System Health Monitoring in RouterOS: A Complete Guide

System Health Monitoring in RouterOS: A Complete Guide

RouterOS Version: 7.x+ Difficulty: Beginner Estimated Time: 20 minutes

Overview

System health monitoring provides real-time hardware status information including temperature, voltage, current, fan speed, and power consumption. Monitoring these values helps you:

• Detect overheating before hardware damage occurs
• Verify power supplies are functioning correctly
• Monitor fan operation to prevent thermal throttling
• Track power consumption for capacity planning
• Integrate with external monitoring via SNMP

Health data is available via CLI, WinBox, API, and SNMP, making it easy to integrate with monitoring systems like MRTG, Cacti, Zabbix, or The Dude.

Key limitation: Health monitoring requires hardware support. CHR (Cloud Hosted Router) and some low-end devices have no sensors and will show an empty health menu.

Menu Reference

Menu	Purpose
/system/health	View sensor readings
/system/health/settings	Configure fan control (v7.9+)

Understanding Health Readings

Common Sensors

Sensor	Type	Description
cpu-temperature	C	CPU die temperature
board-temperature	C	Board/ambient temperature
temperature	C	General temperature sensor
voltage	V	Input voltage
current	A	Input current draw
power-consumption	W	Total power draw
fan1-speed	RPM	Fan rotation speed
psu1-state	ok/fail	Power supply status

What’s Normal?

Sensor	Normal Range	Warning
CPU Temperature	40-70°C	Above 80°C
Board Temperature	20-50°C	Above 60°C
Voltage	10-28V (varies)	Check device specs
Fan Speed	2000-6000 RPM	0 RPM when should be running

Note: CPU temperatures of 70-80°C are normal under load. The “operating temperature” in device specs (-20°C to +70°C) refers to room/ambient temperature, not CPU temperature.

Configuration Examples

Example 1: View Current Health Status

/system/health/print

Example output on a CCR device:

 #  NAME                VALUE  TYPE
 0  power-consumption   50.8   W
 1  cpu-temperature     43     C
 2  fan1-speed          5654   RPM
 3  board-temperature1  29     C
 4  voltage             24.5   V

Example 2: Filter Specific Readings

View only temperatures:

/system/health/print where type="C"

View only fan speeds:

/system/health/print where type="RPM"

Find a specific sensor:

/system/health/print where name="cpu-temperature"

Example 3: Configure Fan Control (v7.9+)

Control when fans start and reach full speed:

# View current settings
/system/health/settings/print

# Set temperature where fans start spinning
/system/health/settings/set fan-target-temp=55

# Set temperature where fans reach maximum speed
/system/health/settings/set fan-full-speed-temp=65

# Prevent fans from completely stopping (reduces cycling)
/system/health/settings/set fan-min-speed-percent=15

Fan control is available on: CRS3xx, CRS5xx, CCR2xxx (v7.9+), CCR1036/CCR1016 (v7.14+)

Example 4: Enable CPU Overtemperature Protection

For ARM/ARM64 devices, enable automatic protection if CPU overheats:

# Enable overtemperature monitoring
/system/health/settings/set cpu-overtemp-check=yes

# Set threshold (default 105°C)
/system/health/settings/set cpu-overtemp-threshold=100

# Delay after boot before monitoring (prevents false triggers)
/system/health/settings/set cpu-overtemp-startup-delay=2m

Example 5: Create Temperature Alert Script

Send email when temperature exceeds threshold:

# First configure email (required)
/tool/e-mail/set server=smtp.example.com from[email protected]

# Create alert script
/system/script/add name=temp-alert policy=read,write,test source={
    :local cpuTemp [/system/health/get [find name="cpu-temperature"] value]
    :local threshold 75
    :if ($cpuTemp > $threshold) do={
        :log warning "CPU temperature high: $cpuTemp C"
        /tool/e-mail/send to="[email protected]" \
            subject="[ALERT] Router temperature high" \
            body="CPU temperature: $cpuTemp C (threshold: $threshold C)"
    }
}

# Schedule to run every 5 minutes
/system/scheduler/add name=temp-check interval=5m on-event=temp-alert

Example 6: Temperature Alert with Rate Limiting

Prevent email spam by limiting alerts:

/system/script/add name=temp-alert-limited policy=read,write,test source={
    :global lastTempAlert
    :local cpuTemp [/system/health/get [find name="cpu-temperature"] value]
    :local threshold 75
    :local cooldown 3600

    :if ($cpuTemp > $threshold) do={
        :local now [/system/clock/get time]
        :local currentSecs ([:pick $now 0 2] * 3600 + [:pick $now 3 5] * 60)

        :if (($lastTempAlert = nil) or (($currentSecs - $lastTempAlert) > $cooldown)) do={
            :log warning "CPU temperature high: $cpuTemp C"
            /tool/e-mail/send to="[email protected]" \
                subject="[ALERT] Router temperature high" \
                body="CPU temperature: $cpuTemp C"
            :set lastTempAlert $currentSecs
        }
    }
}

Example 7: Monitor via SNMP

Get SNMP OIDs for external monitoring tools:

/system/health/print oid

Output shows OID for each sensor:

 #  NAME             VALUE  TYPE  OID
 0  cpu-temperature  43     C     .1.3.6.1.4.1.14988.1.1.3.11
 1  voltage          24.5   V     .1.3.6.1.4.1.14988.1.1.3.8

Poll from external system (Linux example):

# Get CPU temperature (returns decidegrees - divide by 10)
snmpget -v2c -c public 192.168.1.1 .1.3.6.1.4.1.14988.1.1.3.11.0

Example 8: Configure SNMP Temperature Traps

Send SNMP trap when temperature exceeds threshold:

# Enable SNMP
/snmp set enabled=yes

# Configure trap destination
/snmp set trap-target=192.168.1.100 trap-community=public trap-version=2

# Enable temperature exception traps
/snmp set trap-generators=temp-exception

Trap triggers at 100°C or the cpu-overtemp-threshold value.

Example 9: Check Power Supply Status (Dual-PSU Devices)

# View PSU status
/system/health/print where name~"psu"

Expected output:

 #  NAME          VALUE  TYPE
 0  psu1-state    ok
 1  psu2-state    ok
 2  psu1-voltage  24.2   V
 3  psu2-voltage  24.3   V

If a PSU fails, psu1-state or psu2-state will show fail.

SNMP Integration

Common Health OIDs

Reading	OID	Notes
voltage	.1.3.6.1.4.1.14988.1.1.3.8.0	Decivolts (÷10)
temperature	.1.3.6.1.4.1.14988.1.1.3.10.0	Decidegrees (÷10)
cpu-temperature	.1.3.6.1.4.1.14988.1.1.3.11.0	Decidegrees (÷10)
power-consumption	.1.3.6.1.4.1.14988.1.1.3.12.0	Deciwatts (÷10)
fan-speed	.1.3.6.1.4.1.14988.1.1.3.17.0	RPM
psu1-state	.1.3.6.1.4.1.14988.1.1.3.15.0	0=fail, 1=ok
psu2-state	.1.3.6.1.4.1.14988.1.1.3.16.0	0=fail, 1=ok

Important: SNMP returns values multiplied by 10. CLI shows 24.5V; SNMP returns 245.

Use in Monitoring Tools

Zabbix/Cacti/MRTG: Use the OIDs above with a multiplier of 0.1 for voltage/temperature/power.

The Dude: Supports MikroTik health OIDs natively; no configuration needed.

Common Problems and Solutions

Problem 1: Health Menu is Empty

Cause: Device has no hardware monitoring support.

Solution: Check device specifications at mikrotik.com. CHR and some low-end RouterBOARDs have no sensors.

Problem 2: Scripts Broken After v7 Upgrade

Cause: RouterOS v7 changed health menu structure.

v6 syntax (broken):

:local temp [/system health get temperature]

v7 syntax (correct):

:local temp [/system/health/get [find name="cpu-temperature"] value]

Problem 3: SNMP Values Look Wrong

Cause: SNMP returns decivolts/decidegrees (multiplied by 10).

Solution: Divide SNMP values by 10 in your monitoring tool.

Problem 4: Fan Shows 0 RPM

Possible causes:

1. Temperature below fan-target-temp (fans not needed)
2. Device doesn’t support fan control
3. Fan hardware failure

Check:

/system/health/settings/print
# If fan-target-temp is higher than current temp, fans won't spin

Problem 5: Cannot Control Fan Speed Directly

Cause: MikroTik doesn’t allow direct RPM control.

Solution: Use temperature thresholds to influence behavior:

/system/health/settings/set fan-target-temp=50 fan-full-speed-temp=60

Problem 6: Temperature Email Spam

Cause: Script runs repeatedly while temperature is high.

Solution: Add rate limiting (see Example 6) or track recovery:

# Only alert once until temperature recovers
:global tempAlertSent
:if ($cpuTemp > 75) do={
    :if ($tempAlertSent != true) do={
        # Send alert
        :set tempAlertSent true
    }
} else={
    :set tempAlertSent false
}

Problem 7: PoE Voltage Reads Lower Than Expected

Cause: PoE-powered devices have protection circuitry causing voltage drop in readings.

Solution: This is expected behavior. Actual input voltage is higher than displayed.

Fan Control Settings Reference

Setting	Default	Description
fan-target-temp	58°C	Temperature where fans start
fan-full-speed-temp	65°C	Temperature where fans reach max
fan-min-speed-percent	12%	Minimum fan speed (prevents cycling)
fan-control-interval	30s	Seconds between temp readings

Supported devices (v7.9+): CRS3xx, CRS5xx, CCR2xxx

Additional support (v7.14+): CCR1036, CCR1016

Verification Commands

# Check if health monitoring is available
/system/health/print
# Empty = no hardware support

# View all temperature sensors
/system/health/print where type="C"

# Check fan status
/system/health/print where type="RPM"

# View fan control settings
/system/health/settings/print

# Get SNMP OIDs
/system/health/print oid

# Check PSU status (dual-PSU devices)
/system/health/print where name~"psu.*state"

Summary

Health monitoring in RouterOS provides essential hardware status information:

1. View readings with /system/health/print
2. Configure fans with /system/health/settings (v7.9+)
3. Create alerts using scripts and scheduler

Key points:

• Available sensors vary by device model
• SNMP values are multiplied by 10 (divide for actual values)
• Fan control is indirect via temperature thresholds
• v7 changed health menu structure (update scripts accordingly)
• Alert scripts need rate limiting to prevent spam

Related Topics

Monitoring Integration

• SNMP Configuration - export health data to monitoring systems
• Netwatch - complement with connectivity monitoring

Alerting

• Email Tool - send temperature alerts
• Scheduler - run health checks periodically
• Scripts - custom health monitoring logic

System Reliability

• Watchdog - automatic recovery on system issues
• System Backup - backup before potential hardware issues

Configuration and Logging

• Logging - log health-related events