adminLinux Boot and Systemd Startup Flow Checklist
A Linux server does not become ready all at once. It boots the kernel, initializes systemd, starts units, mounts disks, brings up networking and then starts services like Nginx, MySQL, Docker or automation workers.
Core principle
If something works manually but fails after reboot, the problem is often startup order, disabled service, missing dependency, wrong unit file or a service that starts before the network or disk is ready.
Checklist
- Check server uptime and last reboot time.
- Check failed systemd units.
- Check whether important services are enabled on boot.
- Check startup logs from the current boot.
- Check whether network is ready before dependent services start.
- Check whether mounted disks are available before services need them.
- Check service dependencies.
- Check slow boot services.
- Restart only the affected service after understanding the failure.
- Document the startup chain for critical services.
Reusable lesson
This applies to websites, databases, Docker stacks, n8n, custom workers, monitoring agents and any background service that must survive reboot.
When to Use This Checklist
Use this checklist when a service works manually but fails after VPS reboot, system update or startup sequence change.
Required Tools
SSH access, Ubuntu VPS, systemctl, journalctl, systemd-analyze, service names, reboot history
Before You Start
Do not keep restarting services without checking whether they are enabled, failed or blocked by startup dependencies.
Structured Checklist Steps
- Check uptime.
- List failed units.
- Check service enable status.
- Read boot logs.
- Check network dependency.
- Check mount dependency.
- Check unit dependencies.
- Check slow boot services.
- Restart affected service.
- Document startup chain.
Verification Steps
- No critical failed unit remains unexplained.
- Required services are enabled.
- Boot logs show no repeated startup failure.
- Service survives reboot.
- Startup dependencies are documented.
Rollback Plan
If a systemd change breaks boot behavior, restore the previous unit file, run daemon-reload and restart only the affected service.
Common Mistakes
- Assuming reboot fixes everything.
- Forgetting to enable services.
- Ignoring failed units.
- Not checking journal logs from current boot.
- Starting app services before network or disk is ready.
Related Commands
uptime
systemctl --failed
systemctl is-enabled nginx
journalctl -b --no-pager | tail -n 100
systemd-analyze blame
systemctl list-dependencies nginx
sudo systemctl daemon-reload