How to Copy Data Between Two Linux Servers Using Root Access (With Resume Support)

Cloud Services Support
, ,
1

ICTN800
ICTN8001536×1024 405 KB

How to Copy Data Between Two Linux Servers Using Root Access (With Resume Support)

When migrating data between Linux servers, unstable internet connections or SSH drops can interrupt long transfers. A professional migration strategy must support automatic resume without re-copying data.

This guide explains how to securely copy data from Server A to Server B using root access over SSH, while masking sensitive details and ensuring transfers resume seamlessly if connectivity is lost.

Scenario Overview

  • Source server (Server A) Directory to copy:

    /home/USERNAME/PROJECT_DATA

  • Destination server (Server B) Target directory:

    /home

  • Access method: SSH (root)

  • Server IP address: Masked (e.g. XXX.XXX.XXX.XXX)

  • Connectivity: May be unstable (resume required)

Why rsync Is Ideal for Unstable Connections

Unlike scp or FTP, rsync is designed to continue exactly where it stopped. If the internet drops, SSH disconnects, or your terminal closes, you can simply rerun the same command.

rsync provides:

  • Delta-based transfers (copies only missing parts)
  • Partial file preservation
  • Resume support for large files
  • Verification of already transferred data

Step 1: Enable Root SSH Access on the Destination Server

On Server B, edit the SSH configuration:

nano /etc/ssh/sshd_config

Ensure root login is temporarily enabled:

PermitRootLogin yes

Restart SSH:

systemctl restart sshd

:locked_with_key: Root login should only be enabled during migration.

Step 2: Start the Transfer With Resume Support (Run on Server A)

Use the following resume-safe rsync command:

rsync -avzP --partial --inplace \
/home/USERNAME/PROJECT_DATA \
root@XXX.XXX.XXX.XXX:/home/

Important flags explained:

  • -a preserves permissions, ownership, timestamps
  • -v verbose output
  • -z compresses data
  • -P shows progress and enables partial transfers
  • --partial keeps partially transferred files
  • --inplace writes directly to destination files (faster resume)

:white_check_mark: If the internet disconnects, rerun the same command and rsync will continue automatically.

Step 3: Protect Long Transfers With screen or tmux (Recommended)

To avoid losing progress when your SSH session closes:

screen -S migration

Run rsync inside the screen session, then detach:

Ctrl + A then D

Reconnect anytime:

screen -r migration

Step 4: Verify the Transfer After Completion

On Server B:

du -sh /home/PROJECT_DATA

On Server A:

du -sh /home/USERNAME/PROJECT_DATA

Matching sizes confirm a successful and complete transfer.

Step 5: Secure the Destination Server After Migration

Once the data is confirmed, disable root SSH login again:

nano /etc/ssh/sshd_config

Set one of the following:

PermitRootLogin prohibit-password

or

PermitRootLogin no

Restart SSH:

systemctl restart sshd

Best Practices for Reliable Server Migrations

  • Always use rsync instead of scp for large data
  • Use --partial and --inplace for resume safety
  • Wrap long jobs in screen or tmux
  • Verify disk space before starting
  • Mask IPs, usernames, and paths when publishing guides
  • Keep backups before migration

Conclusion

A reliable server migration must assume network interruptions will happen. By combining rsync’s resume capabilities with SSH root access and proper security practices, you ensure data moves safely, efficiently, and without restarting from zero.

This method is trusted across VPS, shared hosting, and enterprise Linux environments.