New Proxmox Host - Cant locate old VMs

[MarcMorgan]

Hi All

I may be daft on this or a little inexperience and hoping for you to guide me:

I have Proxmox installed on a host, i installed Proxmox backup and backed up my VMs (snapshot) to a external USB

I then installed proxmox on a new hardware (i cant seem to find a way to insert the USB and allow the new Proxmox server to find/see these backs so i can import them.

I thought ok I will use the existing hardware that previous had Proxmox installed (had to reinstall on new SSD) then re-import them again either via the above method or via the below setup:

Server: Dell R710
Internal SAS drive that was used for VM-Storage
I know this SAD drive had not been formatted in anyway
The new Proxmox install sees the disks as /dev/sda (sda1 - partition/sda2 - ZFS)

I just cant see to find a way to eithe rimport from disk - allow this disk to be used as VM storage (LVM etc)

I clearly dont want to format/wipe the disk unless i can re-load from the external USB backup.

Please can someone help

I also for some reason on new hardware get an error after i created a LVM/VG for VM-Storage - when i create a VM and come to the 'finish' part where it goes to create it, error about

LVM2 broken - signature - i have tried to fdisk the drive and wipe it but doesn't seem to do anything, so i cant even create new VMs on my new hardware because of this error 'Aborted wiping of PMBR. 1 existing signature left on the device'​

 

[fabian]

if your internal disk was previously used with LVM or LVM-thin (and it really hasn't been formatted/wiped/.. in the meantime) you should just need to add the storage entry again (possible via the GUI: Datacenter -> Storage -> Add ...)

the failure to create new disks is a bug, if you upgrade to current PVE packages it is fixed.

 

[MarcMorgan]

So i have imported like you said but i cant see the VM's under the VM Disk or CT volumes

In summary it shows disk space used but unable to see anything

 

[fabian]

please provide
pveversion -v
cat /etc/pve/storage.cfg
pvs
vgs
lvs

 

[MarcMorgan]

 

[fabian]

there don't seem to be any guest volumes on either volume group..

 

[MarcMorgan]

Yeah but i know there is thats whats confusing about it

 

[MarcMorgan]

So looking further into this i noticed it had pulled from my newly created SSD for proxmox and not from the drive/disk they are stored, so i need to import/load /dev/sda (which is where all of them sit)

If i go to storage/add it doesnt show within the list.

 

[fabian]

that is not LVM but ZFS.. you need to check the output of zpool import, then you can add one or more ZFS storages using the GUI (like you did for the LVM ones).

 

[MarcMorgan]

if your internal disk was previously used with LVM or LVM-thin (and it really hasn't been formatted/wiped/.. in the meantime) you should just need to add the storage entry again (possible via the GUI: Datacenter -> Storage -> Add ...)

the failure to create new disks is a bug, if you upgrade to current PVE packages it is fixed.

I updated PVE and still got same message when trying to create a vm

Error:

WARNING: gpt signature detected on /dev/VM-Storage-Disks/vm-100-disk-0 at offset 512. Wipe it? [y/n]: [n]
Aborted wiping of gpt.
1 existing signature left on the device.
Failed to wipe signatures on logical volume VM-Storage-Disks/vm-100-disk-0.
TASK ERROR: unable to create VM 100 - lvcreate 'VM-Storage-Disks/vm-100-disk-0' error: Aborting. Failed to wipe start of new LV

 

[MarcMorgan]

that is not LVM but ZFS.. you need to check the output of zpool import, then you can add one or more ZFS storages using the GUI (like you did for the LVM ones).

The zpool import shows only my pools that were created when i had a VM - TrueNAS - I need these anyway but i am also missing the VMs

 

[fabian]

I updated PVE and still got same message when trying to create a vm

Error:

WARNING: gpt signature detected on /dev/VM-Storage-Disks/vm-100-disk-0 at offset 512. Wipe it? [y/n]: [n]
Aborted wiping of gpt.
1 existing signature left on the device.
Failed to wipe signatures on logical volume VM-Storage-Disks/vm-100-disk-0.
TASK ERROR: unable to create VM 100 - lvcreate 'VM-Storage-Disks/vm-100-disk-0' error: Aborting. Failed to wipe start of new LV

those are now again completely different storages? if you don't give me consistent information it is kind of hard to understand what's going on and help...

please post the following:

• pveversion -v
• cat /etc/pve/storage.cfg

of the node where this is happening..

 

[fabian]

The zpool import shows only my pools that were created when i had a VM - TrueNAS - I need these anyway but i am also missing the VMs

please post the output of zpool import and lsblk

 

[MarcMorgan]

those are now again completely different storages? if you don't give me consistent information it is kind of hard to understand what's going on and help...

please post the following:

• pveversion -v
• cat /etc/pve/storage.cfg

of the node where this is happening..

OK let me explain

I have 2 Dell R710 servers

1 high spec - this is going to be running my new system/VMs everything - New SSD for Host and new SSD (1TB) for VM storage
Created a VG with the 1TB but when i create VM i get this error

the other server is my old server (internal Sas drives - I used one of these sas drives for all my vms and had truenas installed as a vm (inside this truenas vm i used for disk storage using other internal sas drives available on this server

Basicallly all i am trying to do it bring the VMs across from old to new (i have a external USB of these backups) - dont know how to import
If i can import these VMs from my old server to my new one im happy

Issues i, having on the new server is i cant even create a VM because im getting this signature error and i really dont understand why (fresh install on fresh disk)

Cant load VMs on old server (after reinstall on proxmox on new ssd)

Even if i have to lose my Vms from the old server - so be it

I just need to be able to create new VMs on my new one but getting signature errors



Thanos is new server - Loki is old server

 

[fabian]

your THANOS host ist not upgraded to current packages.. the fix is in libpve-storage-perl 7.0.8-1, which is available on all repos. you still haven't given the output of zpool import on LOKI

 

[MarcMorgan]

I have managed to import 2 of them but no way to access them via windows machine yet

I have ran apt-get update and upgrade but showing nothing to be upgraded so how can i manually download/install this libpve-storage-perl 7.0.8-1,

I am sorry about these questions, when i did this forst time round had no issues at all, and i thought a new system/move everything across would be easy :/

Happy for a remote connection if you like bud

 

[MarcMorgan]

FYI - Updated pvelib and this has worked )

 

[MarcMorgan]

So i guess the next question if so kinda could be asked is

Is there anyway i could:

Move my zpool to my other (new server) - I have just built a TrueNAS vm so i guess i would need to move these across from old to new sever OR can i share mount access

Keep them on my old system and allow access to link to my new nas?

 

[fabian]

you probably want to move them over - but that is general linux administration and not really PVE specific