[External] - VMware Cloud on AWS - Getting Error: The attempted operation cannot be performed in the current state ("Powered on").
Article ID: 302850
Updated On:
Products
VMware Cloud on AWS
Issue/Introduction
To follow the workaround section in order to continue being able to expand a disk for the affected virtual machine.
Symptoms:
When a Virtual Machine workload has been migrated via VMware HCX RAV (Replicated-Assisted vMotion) migration type, when trying to expand a disk one can get the following error:
The attempted operation cannot be performed in the current state ("Powered on").
The above error may also be experienced for newly provsioned (newly created) VMs running on SDDC Version 1.8v3, this would be expected and this issue would be resolved and fixed iin a future SDDC release.
Symptoms:
When a Virtual Machine workload has been migrated via VMware HCX RAV (Replicated-Assisted vMotion) migration type, when trying to expand a disk one can get the following error:
The attempted operation cannot be performed in the current state ("Powered on").
The above error may also be experienced for newly provsioned (newly created) VMs running on SDDC Version 1.8v3, this would be expected and this issue would be resolved and fixed iin a future SDDC release.
Cause
Root cause for error message: The attempted operation cannot be performed in the current state ("Powered on").
It is likely that these VMs were migrated via RAV - Replicated-Assisted vMotion (migration method in HCX). And this is due to an issue in path conversion behavior - we provide different behavior for absolute and relative datastore path format. When we get an absolute path like this: "[]/vmfs/volumes/vsan:UUID/OtherDisk/disk1.vmdk", we will (attempt to) do a full canonicalization, but when we get a relative path like this: "[vsanDatastore] OtherDisk/disk1.vmdk", we do not canonicalize the relative part of the path. So when we do canonicalization, we call realpath, but it will fail because the file doesn't exist yet during creation, so we fall back to the uncanonicalized path. Because of this, this where where we do the workaround of creating/deleting the snapshot of the VM, or power-down the VM and edit the disk settings (which fixes the path call logic in the cloud vCenter).
It is likely that these VMs were migrated via RAV - Replicated-Assisted vMotion (migration method in HCX). And this is due to an issue in path conversion behavior - we provide different behavior for absolute and relative datastore path format. When we get an absolute path like this: "[]/vmfs/volumes/vsan:UUID/OtherDisk/disk1.vmdk", we will (attempt to) do a full canonicalization, but when we get a relative path like this: "[vsanDatastore] OtherDisk/disk1.vmdk", we do not canonicalize the relative part of the path. So when we do canonicalization, we call realpath, but it will fail because the file doesn't exist yet during creation, so we fall back to the uncanonicalized path. Because of this, this where where we do the workaround of creating/deleting the snapshot of the VM, or power-down the VM and edit the disk settings (which fixes the path call logic in the cloud vCenter).
Resolution
The long-term resolution will be fixed in an upcoming VMware Cloud on AWS SDDC version release.
Workaround:
Workaround Workflow 1:
a. Select VM Name
b. Right-click --> "Power-off" or "Shutdown Guest OS"
c. Select "Edit VM settings" and scroll down to "Hard Disk" section to expand the disk to preferred disk size
or
Workaround Workflow 2:
(Under the assumption that the VM workload would be left powered-on/running)
a. Select VM Name
b. Right-click --> "Snapshots" --> "Take Snapshot" (Snapshot the virtual machine's memory can be UNCHECKED)
c. Allow the snapshot to be created
d. Right-click the VM name (from #1), --> Select "Delete All Snapshots"
e. Re-attempt in expanding the disk through native cloud vCenter UI for the RAV migrated VM, expanding the disk will work as expected now
Workaround:
Workaround Workflow 1:
a. Select VM Name
b. Right-click --> "Power-off" or "Shutdown Guest OS"
c. Select "Edit VM settings" and scroll down to "Hard Disk" section to expand the disk to preferred disk size
or
Workaround Workflow 2:
(Under the assumption that the VM workload would be left powered-on/running)
a. Select VM Name
b. Right-click --> "Snapshots" --> "Take Snapshot" (Snapshot the virtual machine's memory can be UNCHECKED)
c. Allow the snapshot to be created
d. Right-click the VM name (from #1), --> Select "Delete All Snapshots"
e. Re-attempt in expanding the disk through native cloud vCenter UI for the RAV migrated VM, expanding the disk will work as expected now
Feedback
Yes
No
Powered by
