OpenWrt System Maintenance and Troubleshooting

Document Scope: Daily maintenance, monitoring setup, and troubleshooting practices for Raspberry Pi + ImmortalWrt.

Table of Contents

• System Information
• Automated Monitoring
• Troubleshooting
• Storage and SD Card Maintenance
• Routine Maintenance
• Quick Reference Commands
• Important Paths

System Information

1. Hardware

• Device: Raspberry Pi 4B (aarch64_cortex-a72)
• System: ImmortalWrt 24.10.3 r33451-5531f6bc76a3
• Memory: 4GB
• Storage: 32GB SD card

2. Network

• LAN IP: 192.168.1.1
• SSH Port: 22
• Web UI: http://192.168.1.1

Automated Monitoring

Three scripts are used to keep the system stable.

1. Health Monitor (every 30 minutes)

• Script: /root/health_monitor.sh
• Checks:
  • Memory usage alert (>90%)
  • Service status (openclash, uhttpd, network, dnsmasq)
  • Connection count alert (>1000)
  • Auto-restart if service is unhealthy

Check logs:

tail -50 /root/health_monitor.log

2. LuCI Watchdog (every 5 minutes)

• Script: /root/luci_watchdog.sh
• Checks:
  • uhttpd process
  • LuCI CGI executable status
  • HTTP response
  • Permission and service auto-fix

Check logs:

tail -50 /root/luci_watchdog.log

3. Smart Backup (daily, change-triggered backup logic)

• Script: /root/smart_backup.sh
• Core features:
  • Create a new restore archive only when configuration changes are detected
  • Keep the backup output focused on sysupgrade-restorable .tar.gz archives
  • Keep the latest 3 archives locally and the latest 7 in GitHub
  • Support branch-based separation such as campus and home

Check logs:

tail -50 /root/smart_backup.log

Run backup manually:

/root/smart_backup.sh --force

4. Suggested scheduling

5. Crontab

# Run smart backup every day at 15:00
0 15 * * * /root/smart_backup.sh

2. Boot-time self-check (/etc/rc.local)

# Add before exit 0
/root/smart_backup.sh &

3. Inspect current cron jobs

crontab -l

Troubleshooting

1. haproxy crash loop

Symptom:

• System log shows haproxy::instance1 is in a crash loop

Fix (if you do not need load balancing):

/etc/init.d/haproxy stop
/etc/init.d/haproxy disable

Verify:

/etc/init.d/haproxy enabled && echo "enabled" || echo "disabled"

2. Wi-Fi connected but no internet

# 1) Check OpenClash status
/etc/init.d/openclash status

# 2) Restart OpenClash
/etc/init.d/openclash restart

# 3) Test DNS
nslookup google.com

# 4) Check related logs
logread | grep -i "openclash\\|error"

3. LuCI Web UI not accessible

# 1) Check uhttpd
/etc/init.d/uhttpd status

# 2) Process check
pgrep uhttpd

# 3) LuCI CGI permission
ls -la /www/cgi-bin/luci
chmod +x /www/cgi-bin/luci

# 4) Restart uhttpd
/etc/init.d/uhttpd restart

# 5) Watchdog log
tail -20 /root/luci_watchdog.log

4. Performance inspection

# Uptime and load
uptime

# Memory
free -h

# Disk usage
df -h

# CPU temperature (divide by 1000)
cat /sys/class/thermal/thermal_zone0/temp

# Running services
ls /etc/rc.d/ | grep "^S"

# Recent system logs
logread | tail -50

# Top memory consumers
ps aux | sort -k4 -r | head -10

Storage and SD Card Maintenance

Do not assume every Raspberry Pi image uses the same partition layout.

The two common cases are:

• ext4 image: mount shows /dev/root on / type ext4
• squashfs image: mount shows overlayfs:/overlay on / type overlay

These systems should not be expanded in the same way:

• an ext4 root filesystem is usually best handled by expanding the root partition
• a squashfs system should first distinguish between a separate data partition and extroot

Before changing partitions, always inspect:

mount
df -h
block info
uci show fstab

If you mainly want more room for packages, Git repositories, or backup archives, start with:

• Storage Expansion and Partitioning Guide

Only open the extroot guide after confirming that the system is still on the squashfs + overlay path:

• Overlay / Extroot Expansion (Advanced)

Routine Maintenance

Suggested daily checks:

1. check free space with df -h
2. check key services such as openclash and uhttpd
3. check the smart backup log
4. review recent system logs
5. create a backup before network or plugin changes

Related docs:

• Manual Backup and Restore
• ImmortalWrt GitHub Auto Backup (Legacy Reference)

Additional guidance:

1. Write and log management

2. keep backup retention limited to what you actually need

3. use /tmp for temporary processing where possible

4. restart services instead of rebooting the whole router

5. trim logs periodically if they are no longer useful

Optional cleanup:

> /root/health_monitor.log
> /root/luci_watchdog.log
> /root/smart_backup.log

2. Weekly

tail -100 /root/health_monitor.log
tail -100 /root/luci_watchdog.log
logread | tail -100
df -h
free -h
/etc/init.d/openclash status
/etc/init.d/uhttpd status

3. Monthly

First identify your OpenWrt branch:

• OpenWrt 24.10 and earlier stable releases: use opkg
• OpenWrt 25.12 and newer: use apk

# Check recent restore archives
ls -lh /root/*.tar.gz 2>/dev/null

# Optional log cleanup
> /root/health_monitor.log
> /root/luci_watchdog.log
> /root/smart_backup.log

# Check package updates (24.10 and earlier)
opkg update
opkg list-upgradable

# Check package updates (25.12 and newer)
apk update
apk list --upgradeable

# Check uptime
uptime

Quick Reference Commands

1. Service management

# OpenClash
/etc/init.d/openclash start|stop|restart|status

# LuCI
/etc/init.d/uhttpd start|stop|restart|status

# Network
/etc/init.d/network restart

# DNS/DHCP
/etc/init.d/dnsmasq restart

2. System operations

# Reboot
reboot

# Release info
cat /etc/openwrt_release
uname -a

# Network info
ifconfig
ip addr show

3. Backup commands

# Manual backup
/root/smart_backup.sh

# Backup log
tail -50 /root/smart_backup.log

Important Paths

/root/
├── health_monitor.sh
├── health_monitor.log
├── luci_watchdog.sh
├── luci_watchdog.log
├── smart_backup.sh
├── smart_backup.log
└── *.tar.gz

/etc/
├── config/
└── init.d/

/var/log/

Support

• SSH: ssh root@192.168.1.1
• Web UI: http://192.168.1.1
• Backup repository: https://github.com/Ronchy2000/Immortalwrt-AutoBackup