11 Responses

  1. Graham Mitchell
    Graham Mitchell at |

    Chris, SDelete is a great tool for simple machines (which we don’t have). The one big failing of SDelete, which makes it almost useless for us, is that it will not work on disks that are mounted at mount points. So in your example, your W drive was just a 4GB drive, and the real storage was on a 32GB drive mounted in a directory on that drive, then SDelete wouldn’t reclaim the space on the mounted drive.

    I’m not really sure why this limitation is in the program, since every drive letter is just a mount point any way, but unfortunately it just doesn’t work.

    Reply
  2. Ed Grigson
    Ed Grigson at |

    It’s also worth pointing out this a similar issue affects Linux VMs, and a similar solution is available (using ‘dd’ instead of sdelete). A quick Google should tell you all you need to know.

    Reply
  3. David Pasek
    David Pasek at |

    Thanks for another nice blog post and your continuous contribution to vmware community. Just small comment … It is about reclaiming space of thin provisioned vDisk inside guest OS sitting on NFS datastore. Absolutely same trick can be used for block storages (FC, iSCSI). If block storage supports hardware acceleration (aka VAAI) storage vMotion can leverage cloning block VAAI primitive. That’s IMHO advantage of VAAI ready block storage against NFS for this particular use case.

    Reply
  4. How To Reclaim Unused Disk Space on NFS Storage | Storage CH Blog

    […] Read on here […]

  5. IT_Architect
    IT_Architect at |

    >Keep in mind that a storage array that’s able to compress and/or deduplicate the underlying NFS volumes will make this effort largely irrelevant for most situations.<

    That is not the case for me. I have Windows VMs that is present NFS volumes, and the NFS volumes are underpinned by NTFS compressed volumes. While VMware and the OS both agree on the amount of free space, after so many backups to the NFS volumes using vkmfstools, The backups fail and say they are out of space when copying a 37GB VM to virtual disk with hundreds of GB of free space according to Windows and ESXi. You can delete a couple of the backups, but it will still fail. When it fails, it will always fail at the exact same point in the backup. After some days, it will often fix itself and resume backups. This has been going on for a long time. The answer here is what I thought I found at first, but that is not the case.

    Reply
  6. zhangdl
    zhangdl at |

    Hello chris
    Could you share how to use netapp vsc to shink VMDK ?

    Reply
  7. Reclaim Unused Guest Disk Space on NFS Storage, without vm downtime. | Virtual ManaGeMenT

Share your point of view!