This is using RedHat Linux Enterprise server 7.x. The server hardware is HP Proliant DL380 G7. We are thinking of cloning the whole server to another new server hardware. So far, not backup software is installed in this server. What would be the workable way to clone it? can we use rsync? or any proven commercial software.
Appreciate for any help.
LinuxServer Hardware
Last Comment
MichaelBalack
8/22/2022 - Mon
arnold
dump|restore is one way
You would need to first partition the new drives and then use dump | restore to "clone" the data from one to the other
dd is a low level media copy takes longer
and will need to be used to set the boot marker ....... on the new server...
The other option, is you can use rpm -qa and then add these packages to the new server.
Clonning while faster, could replicate/copy an existing issue yet unseen to the next one.
Do you have a kickstart installer setup in your environment where you can use it to build instances of servers that you need to deploy?///
Do you need the server you are cloning to be up an operational, or can it be unreachable/unusable during the process?
Dr. Klahn
Is the system drive a single drive, or is it a RAID set?
MichaelBalack
ASKER
Hi Dr. Klahn,
I am new to the server, I have to check. I let you know later.
If you are moving servers anyway, I would virtualize it with VMWare. You can use the free edition on the new server, then do a clone via CD or USB boot to copy it over.
Having the server on a recognized Virtual platform will make life so much easier in the future with snapshots and easy backups.
Vmware Free Edition on the new server.
Create a "Standalone Converter" disk for vmware.
Boot to it.
Point the software to the new server.
It will clone!
Then you can use things like VEEAM free to backup the server as an image if needed for fast recovery.
I got the problem to access to the given linux server as root's password changed. I will run those commands once I can access.
arnold
if you have sudoers configured for your user, and you have a login, login and run sudo -s provide your password and you should have elevated rights... similar to root.
Access to the system is required and booting from a separate BOOT media can help regain access provided you take note on how the disks/partitions are arranged using df -k which does not require root level access pay attention where / is as you would need to access /etc/passwd /etc/shadow to update root's password or /etc/sudoers to configure your user with elevation of rights access...
rpm -qa | grep -i sudo ......
root might not be allowed remote logins versus you not having the password.
su -
root's password if you have it should work/..
You have lvm overlay, you can use dump to create a full backup, and use restore to restore.
How is the RAID on the G7 configured?
If RAid 1 or 10,
You could use one of each mirrored pair, as the source for the new server make sure the new server is not connected to the network.
Pulling on of the existing mirrored raid Pairs and using it to boot the new server. While replacing the pulled drive when the system is running to allow for the raid rebuild.....
Note you Sid you have redhat 7, but your redhat release says you have redhat 6..
MichaelBalack
ASKER
Hi Arnold,
Is using RAID 1, since 2 pieces of 300 GB SAS HardDisks
Sorry, my mistake. It is Redhat Enterprise server 6.6
MichaelBalack
ASKER
Hi Arnold,
Forgot to share with you, the new server is located in different country, so, pulling the hdd method is not possible.
You would need to first partition the new drives and then use dump | restore to "clone" the data from one to the other
dd is a low level media copy takes longer
and will need to be used to set the boot marker ....... on the new server...
The other option, is you can use rpm -qa and then add these packages to the new server.
Clonning while faster, could replicate/copy an existing issue yet unseen to the next one.
Do you have a kickstart installer setup in your environment where you can use it to build instances of servers that you need to deploy?///
Do you need the server you are cloning to be up an operational, or can it be unreachable/unusable during the process?