I have a server with hundreads of vms and the snapshot is very useful to me, on incidents i can restore a vm to a point on time in no time, but i have many datastores where operational data is stored and all bkp is done on LTO tapes, but the restore process of clean data (without virus) take more than 8 hours each time and time is growing as daily data grow too.
I have a lot of storage servers here and looking to make snapshots of at least my last datstore state to allow fast recovery and cant find one.
I know there is the vm snapshot but there is datastore snapshot on esxi vmware too ?
-
1Datastore-level snaps would need to be performed in your storage layer. – EEAA May 13 '17 at 20:52
3 Answers
Snapshots aren't backups! You really need to follow 3-2-1 backup rule to ensure your data is safe. To accelerate your tape backup and recovery process I'd suggest to go from Disk-to-Tape to Disk-to-Disk-to-Tape: deploy some VTL offloading "cold" backups to either physical tape or AWS S3/Glacier.
![](../../users/profiles/201494.webp)
- 12,510
- 1
- 20
- 46
As BaronSamedi mentioned, to boost up the backup and recovery process you need to use Disk-to-Disk-to-Tape backup. For such case, Starwind VTL is the most suitable solution with an optional offload to Azure and AWS.
https://www.starwindsoftware.com/virtual-tape-library-feature
Also, there is a solution called AcloudA, it will act as a standard SATA or SAS HDD and fit into standard 3.5" drive bay basically presenting the cloud storage as internal memory to the server.
![](../../users/profiles/201494.webp)
- 12,510
- 1
- 20
- 46
![](../../users/profiles/199697.webp)
- 5,181
- 13
- 11
As EEAA correctly stated, you have to do this one level below, on the storage layer. To be more precise, you have to do both - snapshot the VM on ESXi and snapshot the storage on your storage layer:
- Snapshot your VM with the builtin snapshot utility of ESXi.
- Snapshot your storage where your VM and its snapshots reside (typically, the backing datastore that is exported via NFS or iSCSI and used by ESXi to present a VM datastore). Of course, those snapshots have to be atomic and consistent, so you need a file system that supports those properties (for example ZFS).
- Delete/consolidate your ESXi VM snapshot so that it is removed and that it does not slowdown VM operation.
- Backup your storage snapshot to another server/medium in a rotation of your choice so that it works as a real backup.
Restore is the reverse process.
![](../../users/profiles/333167.webp)
- 2,452
- 12
- 31