Backup / Restore.... (DR)

Hello,

We have a backup/restore question.
Next week we will deploy a dedicated cluster of open-e SAN's with the Active-Active plugin and replication between them.
If the data would be corrupt, we want to be able to restore the whole boxes from a 3rd machine (disaster recovery).
We will have Xenserver ISCSI block mode device of 3-4TB that will hold all Xenserver images in a ext3 or lvm image.

I watched the backup presentation and the fact that restores can be done, but that is only to restore certain files (but will that work on Xenserver?).

How would we make reliable backups to:
1. be able to restore individual Xenserver vm's (data inside the Xenserver ISCSI volume)
2. be able to restore the complete open-e san servers from auxilary storage (iscsi).

Thanks in advance. Our fear is we loose all data in a fatal crash and are unable to restore. Or that it will be very hard to restore individual VM's when corrupted...

We have a backup server with Linux which we can build like a ISCSI target or f.e. rsync server. Would rsync be able to do the trick?

We could run rsync then f.e. every day or two days. And in case of disaster restore all content back?

Regards,

This would require a third party tool to replicate to the third site.

Hello,

All explenations about backup/restore are about NAS, just plain files on a filesystem and then having the data after a restore on a windows machine...

What we want: suppose an ISCSI volume used by Xenserver 6.1 got corrupted, the Ext3 filesystem is broken and now several VPS's broke down.

How (and i havent seen this in any presentation) can we restore a whole volume with all Ext3 data in it?

It makes no sense to have ext3 data available on a windows server share we want whole volumes to be able to be restored quickly.

Question is how?

So:
1 big volume group with all data in it...

Volume 1: 100GB (VPS's 1-5)
Volume 2: 200GB (VPS's 6-10)
Volume 3: 250GB (VPS's 11-15)

Volume 2 of 200GB gets corrupted and Xenserver cannot use VPS 6-11 any more and snapshots are no use since the volume is broken...

How can we within a short time, click and say:

Delete Volume 2, restore Volume 2, start the ISCSI target en connect Xenserver en run the VPS's again....

In this way we will have:
1. Xenserver Snaps and,
2. Complete volume backups.

If the VPS is broken we will have Xenserver snaps if the volume on open-e gets broken we have the backup/restore in place of all SCSCI volumes...

Thanks!