Did you check the hierarchy of these files with 50KB?
I believe that when ACS takes a snapshots it will create a hierarchy of VHDs, and since you are looking for the leaf right after the snapshot, it is for sure empty (or almost empty)
On 10/28/2017 9:24 AM, Vivek Kumar wrote:
Hello Team, I am using cloudstack 4.7.1 with xenserver 6.5, Recently I come up with a strange issue. I had a VM with four data disk so we planned to scale it up, I shutdown my VM and changed the service offering and before doing that we had taken snapshot, after all this I just turned on my VM and it gave me error that VMs cant be up due to insufficient capacity. So I have checked the logs but couldn't get any clue. I see the path of these volumes from my DB and checked it in my storage and I found all these VHD are having no data earlier it used to be like 100GB, 25GB and like this , but now size of disks are around 50KB for all VHDs. So how come it deleted all old VHDs and replaced it with new VHd, I checked on my xenserver about these VHDs so that I can map the correct VHD in cloud stack. We found the relevant VHDs and copied it to another location so that we can revcover data in case of any issue. After some time I noticed that it deleted all old VHDs as well - I Dont understand why it got deleted even if I manually create VHD it is also getting deleted. So we tried it import all VHDs in a xenserver and found no data inside these VHDs. We were using LVM with all these four disk, basically I mounted 3 data disk and create a volume group and then merged it into a root file system. I agree that its not a good practice because data disk should not depend on OS disk. Can someone please help me to figure out the reason behind it. Vivek Kumar
-- Rafael Weingärtner
