VCF 4.0 - HostPreparationException: No eligible cache disks found on ESXi

[abuzzi]

Hello,

I'm attempting to perform a bringup on 4x servers each equipped with:

- 1x 32gb flash (used to boot esxi)

- 2x 900gb SSD

Issue I'm getting is:

"Host preparation failed with the following errors: com.vmware.evo.sddc.common.hostservices.error.HostPreparationException: No eligible cache disks found on ESXi host"

From a google search it seems the VCF bringup requires two SSD with different size in order to use the smalled for cache and the other for capacity... is that true ?

If so... how can I make one disk appearing smaller ? (if ever possible)

Thx,

A.

[tenthirtyam]

Since the SSDs are the same size, you'll need to tag the capacityFlash of each host:

Get the Disks:

esxcli storage core devices list

Check the status of a specific SSD disk.

vdq -qH -d <disk>

Example: <disk> = t10.ATA_____INTEL_SSDSxxxx......

Results:

t10.ATA_____INTEL_SSDSxxxx......

..............

   IsCapacityFlash?:  0

Tag the SSD as capacityFlash

esxcli vsan storage tag add -d <disk> -t capacityFlash

Tag the SSD as capacityFlash

esxcli storage core devices list

t10.ATA_____INTEL_SSDSxxxx......

..............

   IsCapacityFlash?:  1

[abuzzi]

Hello Ryan,

it seems both of them are set for CapacityFlash usage:

[root@esxi153:~] esxcli storage core device list | grep naa.55

naa.55cd2e4151461075

   Display Name: Local ATA Disk (naa.55cd2e4151461075)

   Devfs Path: /vmfs/devices/disks/naa.55cd2e4151461075

naa.55cd2e415146402c

   Display Name: Local ATA Disk (naa.55cd2e415146402c)

   Devfs Path: /vmfs/devices/disks/naa.55cd2e415146402c

[root@esxi153:~]

[root@esxi153:~] vdq -qH -d naa.55cd2e4151461075

DiskResults:

      DiskResult[0]:

            Name:  naa.55cd2e4151461075

        VSANUUID: 

           State:  Ineligible for use by VSAN

          Reason:  Has partitions

          IsSSD?:  1

IsCapacityFlash?:  1

          IsPDL?:  0

        Size(MB):  915715

     FormatType:  512e

[root@esxi153:~]

[root@esxi153:~] vdq -qH -d naa.55cd2e415146402c

DiskResults:

      DiskResult[0]:

            Name:  naa.55cd2e415146402c

        VSANUUID: 

           State:  Ineligible for use by VSAN

          Reason:  Has partitions

          IsSSD?:  1

IsCapacityFlash?:  1

          IsPDL?:  0

        Size(MB):  915715

     FormatType:  512e

[root@esxi153:~]

So question now for you is how I revert that...

[root@esxi153:~] esxcli vsan storage tag add -d naa.55cd2e415146402c -t

Error: -t must have value

Usage: esxcli vsan storage tag add [cmd options]

Description:

  add                   Add a vSAN storage tag.

Cmd options:

  -d|--disk=<str>       Name of the disk to which the tag should be added. (required)

  -t|--tag=<str>        Tag to be added to disk. Supported tags are  [capacityFlash]. (required)

[root@esxi153:~]

Thank you,

A.

[tenthirtyam]

Use the following:

esxcli vsan storage tag remove -d <device name> -t capacityFlash

Example:

[root@esxi153:~] esxcli vsan storage tag remove -d naa.55cd2e415146402c -t capacityFlash

[root@esxi153:~] vdq -qH -d naa.55cd2e415146402c

DiskResults:

      DiskResult[0]:

            Name:  naa.55cd2e415146402c

        VSANUUID: 

           State:  Ineligible for use by VSAN

          Reason:  Has partitions

          IsSSD?:  1

IsCapacityFlash?:  0

          IsPDL?:  0

        Size(MB):  915715

     FormatType:  512e

[GreatWhiteTec]

Reason is pointing to disks having partitions. vSAN won't be able to use those devices until you delete the partitions.

You can delete the partitions from the ESXi/vCenter UI.

If this was a vSAN cluster previously you can delete the disk groups and partitions with these commands https://greatwhitetec.com/2020/02/19/devices-unavailable-for-vsan/

[abuzzi]

Hello Ryan,

k, based on your instructions I verified one disk is set for Capacity:

[root@esxi153:~] vdq -qH -d naa.55cd2e4151461075 | grep IsCapacityFlash

IsCapacityFlash?:  1

[root@esxi153:~] vdq -qH -d naa.55cd2e415146402c | grep IsCapacityFlash

IsCapacityFlash?:  0

[root@esxi153:~]

[root@esxi154:~] vdq -qH -d naa.55cd2e4151463bd6 | grep IsCapacityFlash

IsCapacityFlash?:  1

[root@esxi154:~] vdq -qH -d naa.55cd2e4151464a25 | grep IsCapacityFlash

IsCapacityFlash?:  0

[root@esxi154:~]

[root@esxi183:~] vdq -qH -d naa.55cd2e4150b84871 | grep IsCapacityFlash

IsCapacityFlash?:  0

[root@esxi183:~] vdq -qH -d naa.55cd2e4151460eba | grep IsCapacityFlash

IsCapacityFlash?:  1

[root@esxi183:~]

[root@esxi184:~] vdq -qH -d naa.55cd2e415141631e | grep IsCapacityFlash

IsCapacityFlash?:  1

[root@esxi184:~] vdq -qH -d naa.55cd2e4150b84933 | grep IsCapacityFlash

IsCapacityFlash?:  0

[root@esxi184:~]

unfortunately when I restarted the "bringup" it throw up new complain:

Host preparation failed with the following errors: com.vmware.evo.sddc.common.hostservices.error.HostPreparationException: No eligible capacity disks found on ESXi host 10.58.6.153. com.vmware.evo.sddc.common.hostservices.error.HostPreparationException: Failed to delete partitions on ESXi host 10.58.6.154.

Checking again the ESXi153 status I see:

[root@esxi153:~] vdq -qH -d naa.55cd2e4151461075 | grep IsCapacityFlash

IsCapacityFlash?:  1

[root@esxi153:~] vdq -qH -d naa.55cd2e415146402c | grep IsCapacityFlash

IsCapacityFlash?:  0

[root@esxi153:~]

So not sure how to progress now... do you have any idea ?

Thx,

A.

[abuzzi · Accepted solution]

Hello,

I finally passed the vsan showstopper.

Here my notes in case might be of any help to anyone facing same issue.

First use "vdq -q" and beeing sure one SSD has "IsCapacityFlash: 0" (this will be used for cache). Be also sure one disk (SSD or not SSD) has "IsCapacityFlash: 1".

Both disks has to have "State: Eligible for use by VSAN":

[root@esxi153:~] vdq -q

[

   {

      "Name"     : "naa.55cd2e4151461075",

      "VSANUUID" : "",

      "State"    : "Eligible for use by VSAN",

      "Reason"   : "None",

      "IsSSD"    : "1",

"IsCapacityFlash": "1",

      "IsPDL"    : "0",

      "Size(MB)" : "915715",

   "FormatType" : "512e",

   },

   {

      "Name"     : "naa.55cd2e415146402c",

      "VSANUUID" : "",

      "State"    : "Eligible for use by VSAN",

      "Reason"   : "None",

      "IsSSD"    : "1",

"IsCapacityFlash": "0",

      "IsPDL"    : "0",

      "Size(MB)" : "915715",

   "FormatType" : "512e",

   },

   {

      "Name"     : "mpx.vmhba32:C0:T0:L0",

      "VSANUUID" : "",

      "State"    : "Ineligible for use by VSAN",

      "Reason"   : "Has partitions",

      "IsSSD"    : "0",

"IsCapacityFlash": "0",

      "IsPDL"    : "0",

      "Size(MB)" : "30432",

   "FormatType" : "512e",

   },

]

[root@esxi153:~]

Be sure no partition are present, ie:

[root@esxi153:/dev/disks] ll naa.55cd2e415146*

-rw-------    1 root     root     960197124096 May 21 09:40 naa.55cd2e4151461075

-rw-------    1 root     root     960197124096 May 21 09:40 naa.55cd2e415146402c

[root@esxi153:/dev/disks] partedUtil get /dev/disks/naa.55cd2e4151461075

116737 255 63 1875385008

[root@esxi153:/dev/disks] partedUtil get /dev/disks/naa.55cd2e415146402c

116737 255 63 1875385008

[root@esxi153:/dev/disks]

Additionally be sure that Scratch location is not holding any disk/datastore you plan to use for vsan. Ie I put mine into /tmp

[root@esxi153:~] vim-cmd hostsvc/advopt/view ScratchConfig.ConfiguredScratchLocation

(vim.option.OptionValue) [

   (vim.option.OptionValue) {

      key = "ScratchConfig.ConfiguredScratchLocation",

      value = "/tmp"

   }

]

[root@esxi153:~]

Also disable coredump:

esxcli system coredump partition set -u

And reboot the ESXi... and retry the "bringup"

Thank you for all the precious suggestions I got in such post thread.

Thx,

A.

[jemisdafs]

This code is working on my dashboard of the baby website. You can check and give reviews.