cloudstack-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Likitha Shetty (JIRA)" <>
Subject [jira] [Commented] (CLOUDSTACK-6224) VM Snapshot inconsistent size
Date Thu, 19 Jun 2014 10:41:24 GMT


Likitha Shetty commented on CLOUDSTACK-6224:

We compute the vm_snapshot_chain_size for each of the volumes that belong to a VM using method
getVMSnapshotChainSize(). The bug in this method is that we look for the file associated with
a volume in the entire datastore instead of looking for it inside in the VM folder in the
datastore. And since we don't match on the entire VMDK path in case of a ROOT volume we add
up the sizes of all the ROOT volumes present in the datastore and not just the ROOT volume
that belongs to the Virtual Machine.

Mike, could you pick this up please?

> VM Snapshot inconsistent size
> -----------------------------
>                 Key: CLOUDSTACK-6224
>                 URL:
>             Project: CloudStack
>          Issue Type: Bug
>      Security Level: Public(Anyone can view this level - this is the default.) 
>          Components: Hypervisor Controller, Management Server
>    Affects Versions: 4.2.1
>         Environment: Cloudstack 4.2.1
> XenServer 6.2
> Vcenter 5.5 ESXi
>            Reporter: Artjoms Petrovs
>            Assignee: Mike Tutkowski
>              Labels: snapshot, vmware, xen
> During the creation of VM Snapshot [VMware], resulting size is written in table „volumes”,
column  vm_snapshot_chain_size. It seems that size of a VM Snapshot is calculated manually
via the method getVMSnapshotChainSize(..)  and it gives overexpected result ( hundreds of
terabytes ) and is much larger than the filesize, that we can see in VMware itself. By calculating
the vmdk and vmsn file sizes manually. Xen VM snapshots [disks only] give similar results.
For me it seems that either the method works incorrectly, either it loops between some simlinks.

This message was sent by Atlassian JIRA

View raw message