Managing your VPS (start, stop, reboot, etc.)

Most of the time a VPS should simply stay on. You'll still need to restart it now and then: after a kernel update, when a service hangs, or when the server stops answering. This guide explains how to reboot a Linux VPS safely from the command line, why you should almost never power it off yourself, and what to do when it stops responding.

1. Restart a service, not the server

Most problems that make people reach for a reboot are one service misbehaving. Restarting just that service is faster and doesn't interrupt anything else:

sudo systemctl status nginx          # is it running? read the last log lines
sudo systemctl restart nginx         # restart it
sudo journalctl -u nginx -n 50       # the service's recent log

Replace nginx with the service you need: apache2 or httpd, mysql or mariadb, php8.3-fpm (use your version), docker and so on. Reboot the whole VPS only when a kernel update needs it, when several services are stuck, or when the system itself is misbehaving.

2. Does the server need a reboot?

After updates, check whether a reboot is actually required:

A new kernel only takes effect after a reboot. Library updates, for example to OpenSSL, often need only the affected services restarted.

3. Reboot safely

4. Stopping and starting a VPS

If you need the VPS stopped for a planned reason, or started again after it was switched off, open a ticket with the VPS IP address and support will do it. The same applies to a force restart when the server is frozen.

5. After the reboot: health checks

Run these once you are back in:

uptime                         # should show a boot a few minutes ago
systemctl --failed             # services that failed to start
df -h                          # disks mounted and not full
free -h                        # memory available
sudo journalctl -b -p err      # errors logged during this boot

Then open your websites and test anything that matters, such as a contact form or a login. If a service failed, systemctl status name and journalctl -u name usually say why. A common cause is a configuration file that was edited weeks ago but never reloaded until now.

6. When the VPS stops responding

Work through these before asking for a restart:

For console access when SSH is broken, see connecting to a VPS when SSH is broken. After a forced restart, run the health checks above; the filesystem check usually runs by itself during boot.

7. Watching your VPS day to day

A few commands show whether the server is healthy without any extra software:

• htop or top: live CPU and memory use per process. Press M in top to sort by memory.
• free -h: if available memory is near zero and swap is full, the kernel will start killing processes.
• df -h and du -sh /var/log/*: a full disk stops databases and mail.
• sudo journalctl -k | grep -i "out of memory": shows processes the kernel killed for memory.

For graphs and alerts over time, see production observability with Prometheus, Grafana and Loki.

8. Domain India VPS

Domain India VPS plans run on KVM with full root access and NVMe storage, and are self-managed: you run the software and routine reboots, and support helps with platform tasks such as starting or force-restarting the VPS. If a server keeps running out of memory, a larger plan may be the fix. Prices on the cards are live and exclude 18% GST.

Ready to take control of your server? Keep this checklist handy, compare VPS plans if you need more memory, or open a ticket if your VPS needs a start or restart from our side.