Recent content by Alessandro 123

I'm using ZFS in both. Yes, backup/restore could be good, but the backup feature only backups to a local storage and i don't have free space to hold a full backup

I have some servers planned for shutdown. Meanwhile, i have to migrate some VMs from pve-manager/6.4-4/337d6701 TO pve-manager/5.2-2/b1d1c7f4 They are not in cluster (but are on the same network). I can shutdown the VM. Which is the best way to move from these 2 servers ?

Hi, I need to export some VMs (qemu with zfs) to Google Compute Engine. Do you know if this is possible in some way or should I move the VM content to a brand new created on GCE ?

Do you have any log in XenServer ? I did not less than 150 exports with no issue with both version (there is a new one that does the export directly to the ZFS volume)

Already saw that, but is talkinig about qcow disk images, in the question i've wrote ZFS storage for a reason........ i'm not using qcow but zfs volumes

As subject, any idea how to export a running VM with ZFS storage to an OVA/OVF file ?

No, wait, I need a direct export , not a mount to use when backupping and restoring. It's better, in my environment, risk to have to restart everything from scratch than the downtime needed to backup and then restore in 2 steps.

No shared storage.

Thank you. Can I do the same PULLING the backup from a remote server instead of pushing it ?

Is possible to do this in one step ? Something like newhost$ vmbackup oldhost 111 | vmrestore Just to avoid to backup hundreds of GB locally and the restore hundreds of GB to the new host (a lot of hours needed as this is a 2 step operation via network)

Which is the best method (I can poweroff the VM) to migrate a VM from 2 standalone (not in a cluster) PVE hosts ? Both hosts are on the same network. There is also a way to livemigrate ?

I'm experiencing a strange issue. A working machine , suddenly loose the LVM metadata (on random disks, not always the same). In example: I properly boot the machine, I use it, then reboot and 1 LVM PV metadata is lost. I have to recreate the metadata on disk recovering the LVM configuration...

well, adding 2 VMs more, means a total RAM usage of 52GB on a 64GB host. There are 12GB free for the host.

Should I reduce the ZFS ARC cache manually or it does it on its own ? In other words, can I add 2 VM with 16GB each ?

Hi guys, I have a PVE node with these VM running on it: So, there are 2+2+1+16GB allocated. Total 21GB But why here I can see 51.96GB used ?