[SOLVED] Snapshot overview

G

godfather007

Renowned Member
Proxmox Subscriber
Oct 17, 2008
96
6
73
48
Netherlands
Hi,

I'm moving away from vmware to proxmox, so i have build a TST-setup with CEPH to gather some info before doing this...

With VMware, it was advised to keep open snapshot at a minimum. The danger of snapshots is storage-corruption when they grow too big.

Is there an easy way to find "running snapshots" instead of verifying each vm?

Thanks
 
Last edited:
What storage are you using? There are ZFS, LVM, qcow2 and CEPH snapshots. So there are 4 completely independen snapshot engines.
 
  • Like
Reactions: godfather007
Hi,
I'm not really sure what you mean by "running snapshots".

In PVE, you can include the VM's state+RAM when taking a snapshot, so that you will be in the exact same state when rolling back. If you don't include the VM's state, you have to make sure that the file systems within the VM are in a consistent state at the time the snapshot is taken. If the QEMU guest agent is installed in the VM and enabled, this will be done using the fsfreeze/fsthaw commands automatically.

If you don't include the VM's state nor use the guest agent, then storage corruption within the VM's file systems can happen. I don't think there's much need to worry about corruption on the storage block layer for Ceph, it's widely used and robust.
 
Allright, sounds relaxing enough.

The question was based on an experience i've come across when using VMFS and "open/running" snapshots were forgotten and kept growing till the datastore became filled and beacme read-only. All VM's came to a stop at that point...
 
You'll still have to keep an eye out for storages getting full, and when you include RAM you'll need more space of course (use the guest agent method if RAM would take up too much space). In any case, Ceph will warn you when OSDs don't have much space left.
 
You must log in or register to reply here.
Top Bottom