**Sorry, urgent** What happens (to disks) at the end of a bare metal restore process?

[pkelly_sts]

I'm in a situation where we have a physical server that's dying (suspect storage issue) and we're desperately trying to get it restored to a VM before the original gives up (SBS2011 server that we've recently inherited, so major single point of failure) and whilst using the kit available to us on a new client site which is VERY basic right now.

We've tried a few approaches which I won't go into here but at the moment we're currently running a bare metal recovery from a basic Synology BUT the source is a backup copy (from the original on another synology which had many restore points) and this backup copy has just a single restore point in it so from a READ perspective it /should/ be very efficient, and indeed the restore job started off very quickly as I expected to.

However it's slowed down significantly over time, from 120MB/s starting out on the 884Gb C: drive to currently running at 25MB/s 94Gb into a 976Gb G: drive, and gradually slowing down. I can see that the source synology is NOT strained whatsoever, throughput, IOPs or utilisation. I will openly admit that the source synology also only has a single physical disk in it but I reiterate, it only contains a single RP so should have no problem providing max throughput (Synology DS218+ that we threw a drive into to try to give us somewhere to send a backup copy job whilst we were trying other things).

I was wondering, given the original physical server has a C: 884Gb, D: 100Mb, G: 976Gb, ALL on a single logical disk (3-disk array I believe) and C: & D: have already restored via bare-metal. **IF** I were to cancel the restore now, reboot the VM, start again but ONLY select the G: volume for restore, do I stand a good chance of ending up with a bootable system?

If the restore process does some disk boot-sector reconfiguration etc. at the /end/ of the process, and that doesn't happen as part of the initially cancelled bare metal restore, then I can see how the VM could end up non-bootable.

Is there any mileage in my thinking?

[Dima P.]

Hello pkelly_sts,

All the post-restore activities (such as driver injection, registry update etc) are performed at the end of the restore process, so I'd definitely wait till the end of the Bare Metal Recovery. Cheers!

[pkelly_sts]

I thought that might be the case so I ended up sticking it out & noticed exactly that happening at the end but thanks for the confirmation.