ESXi

 View Only
  • 1.  Dealing with orphaned snapshots

    Posted Feb 01, 2017 01:11 AM

    I have a VM running on an ESXi 5.5 that had one disk pointing to a snapshot file after running a 'Delete All' command to consolidate all the snapshots. All the other disks on the VM consolidated fine. The VM was then showing a 'Virtual machine disks consolidation is needed.' message. I checked the VMX file for the VM and the one 3.5TB disk was pointing to a '-000002-sesparse.vmdk' disk.

    I checked the VMware support pages and found:

    Determining if there are leftover delta files or snapshots that VMware vSphere or Infrastructure Client cannot detect (1005049)

    Reading through that and running the commands showed there was a 'sesparse.vmdk' file in the deltadisks folder. The guide then pointed to:

    Committing snapshots when there are no snapshot entries in the Snapshot Manager (1002310)

    At first I ran the suggested command line fixes, but when I ran the 'snapshot.get' command it did not list any snapshots like what was listed in the guide. So I stopped using the command line and tried the first suggestion to create a new snapshot and then run 'Delete All' again.

    The 'Take Snapshot...' command started and then hung at 95% for a while then gave the following failure notice:

    'The guest OS has reported an error during quiescing. The error code was: 5 The error message was: 'VssSyncStart' operation failed: IDispatch error #8449 (0x80042301)'

    Now all the disks have snapshot delta files and the initial disk with the problem has a new delta disk starting with '-000001' (Down from -000002).

    Here is the folder structure for the VM (the trouble disk is on a separate LUN and all other disks are stored with the VM):

    Root VM Folder

    BAKEDBEAN.nvram

    BAKEDBEAN.vmsd

    BAKEDBEAN.vmx

    BAKEDBEAN.vmx.lck

    BAKEDBEAN.vmx~

    BAKEDBEAN.vmxf

    BAKEDBEAN_4.vmdk

    BAKEDBEAN_4-000001.vmdk

    BAKEDBEAN_4-000001-ctk.vmdk

    BAKEDBEAN_4-000001-delta.vmdk

    BAKEDBEAN_4-ctk.vmdk

    BAKEDBEAN_4-flat.vmdk

    BAKEDBEAN_5.vmdk

    BAKEDBEAN_5-000001.vmdk

    BAKEDBEAN_5-000001-ctk.vmdk

    BAKEDBEAN_5-000001-delta.vmdk

    BAKEDBEAN_5-ctk.vmdk

    BAKEDBEAN_5-flat.vmdk

    BAKEDBEAN_6.vmdk

    BAKEDBEAN_6-000001.vmdk

    BAKEDBEAN_6-000001-ctk.vmdk

    BAKEDBEAN_6-000001-delta.vmdk

    BAKEDBEAN_6-ctk.vmdk

    BAKEDBEAN_6-flat.vmdk

    BAKEDBEAN_7.vmdk

    BAKEDBEAN_7-000001.vmdk

    BAKEDBEAN_7-000001-ctk.vmdk

    BAKEDBEAN_7-000001-delta.vmdk

    BAKEDBEAN_7-ctk.vmdk

    BAKEDBEAN_7-flat.vmdk

    BAKEDBEAN-aux.xml

    BAKEDBEAN-ca8e35fd.hlog

    BAKEDBEAN-ca8e35fd.vswp

    vmware.log

    vmware-16.log

    vmware-17.log

    vmware-18.log

    vmware-19.log

    vmware-20.log

    vmware-21.log

    vmx-BAKEDBEAN-3398317565-1.vswp

    Separate LUN for this disk (3.5TB thick provisioned disk / 4TB total LUN size)

    BAKEDBEAN.vmdk

    BAKEDBEAN-000001.vmdk

    BAKEDBEAN-000001-ctk.vmdk

    BAKEDBEAN-000001-sesparse.vmdk

    BAKEDBEAN-000002.vmdk

    BAKEDBEAN-000002-ctk.vmdk

    BAKEDBEAN-000002-sesparse.vmdk

    BAKEDBEAN-ctk.vmdk

    BAKEDBEAN-flat.vmdk

    One other note to add is that I am using Veeam for backup, so CBT is turned on. I know this has caused problems on this same VM before when we migrated it to a new vCenter and at one point had to edit the VMX to turn off the CBT setting.

    Any tips on what I can do to consolidate this VM?

    Thank you,


    Brent



  • 2.  RE: Dealing with orphaned snapshots

    Posted Feb 01, 2017 02:42 PM

    Shutdown the VM and try to consolidate.

    less BAKEDBEAN.vmx | grep vmdk

    Check the disk consistency/Chain with below command

    vmkfstools -e -v10 BAKEDBEAN-000002.vmdk

    Note :- Once the VM is down and consolidated started  wait until it completes. Do not power on VM or cancel the task. 



  • 3.  RE: Dealing with orphaned snapshots

    Posted Feb 01, 2017 04:33 PM

    ArjunDooti,

    After shutting down the VM, the snapshot manager the VM still showed no snapshots to consolidate, so I ran another snapshot while the VM was off. This last snapshot showed up in the manager, so then I consolidated the snapshot and all the vmdks are back to normal. No more orphaned deltas.

    Thank you,

    Brent