New Config questions

L

Lyndhurst PS

New Member
May 15, 2025
6
0
1
Hi all,

I am new to proxmox but not to qemu, I have been using this command line version for at least a decade. I love the user interface and want to migrate to this on my new server. The way i currently have it setup is on a debian install I use zfs filesystem. I need to be able to use snapshots or backup but exclude certain volumes. Here is why, I run on prem exchange and we have a massive 8TB volume for our databases. Currently the way snapshots work is that each drive can be restored individually from a snapshot, with proxmox, i dont see a way to exclude that big volume if needed. I would in most instances never have to restore my databases it would always be the OS from a bad update or some kind of attack. Any thoughts?
 
Yes i read that post, here is another question, is it possible to use a drive file without importing it into the machine?
 
Lyndhurst PS said:
Yes i read that post, here is another question, is it possible to use a drive file without importing it into the machine?
Click to expand...
I am not quite sure what you mean. It's unlikely that you can achieve that with local storage.
Here is another post that may be helpful:
fabian

Post in thread 'Passthrough disks and snapshots'

first, only containers do a storage snapshot when doing a 'snapshot' mode backup, VMs use a qemu-internal mechanism instead that doesn't require storage support.

second - the main reason why "skip volume when doing a snapshot" is not available as a feature is that handling of all sorts of related features quickly becomes tricky. for example rollback - is the expectation to leave the disk as it is? or to re-create it empty (like when restoring a backup with excluded mountpoints/volumes)? should it be configurable? what if the flag changed since the rollback target snapshot was taken (in...
  • Like

Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox
 
One more question, when using zfs filesystem, is there a way to see the actual disk image file or extract it somehow if i was to go old school and use zfs auto snapshot from cli? I know that this works using block write versus just a file storage so where and how would i find or extract the actual disk image? Is there an example anywhere?
 
Lyndhurst PS said:
One more question, when using zfs filesystem, is there a way to see the actual disk image file
Click to expand...
Technically everything in Linux is a file, but most would not use that term here. With PVE when ZFS is used for VM storage, you are using ZFS zvols which are raw slices. You can place any filesystem on top of them, from ext2 to ntfs.
The Zvols can be managed by regular ZFS toolset, i.e. : zfs list

You can use native ZFS tools to take and remove snapshots, but keep in mind that there will be no automatic synchronization with VM state.

Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox
 
What im asking and I think this is my ignorance now with how the file system works, is when i make a manual snapshot through cli using zfs snapshot command, when i go into the .zfs folder, there is nothing in there. What im looking to do is essentially like run a hybrid i guess where if i need to roll back my OS drive to earlier in the day from a snapshot, to be able to do that without trying to roll back my big storage drive thats multiple TBs of data. The reasoning for this is that usually the data drive is not an issue. An example would be a failed update or in the instance of my EDR software causing a complete lockup and needing to roll back to earlier in the day. I dont mind if my snaps dont sync with the gui, im ok with downing my VM instance, extracting the OS image from zfs snapshot that i made using cron etc. then reimporting that to my vm and reupping it. Sorry for the convoluted mess this post turned into LOL. Thank you for bearing with me here. Ive been doing this through CLI for over a decade now and I really want to use this front end for it.
 
I’m not a ZFS expert, there are others here who work with it daily and can provide more detailed insights.

That said, if you can share a full, end-to-end example of the commands you’re running and the actual output, it’ll be easier for others to help. If you go this route, please make sure to format your post using text and CODE tags for readability.

Also, keep in mind that the .zfs directory is hidden by default. Have you tried accessing it using the full path, like:
ls /rpool/data/vm-100-disk-0/.zfs/snapshot/?

In general, the best approach is to test things directly. You can set up a PVE instance with a ZFS pool in a virtual environment and experiment freely - without risking your production setup.

Good luck!

Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox
 
You must log in or register to reply here.
Top Bottom