

Cron Job Setup

# Edit crontab
crontab -e

# Add daily check at 2 AM
0 2 * * * /usr/local/bin/hardware-check.sh > /var/log/hardware-check.log 2>&1

# Or hourly monitoring
0 * * * * /usr/local/bin/hardware-monitor.sh

Conclusion

Hardware sensor and resource monitoring is the core of server operational stability. Early detection of temperature rises, voltage abnormalities, disk degradation, etc., can prevent failures before they occur.

The commands introduced above are easily applicable in most Linux environments, and can be visualized using Grafana + Prometheus + node_exporter if needed.

Key Points:

✅ Linux servers are more directly affected by hardware conditions than virtual environments

✅ Regular sensor checks cost nothing but prevent major failures

✅ Especially during summer or sustained high-load periods, monitor CPU throttling and fan status

Best Practices:

Set up automated monitoring
Define alert thresholds
Regular maintenance schedule
Document baseline metrics
Test alerting system

Check your hardware before it gets too hot. Prevention is better than emergency repairs.