What storage type for vm backups in snapshot mode?

Martini2

Member
Jan 28, 2022
18
0
6
Hello,

what storage type do i have to create for vm backups in snapshot mode?
Right now i make backups to a directory type storage with content type set to 'VZDump backup file', but 'snapshot' backups don't work. Proxmox complains 'mode failure - some volumes do not support snapshots' and i only can do stopped or suspended backups.
When i try to make a zfs storage i can only select image and container for the content type. So it can't be a backup location.
 
I dont think it matters what your backup target storage is. The problem is with you source/production volume storage.
Anything that's listed here https://pve.proxmox.com/wiki/Storage with Snapshot=yes should provide ability to backup via snapshot method.

It may be useful for you to post vm config and storage config file, so community can help


Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox
 
the type of the volume storage 'zpool2_cvdata' is zfspool
Code:
storage.cfg
dir: local
        path /var/lib/vz
        content iso,vztmpl
        prune-backups keep-last=2,keep-monthly=11
        shared 0

dir: zpool3_backups
        path /zpool3/backups
        content backup
        mkdir 0
        prune-backups keep-last=8,keep-monthly=11,keep-yearly=3
        shared 0

zfspool: zpool2_cvdata
        pool zpool2
        content rootdir,images
        mountpoint /zpool2
        sparse 1

the config of the container
Code:
arch: amd64
cores: 2
features: mount=nfs;cifs,nesting=1
hostname: mediaplayer
memory: 4096
mp0: /zpool1/shares/media,mp=/mnt/media,backup=0
nameserver: 192.168.178.1
net0: name=eth0,bridge=vmbr0,firewall=1,hwaddr=01:02:03:04:05:06,ip=dhcp,type=veth
onboot: 1
ostype: debian
rootfs: zpool2_cvdata:subvol-203-disk-0,size=32G
swap: 2048
unprivileged: 1
lxc.cgroup2.devices.allow: c 226:0 rwm
lxc.cgroup2.devices.allow: c 226:128 rwm
lxc.cgroup2.devices.allow: c 29:0 rwm
lxc.mount.entry: /dev/dri/renderD128 dev/dri/renderD128 none bind,optional,create=file
lxc.mount.entry: /dev/dri/card0 dev/dri/card0 none bind,optional,create=file
lxc.cgroup2.devices.allow: c 10:200 rwm
lxc.mount.entry: /dev/net dev/net none bind,create=dir

Backups in snapshot mode worked on this system until 2021.12.03 after that proxmox wirtes 'mode failure - some volumes do not support snapshots' in the task viewer.
 
Something changed. You have mp0 on zpool1 that is not in the storage list. I know it's set to backup=0, but snapshots are normally done on entire VM/CT. There is no special "backup snapshot" that would exclude backup=0 filesystems.

I recommend you review your storage configuration and make sure everything lines up properly.


Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox
 
i have added backup=0 while searching for a solution yesterday, but it makes no difference. The mp0 mount existed since the creation of the container years ago and worked until december. The only thing that change are the proxmox updates. The storage is untouched for years. Sorry, i don't understand what you mean with 'lines up properly'.
 
You can store backups on a ZFS pool. Create a dataset like "YourPool/VMs" for your VMs and add that dataset as a "ZFS" storage to store your virtual disks on it. Then create another dataset (for example "YourPool/Backups") and then add a "directory" Storage to your PVE pointing to the mountpoint of that dataset.
 
this is how i have it set up. The vms and cts are on pool1 in the dataset cvdata which is a zfs storage in proxmox. the backups are on pool3 in the dataset backups that is a directory storage. I have periodical backup jobs configured to make backups in snapshot mode. This setup is years old und and worked until last december. Now the backups of the cts throw an error and fail to work in snapshot or suspended mode. I can do the backups in stopped mode but this is not what i want. As this worked the last years i don't think i am affected by the bug mentioned by bbgeek17. Unfortunately the old syslogs are gone and i can not find out what happend.
 
Create a basic CT, minimal config - just rootfs. Try to back it up with snapshot mode. If it works, start adding other stuff you have in other CTs. See where it breaks.
Keep an eye on journalctl -f - I would expect something useful to be there.
 
You can store backups on a ZFS pool. Create a dataset like "YourPool/VMs" for your VMs and add that dataset as a "ZFS" storage to store your virtual disks on it. Then create another dataset (for example "YourPool/Backups") and then add a "directory" Storage to your PVE pointing to the mountpoint of that dataset.

Just fighting with the same issue. I also have the lxc stored on 'local', which is a directory type storage and does not support snapshots. If I understood you correctly, I should create a new dataset for lxcs/vms and another for backups. To do so, I need to:

1. Create the pools (are these commands ok?):

Code:
zfs create rpool/VMs
zfs create rpool/backups

2. Add a directory from the Datacenter -- Storage tab.

3. Move the LXC to the new storage? How should I do that?

Thanks
 

About

The Proxmox community has been around for many years and offers help and support for Proxmox VE, Proxmox Backup Server, and Proxmox Mail Gateway.
We think our community is one of the best thanks to people like you!

Get your subscription!

The Proxmox team works very hard to make sure you are running the best software and getting stable updates and security enhancements, as well as quick enterprise support. Tens of thousands of happy customers have a Proxmox subscription. Get yours easily in our online shop.

Buy now!