[
https://issues.apache.org/jira/browse/CLOUDSTACK-9008?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14983072#comment-14983072
]
Mike Tutkowski commented on CLOUDSTACK-9008:
--------------------------------------------
More details with comments inline:
The VM's root disk in on SR 424db333-31b4-9390-84e9-59c9e2fd2a28. It is the
only VDI on the SR as can be seen below.
[root@XenServer-6 ~]# xe vdi-list sr-uuid=424db333-31b4-9390-84e9-59c9e2fd2a28
uuid ( RO) : 40dee945-2c4d-4227-b508-1adb5e7709c2
name-label ( RW): ROOT-50
name-description ( RW):
sr-uuid ( RO): 424db333-31b4-9390-84e9-59c9e2fd2a28
virtual-size ( RO): 21474836480
sharable ( RO): false
read-only ( RO): false
I take a VM snapshot. As can be seen below, there are now three VDIs that make
up the overall root disk (as expected).
[root@XenServer-6 ~]# xe vdi-list sr-uuid=424db333-31b4-9390-84e9-59c9e2fd2a28
uuid ( RO) : b9a80c64-1dab-4d5f-8d8e-9fb53f8f08ac
name-label ( RW): base copy
name-description ( RW):
sr-uuid ( RO): 424db333-31b4-9390-84e9-59c9e2fd2a28
virtual-size ( RO): 21474836480
sharable ( RO): false
read-only ( RO): true
uuid ( RO) : 40dee945-2c4d-4227-b508-1adb5e7709c2
name-label ( RW): ROOT-50
name-description ( RW):
sr-uuid ( RO): 424db333-31b4-9390-84e9-59c9e2fd2a28
virtual-size ( RO): 21474836480
sharable ( RO): false
read-only ( RO): false
uuid ( RO) : ae7b04b7-7f23-40d3-80df-95cab4bda602
name-label ( RW): ROOT-50
name-description ( RW):
sr-uuid ( RO): 424db333-31b4-9390-84e9-59c9e2fd2a28
virtual-size ( RO): 21474836480
sharable ( RO): false
read-only ( RO): false
I revert the VM to its one and only VM snapshot. Data corruption occurs (only
the snapshot VDI remains...we should have a base VDI and another new VDI (three
VDIs in total)).
[root@XenServer-6 ~]# xe vdi-list sr-uuid=424db333-31b4-9390-84e9-59c9e2fd2a28
uuid ( RO) : ae7b04b7-7f23-40d3-80df-95cab4bda602
name-label ( RW): ROOT-50
name-description ( RW):
sr-uuid ( RO): 424db333-31b4-9390-84e9-59c9e2fd2a28
virtual-size ( RO): 21474836480
sharable ( RO): false
read-only ( RO): false
If you look at this in XenCenter (which ignores base VDIs), you only see the
snapshot VDI (you should see the "regular" VDI (the one accepting the new
writes) and the snapshot VDI).
> VM Snapshots no longer work with managed storage
> ------------------------------------------------
>
> Key: CLOUDSTACK-9008
> URL: https://issues.apache.org/jira/browse/CLOUDSTACK-9008
> Project: CloudStack
> Issue Type: Bug
> Security Level: Public(Anyone can view this level - this is the
> default.)
> Components: Management Server
> Affects Versions: 4.6.0
> Environment: XenServer 6.5
> Reporter: Mike Tutkowski
> Assignee: Mike Tutkowski
> Priority: Blocker
> Fix For: 4.6.0
>
>
> When using managed storage for the root disk of a VM, you cannot revert a VM
> to a VM snapshot without encountering a RuntimeException that destroys the
> state of your disk.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
