Veeam Linux Instant recovery in vpshere, vm does not boot

[Hoguss]

Dear Veeam community.

I'am currently evaluating the Veeam backup and restor product.
( I have no case ID as i'm not able to open one )
My objectif is to be able to transfer AWS EC2 windows/linux instance into vpshere for legal/backup purpose.

For that I want to use Veeam capability.
I have done a full instance backup using veeam backup of an OEL 6 Linux EC2.
I have added connexion to my on premise vsphere and did a Instant recovery of this server based on my backup.

The vm was imported correctly into vpshere but my VM cannot boot. After some checks, it appears that my OEL 6 is using UEFI bios.
Or, when vm was created in vmware, BIOS was set.
I'm not able to change the boot option as the only value I have is BIOS.
I have tried to manually change the .VMS setting by adding firmware = efi but it did not work.

Can you please help ?
Thanks

[Hoguss]

After some checks, I was able to change VM type that was Other Linux to OEL 6 and was able to select UEFI.
Unfortunalty my vm is still unable to boot.

I have tried many things until now and yet i'm not able to figure out why it does not want to boot.

Can you please help ?

Thank you

[nitramd]

If your Linux VM was originally set up to boot from BIOS then you won't be able to boot via UEFI.

Booting via UEFI requires a different type of boot partition than BIOS. UEFI requires the boot partition to be formatted Fat32 if I remember correctly. Usually, a boot partition is formatted ext4 or xfs (in RHEL/CentOS cases) - the file system format depends on the distribution.

I think the only way to get your VM to boot is to use BIOS. If you can, check to see if the partition /boot is set as bootable.

If I'm interpreting your distro correctly you're using Oracle Enterprise Linux. You may want to contact Oracle for assistance.

[Hoguss]

Hello,

Thank you for your repply.
I guess my VM is intended to use UEFI since In the original EC2 instance, i have /boot/efi folder.
After some research, I saw that I may have to re-install grub in using a live-cd and troublshoot mode.
Because the disks are listed as /dev/xvdf or dev/xvda in amazon way, while the disk is listed as /dev/sda and /dev/sdb.

The procedure I followed is this one:
after booting as rescue mode with live cd.

cd /
mount -t ext3 /dev/sdb /mnt
mount -t proc proc /mnt/proc
mount -t sysfs sys /mnt/sys
mount -o bind /dev /mnt/dev
chroot /mnt /bin/bash

Then grub-install dev/sdb
But I have lots of errors.

Generally speaking, I would like to know if Veeam can be used for this kind of work ( exporting ec2 linux to on preme vsphere) and if any users could share with me their experience.
Because as far as I can see, it cannot work "out of the box" when performing migration from Cloud to on prem.

Thank you
Hugo

[Gostev]

Yes, most definitely it can be used for this kind of work. Moreover, what your described is the primary use case for this functionality, so it was thoroughly tested too. Thus, in cases when this does not work, you should open a support case and have them troubleshoot your issue - as it is likely a bug or limitation specific to VM configuration or Linux distribution used. Just be sure your configuration is not already listed as unsupported in the Release Notes, for example:

• Instant recovery of non-VMware Linux machines to VMware is not supported for backups of machines
with mkinitrd missing, or with mount points outside of /

Thanks!