Quick wins for a faster PC:
Scan for outdated or missing drivers - takes under a minuteDriver Scan →Repair Windows errors before they cause bigger problemsFix Now →Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Reinstalling Ubuntu on a VPS normally means replacing the server’s system disk from your hosting provider’s control panel. Choose Rebuild, Reinstall, Restore base image, or create a replacement server, select the required Ubuntu image, and confirm the destructive operation. The process erases the existing operating system and anything stored on the replaced disk, so inventory the machine, stop writes, make and verify backups, and keep a recovery path before you click the final confirmation.
Antihost offers Linux VPS plans with full root access, unlimited traffic on a 1 Gbps port and a 14-day free trial.
What a VPS “reinstall” actually does
A VPS provider, rather than Ubuntu itself, usually performs the reinstall. The provider writes a fresh Ubuntu image to the virtual disk and then presents the server as a newly installed machine. The exact label varies: Rebuild, Reinstall, Restore base image, or a new-server workflow can all produce a clean Ubuntu environment.
As an Amazon Associate I earn from qualifying purchases.
This is different from reinstalling packages or repairing a broken boot loader. A rebuild replaces the operating-system disk. Existing users, SSH configuration, firewall rules, web and database configuration, certificates, scheduled jobs, application files, secrets, and locally stored data disappear unless they exist on an independent volume or in a backup that you restore later.
DigitalOcean’s documentation gives the risk plainly: if a Droplet has no backups, snapshots, or local copies and you rebuild it, the data is completely irretrievable by DigitalOcean. Treat the provider confirmation dialog as a destruction warning, not as a routine software update.
#1 Best Overall
Decide whether you need a reinstall, an upgrade, or a replacement
First identify the outcome you want. A reinstall is appropriate when you want a known-clean operating system, the current installation is compromised or unmanageable, or configuration drift is more expensive to untangle than rebuilding. It is not the default answer when the only goal is a newer Ubuntu release.
Reinstall versus in-place release upgrade
| Decision factor | Provider rebuild/reinstall | In-place Ubuntu upgrade |
|---|---|---|
| Resulting system | Fresh image with only the software and configuration you deliberately restore. | Existing system and installed services remain, with packages and release components changed in place. |
| Data and configuration | Everything on the replaced disk must be restored from independent copies. | Configuration is generally preserved, but obsolete packages, custom repositories, and old settings can affect the result. |
| Downtime | Includes image deployment, bootstrap, data restoration, testing, and any DNS cutover. | Usually one maintenance window, but package migrations and reboots can extend it. |
| Application compatibility | You can install only versions supported by the clean image and test them before cutover. | Existing applications must tolerate the release transition and any dependency changes. |
| Rollback | Requires a snapshot, an untouched old server, or a tested backup; the old disk is not a rollback mechanism after replacement. | Rollback is complex unless you have a full image or server snapshot from before the upgrade. |
For Ubuntu Server and cloud images, Canonical recommends sudo do-release-upgrade when the objective is a newer release. Back up first and follow the supported sequence: Ubuntu expects LTS releases to be upgraded sequentially rather than skipped across multiple LTS versions. Canonical says an LTS release receives five years of standard support and security updates, while an interim release is supported for nine months. Those periods describe Ubuntu’s standard support policy, not a guarantee that every VPS image or provider template remains available for that entire time.
Free tools Windows power users keep installed
One-click scans. No signup required.
Rebuild versus creating a replacement VPS
| Consideration | Rebuild the existing VPS | Create a replacement VPS |
|---|---|---|
| Data-loss exposure | High at the moment of confirmation because the existing system disk is replaced. | The old server can remain untouched while you build and test the new one; deleting it later is still destructive. |
| IP address | DigitalOcean documents that rebuilding a Droplet retains its IP address. | A replacement normally has a different address; verify your provider’s behavior and update DNS, allowlists, and integrations. |
| Rollback path | Depends on a snapshot, backup, or separately retained copy made before rebuilding. | Switch traffic back to the old server while the new server is being corrected, provided the old server and its data remain available. |
| Downtime | Traffic is interrupted while the image is deployed and the application is restored. | You can keep the old service online until validation, then make a short cutover window. |
| Testing before DNS cutover | Limited unless you use a temporary hostname, direct IP, or provider console. | Strong: test the replacement privately before changing DNS. |
| Console and recovery access | Existing provider console object remains, but the operating system and SSH host key change. | Both servers remain available for comparison and recovery until you retire the old one. |
| Snapshots and backups | Take a final snapshot or provider backup before the rebuild, then verify independent copies. | You can snapshot the old and new systems separately during migration. |
| Ongoing cost | Usually one VPS resource, subject to the provider’s billing rules. | Two resources overlap during migration. DigitalOcean documents a 672-hour (28-day) bundled-plan monthly charging threshold for eligible Droplets; this is provider- and plan-specific and must not be generalized to other hosts. |
Choose a replacement when the service is important enough to justify a parallel test, when the current server must stay available, or when you need to change the VPS size, region, disk layout, or network design at the same time. Choose a rebuild when retaining the existing resource and, where documented, its IP address is more valuable than a parallel migration.
Prepare a complete inventory before touching the control panel
Do not rely on memory. Record the current state in a file stored outside the VPS. This inventory becomes your rebuild checklist and exposes dependencies that are easy to miss during an emergency.
- Ubuntu and kernel: release and architecture from
cat /etc/os-releaseanduname -a. - Networking: public and private IP addresses, interface names, routes, DNS records, reverse-DNS settings, floating or reserved addresses, and provider firewall rules.
- Accounts and access: usernames, sudo membership, authorized SSH keys, service accounts, console credentials, the SSH private key, and a second administrative path.
- Storage: attached volumes, mount points, filesystem types, disk usage, application upload directories, and any object-storage or network-storage mounts.
- Services: web server, reverse proxy, runtime versions, database engines, queues, caches, container workloads, and listening ports.
sudo ss -tulpnandsystemctl list-units --type=service --state=runningprovide a starting point. - Configuration: relevant files under
/etc, virtual-host definitions, environment files, service unit overrides, package repositories, TLS certificate paths, and application secrets. Never put secret values in a public ticket or shared document. - Automation: user and system cron entries, systemd timers, deployment hooks, backup jobs, monitoring checks, log shipping, and scheduled game-server restarts.
- Software inventory: package selections, language runtimes, container images, manually installed binaries, and application versions. On Debian-based systems,
dpkg --get-selectionscan record package selections, but it is not a substitute for backing up configuration and application data.
Also write down dependencies outside the VPS: DNS provider, mail relay, payment or game-session callbacks, IP allowlists, API tokens, and operators who must be notified. A clean operating system does not recreate any of these relationships.
Back up data in a way you can actually restore
Stop or quiesce writes before taking application backups. A file copy made while a database is changing may be unreadable or internally inconsistent.
The Tool Desk
Outbyte PC Repair FREERepair Windows errors before they cause bigger problemsFix Now →Outbyte Driver Updater FREEScan for outdated or missing drivers - takes under a minuteDriver Scan →Export databases consistently
Use the database engine’s supported dump or backup method, record the command and version, and include users, permissions, extensions, and scheduled jobs where the engine stores them separately. For a small maintenance window, stop the application writers first; for a larger service, use the engine’s consistent online backup facilities. Keep the resulting dump outside the VPS and test opening or restoring it on a disposable system.
Copy files and configuration
Copy application source, uploads, generated assets, deployment manifests, and the required portions of /etc with rsync or scp. Preserve ownership, permissions, symbolic links, and hidden files. Exclude volatile paths such as live process filesystems unless a particular service’s documentation requires them. Encrypt backups that contain private keys, credentials, user data, or database contents.
A typical file transfer has the form rsync -aHAX --numeric-ids /srv/myapp/ backup-host:/safe/location/myapp/; adapt paths, destination, and privileges to your environment rather than copying this command unchanged. For a remote copy over SSH, scp is suitable for a small set of files. Verify the destination by listing files, checking sizes or hashes, and reading a sample archive from a separate machine.
Use provider snapshots without treating them as your only copy
Take a provider snapshot or enable the provider’s backup feature before rebuilding, but retain an independent copy as well. A snapshot tied to the same account, region, or billing system is useful for rapid recovery, yet it does not replace a tested export stored elsewhere. DigitalOcean and Hetzner both document recovery workflows; Hetzner’s Rescue System can expose the disk so you can mount partitions and copy files with scp or rsync before reinstalling.
Prove that the backup is usable
- Check that the backup job completed rather than merely starting.
- Confirm that the destination can be reached with the credentials you will have after the rebuild.
- Restore a database dump to a temporary instance and run representative queries.
- Extract a sample archive and inspect ownership, permissions, certificates, and application uploads.
- Record the backup location, encryption key, retention date, and restore command in your external runbook.
Record access details and plan for a new SSH fingerprint
Keep the provider’s web or serial console available before starting. Save the SSH private key, provider recovery credentials, the intended administrative username, and a second path such as a console session or another administrator’s key. After a rebuild, the machine presents a new SSH host key. DigitalOcean describes this as a new fingerprint; your SSH client may warn that the host identification has changed.
Remove only the old, expected entry from your workstation’s known_hosts file after confirming that the provider operation really targeted the intended server. Do not disable host-key checking globally to silence the warning. Compare the new fingerprint through the provider console or another trusted channel before accepting it.
Rank #4
Run the provider rebuild or create the replacement
DigitalOcean workflow
- Open the intended Droplet in the control panel.
- Choose Settings → Rebuild, or use Actions → Restore base image where that control is shown.
- Select the Ubuntu image and release you have chosen. Confirm the architecture and region-specific image availability before proceeding.
- Type the Droplet name when the confirmation form asks for it. This is an explicit acknowledgement that the disk will be replaced.
- Start the operation and wait for the control panel to report completion before attempting normal SSH access.
Rebuilding keeps the existing Droplet resource and, according to DigitalOcean’s documentation, retains its IP address. It does not preserve the files on the replaced disk.
Hetzner workflow
Hetzner documents both a Rebuild action and the option to create a new server. If the current operating system will not boot, enable the Rescue System first, mount the root partition, and copy required files to independent storage with scp or rsync. For a parallel migration, create the new server, select the desired Ubuntu image, and leave the old server running until validation and DNS cutover are complete.
Other providers
Look for a control-plane action named Rebuild, Reinstall, Reimage, or Restore base image. Read the confirmation text for disk, volume, snapshot, and IP behavior. Some hosts distinguish the boot disk from attached volumes; do not assume an attached volume is safe until the provider explicitly states what the operation changes. If the panel cannot state what is erased, open a support request or use a replacement server instead of guessing.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Bootstrap the clean Ubuntu installation
Use the provider-supplied user and SSH key for the first login. Work from the console if network access is not yet configured.
Best Value
- Confirm identity and release: check the hostname, public IP, Ubuntu release, architecture, and disk layout so you are operating on the intended server.
- Update package metadata and installed packages: run
sudo apt updatefollowed bysudo apt full-upgrade, then reboot if the installed kernel or services require it. - Create a non-root administrator: add a named user, grant it sudo access, install its public SSH key, and test a second session before closing the provider session.
- Enforce key-based SSH: verify key login works, then disable password authentication and direct root SSH only after confirming the alternative administrative path. Keep a console route for recovery.
- Set identity and time: configure the hostname, time zone, and time synchronization required by logs, certificates, databases, and scheduled jobs.
- Configure network filtering: allow SSH from approved sources and open only ports required by the application. Coordinate host firewall rules with provider-level firewalls so one layer does not silently block the other.
- Install required services: add the web server, runtime, database, queue, container engine, game server, monitoring agent, and backup client only as needed. Avoid restoring an old package repository without checking that it supports the selected Ubuntu release.
- Restore configuration carefully: copy files into the new paths, review every secret and permission, recreate systemd units and timers, and regenerate machine-specific values rather than blindly cloning them.
- Restore data: import the tested database dump, mount independent volumes, copy application files, and verify ownership before starting public traffic.
- Re-enable operations: configure TLS certificates, outbound mail, log rotation, monitoring, backups, and alert destinations. Run one backup manually and confirm it reaches the external destination.
Do not copy the entire old /etc over the new installation. Select files deliberately: wholesale replacement can overwrite networking, SSH, package-manager, and system-service defaults that the new release expects.
Recommended Free Tools
Validate the server before sending users to it
Validation is the point at which a replacement server earns the right to receive production traffic. Use a temporary hostname, direct IP with an appropriate test configuration, or an internal network path so you can test without changing public DNS first.
- Access: log in with the non-root account, exercise sudo, verify console access, and confirm the new SSH fingerprint through a trusted channel.
- Network: test private and public routes, provider and host firewalls, listening ports, DNS resolution, and outbound connections required by the application.
- Web or game endpoints: request health checks and representative pages or join a private test session. Confirm static files, uploads, authentication, matchmaking or background workers, and rate limits.
- Database integrity: run migrations only after a fresh backup, check row counts or application-level health checks, and test create, read, update, and delete paths appropriate to the service.
- Certificates: verify the certificate chain, names, renewal timer, and private-key permissions. A certificate copied without its renewal configuration will eventually expire.
- Jobs and workers: trigger one scheduled job, inspect logs, and verify queues, timers, cron entries, and restart policies.
- Mail and notifications: send a controlled test through the configured relay and check that monitoring alerts arrive at the intended destination.
- Backups and monitoring: run both successfully, confirm retention, and ensure alerts identify the new hostname and IP.
- Capacity: inspect CPU, memory, disk, inode, file-descriptor, and provider quota usage under a representative load. A clean image does not correct an undersized VPS.
Keep the old server and its snapshot untouched while these checks run. Once the restored system passes, lower DNS TTL in accordance with your normal change policy, update records and allowlists, monitor the cutover, and remove the old resource only after the new backup has completed and the rollback window has ended.
Recovery paths when the reinstall goes wrong
The provider operation fails or the VPS will not boot
Use the provider console or Rescue System rather than repeatedly rebuilding. Mount the root partition, inspect filesystem and boot errors, and copy any remaining files to independent storage. If the disk is irreparable, restore the provider snapshot or create a replacement server from the last known-good image.
SSH is blocked after hardening
Open the provider console, check the SSH service status and configuration syntax, confirm the administrative key and file permissions, and verify that the host and provider firewalls allow the port from your source address. Keep one console session open while changing remote access settings.
PC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11Crashes, No Sound, or Screen Glitches?
Random freezes, missing sound and display glitches usually trace back to one bad driver. Find and replace yours safely.Free scan · under a minuteThe application starts but behaves incorrectly
Compare environment variables, file ownership, runtime versions, database extensions, systemd dependencies, TLS paths, and outbound firewall rules with the inventory. Restore one component at a time and inspect service logs instead of copying an entire old configuration directory.
DNS still reaches the old server
Check authoritative records, resolver caches, split-horizon DNS, reverse proxies, and hard-coded IP allowlists. If you built a replacement with a different address, keep the old server online until external resolvers and integrations have converged.
Quick Recap
Operational checklist
- Objective identified: clean rebuild, newer release, or parallel replacement.
- Ubuntu release, IPs, DNS, users, keys, services, volumes, jobs, and secrets inventoried externally.
- Database exports and file copies completed, stored independently, and tested.
- Provider snapshot or backup completed; recovery credentials and console access confirmed.
- Rebuild versus replacement decision recorded, including IP and cost implications.
- Ubuntu image and architecture selected; destructive confirmation reviewed.
- New SSH host fingerprint verified through a trusted path.
- Updates, non-root sudo access, key-only SSH, firewall, hostname, time zone, and monitoring configured.
- Application, database, TLS, mail, jobs, backups, DNS, and resource limits validated.
- Old server retained until cutover, backup verification, and rollback window are complete.
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
