Hi, this is a case I also thinking about. It's like an icebox... the easiest way is to deploy cheap primary storage with specific tags, create small compute node in separate dedicated cluster for icebox account. Next, upon archiving transfer the __stopped__ vm to that account snd migrate volumes to dedicated storage.
Stopped VMs don't utilize compute resources, so you can archive a lot of VMs simple. In theory it should work, but has to be tested, maybe limitations exist. чт, 31 янв. 2019 г., 22:15 Cloud List cloud-l...@sg.or.id: > Hi, > > I am using CloudStack with KVM hypervisor. > > I noted that a volume snapshot will be deleted when the volume (and > instance using the volume) is deleted. Is it possible to keep the snapshot > saved somewhere when the volume and instance are deleted? > > My goal is to have a backup copy of the data before an instance (and its > volumes) is deleted, and keep it on the secondary storage (or another > external storage if possible, but not the primary storage) for a certain > period of time. > > So far the solution I can see would be: > 1. To export the snapshot into another volume (which will take up space on > the primary storage). > 2. To save the snapshot as template (which is not practical especially for > data-disk). > > Is there any other solution? Any advice is greatly appreciated. > > Thank you. > > -ip- >
