Link to home
Create AccountLog in
Avatar of kswan_expert

asked on

VMware Data Recovery design and best practise


We currently use Vmware Data recovery 1.2 to back up VMs (20 or so) from a 3 host ESX 4.1 U1 cluster. The backups are dumped to a 500GB volume on a phsical server called NYDR001 and then backed up to tape and sent offsite.

We've just attached an HP P2000 SAN to our vSphere and want to start using this for dumping backups and replicating to another site (which also has a new P2000, identical setup). We'd also like to dump nightly backups of our physical servers to the P2000 as well.  

What's the best way to do this? We were thinking of P2V’ing NYDR001 and running on the P2000 with a 500GB volume for the VDR VMs and 2  more 500GB images for the physical server images.  We’d build another VM on the other site and set up DFS between the VMs to replicate the changes to the 3 volumes containing all the backed up VMs\servers.

Also, saying we have replicated the volumes containing the VM backups to another site, can I restore to another vsphere cluster? Anyone done this?

Is this the best way to do this. Completely open to suggestions on how best we should do this. Maybe we need to look at another product?

Avatar of Andrew Hancock (VMware vExpert PRO / EE Fellow/British Beekeeper)
Andrew Hancock (VMware vExpert PRO / EE Fellow/British Beekeeper)
Flag of United Kingdom of Great Britain and Northern Ireland image

Link to home
Create an account to see this answer
Signing up is free. No credit card required.
Create Account
doing a v2v will not break the old vm usually but its much better to shutdown the vm and back up the vm and restore on the other host
Avatar of kswan_expert


Thanks guys.

We are using CIFs on a physical box. So what we should do is create a 500gb disk for the VDR appliance and back up the vms to here. How do we then back this up to tape, and\or replicate offsite?

Our physical servers we’d be best to look at separately?

thats really the difficulty if you use a SAN based VMDK for backup, because it does not lend itself well to back to tape.

What would you do with the backup you have on tape? If it's as an archive for safekeeping purposes, in case the main repository was damaged or destroyed. Its not as easy as it would suggest to restore from tape and use the vDR. Okay the restore from tape would be easy, but remounting the repostiry and using is awkward at best.

Have you considered two repositorys, with alternative backups to both, and dropping the tape solution?
Yeah I like the peace of mind that tape gives. Is hard to beat.
If you have 2 repositorys i guess they can't be on the same SAN, so what do people do then?
Maybe we need to look at something like VEAM and replicating offisite. Can you send VEAM backups to tape or have people completely done away with backing up to tape. What happens if your backup store gets corrupted some how and is replicated in both sites, how do you go back to clean restore points?

Sorry for all the questions, really appreciate the relpies.
The  2 repositorys can be anywhere, SAN (vmdk based) or Windows Share, or NFS (vmdk).

Yes, people have moved away from the dark ages, we got rid of tape based backups, in 2004, and so did most of our clients, due to cheap storage (but now storage prices are on the increase due to HD shortage, because of Thailand floody, and WD plants!).

If you backup store is corrupted at both sites, your are stuffed.

Veeam Backup and Replication, and replicate VMs to another site.