- Ensure both cluster names are different when created; otherwise, it will create problems.
- Ensure the Ceph storage name is the SAME on both clusters, and that both are configured for either replication or erasure coding.
echo "starceph" > /etc/webmin/vdcsitemanager/ceph-storage-name
echo "0" > /etc/webmin/vdcsitemanager/ceph-erasure-code
echo "cephdata" > /etc/webmin/vdcsitemanager/ceph-storage-name
echo "1" > /etc/webmin/vdcsitemanager/ceph-erasure-code
- Set up one-time SSH auto-login to all Site1 PVE (DC)and Site2 PVE (DR) for automatic login between them and from the VDC VM. This is essential for inter-process communication and to trigger root processes for data synchronization.
ssh-copy-id [email protected]
- Test the SSH connection:
It should log in automatically.
- Login from all PVE servers of Site1 Cluster to each server of Site2 Cluster PVE.
- Login from all PVE servers of Site2 Cluster to each server of Site1 Cluster PVE.
- Login from VDC VM to each server of Site1 Cluster PVE.
- Login from VDC VM to each server of Site2 Cluster PVE.
ssh-copy-id root@site1
ssh-copy-id root@site2
- Navigate to the folder on the VDC VM:
cd /usr/local/src/vdcsitemanager-tools/manager-tools/
- Create the site configuration for VDC to manage all nodes via Webmin (this is a one-time activity).
php onetime-site-build-for-manager.php 1 192.168.30.113
php onetime-site-build-for-manager.php 2 192.168.40.144
- Now, log in to Webmin of VDC (example: 192.168.40.30):
- URL: https://192.168.40.30:833
- Username: manager
- Password: (found in) /usr/local/src/manager-vdcmanager-pass