Call “HostDatastoreSystem.CreateVmfsDatastore” for object “ha-datastoresystem” on ESXi “xxx.xxx.xxx.xxx” failed.

While working with my home vSphere 5.1 lab the other day, I was trying to create a VMFS5 datastores on my local SATA disk. Each time I tried to do that I was just welcomed with the following error:

Call “HostDatastoreSystem.CreateVmfsDatastore” for object “ha-datastoresystem” on ESXi “xxx.xxx.xxx.xxx” failed.

Please note xxx.xxx.xxx.xxx stand for my host IP. To visualize the error below is a screen shot of the error as it has appeared in my home lab

Call "HostDatastoreSystem.CreateVmfsDatastore" for object "ha-datastoresystem" on ESXi "192.168.2.202" failed.

After fuzzing around trying to figure out what happened, I have remembered this particular disk was used by one of my old lab ESXi hosts. As I do all kind of crazy things in my labs, I thought I should try to wipe the disk clean then try to format it with VMFS5 afterward. That has actually solved the problem. This error seems to happen if you have a file system on that LUN/disk that ESXi does not understand, cannot overwrite or if You don’t have a full write access to the disk/LUN(ex: SRM replication target)

Below is how I went about doing it if you are not familiar with the procedure:

Note: In previous versions of ESXi fdisk was your friend in such a situation, though if you try it in vSphere 5.x you will get the following error message:

*** The fdisk command is deprecated: fdisk does not handle GPT partitions. Please use partedUtil ***

fdisk cannot handle GPT partitions

I am sure from the above error you got where I am going next, you will need to use partedUtil instead of fdisk to wipe that disk clean so you can use it. Alright, Alright, here is your one line magical command that will wipe that disk clean:

#PartedUtil mklabel /dev/disks/<disk id> msdos

Below is an example of how it looked like in my environment.

#partedUtil mklabel /dev/disks/t10.ATA____WDC_WD5000AK52000UU340___________WD2DWCAYU6597660 msdos

ESXi Command to wipe a disk clean

Alright in case you have wondered how did I figured out my disk id, all I had to do to get that info is to run the following command just as demonstrated in the previous screenshot:

#ls /dev/disks

After wiping the LUN/Disk clean, now you should be able to use it to create your new VMFS datastore by adding it through the vSphere client(business as usual). Hope this help someone out there!

Comments

  1. Another Solution!!

    Add the same volume as VMFS-3, this should be successful as the previous partition would be VMFS-5.

    Once VMFS-3 is created, you can now delete the datastore to create a VMFS-5 datastore.

    This worked for me.

  2. I was able to re-part the disk to msdos. Didn’t fix my issues with Adaptec 6508e adapter and a jbod disk that gives me this message. I do have a mirrored set working with esxi 5.1 already on same controller so don’t now what gives. I will call adaptec today.

    Also, you have a line of code: #PartedUtil mklabel /dev/disks/ msdos

    the P in PartedUtil needs to be lower case.
    You do have the command higher up in article correctly, just that one is incorrect.

    Thanks for sharing.

Speak Your Mind

*

Sponsors