DR backup or copycluster script fails


Article ID: 5396


Updated On:


CA Infrastructure Management CA Infrastructure Management CA Performance Management - Usage and Administration DX NetOps


Errors are seen trying to run the Vertica backup.

Unable to run local backup for Vertica using vbr.py script.

Unable to run remote backup for Vertica using vbr.py script.

We are trying to set up a Vertica backup using vbr.py.

We are trying to set up a Vertica copycluster using vbr.py.

We have validated that the dradmin user has passwordless ssh enabled and working.

The backups fail with this message:

[[email protected] backup]$ /opt/vertica/bin/vbr.py --task backup --config-file /home/dradmin/backup/backup_snapshot.ini
Error: Errors connecting to remote hosts: <IP>.  Check SSH settings, and that the same Vertica version is installed on all nodes.
Backup FAILED.


The ssh configuration in the sshd_config file is set with No for AllowTcpForwarding.


All supported DX NetOps Performance Management releases


Vertica requires AllowTcpForwarding set to Yes in the /etc/ssh/sshd_config.

To resolve the error edit the /etc/ssh/sshd_config setting AllowTcpForwarding from No to Yes. Save the changes and restart the ssh service for it to read in the new configuration.

To restart it for RH 7.x releases we'd run "systemctl restart ssh".

Now the backup command using vbr.py should succeed.

Additional Information

Vertica Configuring Backup Hosts documentation

Notes/Tips for SSH configuration:

  • AllowTcpForwarding is commonly commented out in the sshd_config file by default. If that is seen ensure the comment (# symbol) is removed from the start of the line.
  • To further restrict to tcp forwarding to just one user you can add the following line to the sshd_config file.
    • Match User dradmin
  • The ssh service MUST be restarted before it will recognize changes to the sshd_config file settings.
  • RH 6.x releases would restart the service with "service ssh restart".