VMware vSphere

 View Only
  • 1.  The need to run a chkdsk equivalent on a attached datastore?

    Posted Apr 21, 2020 09:03 PM

    My storage admin just told me that they are seeing possible corruption (maybe false positive) on a m disk in a v7000 SAN.
    As they cant see inside a LUN they have presented to my hosts the SAN admin is suggesting I should run a chkdsk equivalent on my LUN/Datastore.

    I have never come across this before and a bunch of googling didn't make me any smarter in this matter.

    Now, I am not seeing any corruption of my VM's running on that datastore and obviously they would not see anything as they are abstracted from the storage.

    Would this be a chkdsk from the console of the ESXi host? or perhaps something I can do via powercli.  Any kind of service interruptions to expect or perhaps i should vacate the LUN/Datastore.

    I spoke with my SAN admin and he mentioned he have seen false positives of corruption before on this array.



  • 2.  RE: The need to run a chkdsk equivalent on a attached datastore?

    Broadcom Employee
    Posted Apr 22, 2020 11:05 AM

    Moderator: Thread moved to the vSphere area for visibility.



  • 3.  RE: The need to run a chkdsk equivalent on a attached datastore?

    Posted Apr 22, 2020 12:03 PM

    You can use VOMA utility on esxi hosts to check luns or vmfs datastore metadata checks VMFS file system metadata check. Here is VMware KB VMware Knowledge Base

    Note: when you perform checks make sure you move all vms to another datastore or power off the vms and then perform the check against datatstore.