Cannot create VMFS Datastore on newly installed ESX 6.5 (Licensed)

[ESX_STAGE]

Hi,

we have a new Dell PowerEdge R730xd server setup and preconfigured with DellEMC Customized Image ESXi 6.5 Update 1 A08 (based on ESXi VMKernel Release Build 7388607) .

I am using VSphere Essentials License.

Without any changes, I logged into the Vsphere Web Interface /ui/#/host and attempted to create a new VMFS datastore but got an error:

Failed to create VMFS datastore test - Cannot change the host configuration.

SPECS Used:

Name: Test

Devices: 1 x 10TB Virtual Disk (Raid 10 x 4 Physical Disks) setup by Dell.

Size to create VMFS datastore: 1 TB

VMFS Type:  Both VMFS 5 and 6 failed in separate tests.

No other existing datastores or VM's.

Contacted Dell but awaiting response and as it ESX they may not help.

As we need this server setup ASAP I was wondering if anyone knows of any resolutions or a point in the right direction with this.

I have used ESX different flavors for years and never seen this before.

Any help or suggestions greatly appreciated.

I also have limited command line experience.

[vijayrana968]

Please refer to the KB VMware Knowledge Base

[KyleBrown]

The hyperlinked article no longer exists.

[rchaubey]

can you share vmkernel.log to do further investigation.

Regards,

Randhir

[Finikiez]

You need to check vmkernel.log  according to time when you try to create VMFS datastore.

Usually outdated driver \ firmware for raid controller causes such issues.

If you use server from Dell then try to install ESXi using Dell customized ISO.

[a_p_]

What I could think of is that there's something on the LUN (Virtual Disk) that causes the issue, e.g. some dirty bits which are interpreted incorrectly!?

In this case you could try to clear the 10TB virtual disk's partition table. To do this, select Storage in the inventory, then switch to the Devices tab. With the proper device selected, run "Clear partition table" from the Actions menu.

Devices: 1 x 10TB Virtual Disk (Raid 10 x 4 Physical Disks) setup by Dell.

Size to create VMFS datastore: 1 TB

Please remember that VMware supports only a single datasatore per LUN (Virtual Disk)!

André

[Cookies04]

I had the same problem and in order to fix it I had to run three commands through an SSH connection.  From what I have seen and found this error comes from having disks that were part of different arrays and contain some data on them.  When I ran the commands I was then able to connect the data stores with no issues.

1.     Show connected disks.

          ls -lha /vmfs/devices/disks/

               (Verify the disk is seen.  You will probably see your disk ID then :1.  This is a partition on the disk.  We only need to work about the main disk ID.)

2.     Show the error on disk.

          partedUtil getptbl /vmfs/devices/disks/(disk ID)

               (It will probably indicate that the GPT is located beyond the end of the disk.)

3.     Wipe disk and rewrite with a basic MSDOS partion.

          partedUtil setptbl /vmfs/devices/disks/(disk ID) msdos

               (The output from this should be similar to msdos and the next line will be o o o o)

I hope this helps you out.

[marklkmetz]

Thank you!

[JPhiffer]

Thanks Cookies04​ your solution immediately fixed my issue.

[hicall]

thank you Cookies04​ , this method has been successful with simple solution !

[Anxionbaby]

Hi

I got this error could you help me please

msdos

0 0 0 0

Error: Read-only file system during write on /dev/disks/naa.600508b1001c50fcc5654d86b955aa6e

WriteNewPtable: error removing gpt signatures from (/dev/disks/naa.600508b1001c50fcc5654d86b955aa6e)

[Hopes26]

Hello , i have the same problem , i cannot create my own DATASTORE .

I'm blocked for a long time , i followed a lot of tuto but still same issue , can you please help me to resolve this problem !!

[Anxionbaby]

Hi ,

I was able to fix my issue.

Could you run this     ls -lha /vmfs/devices/disks/

and paste image of your disks. thanks

[a_p_]

According to the screenshot you are trying to overwrite the system/installation disk on an ESXi 7.0 host!?

André

Moderator note: Please consider to open a new discussion instead of asking for help on an already 2 years old one. A new discussion you will get much more attention.

[Steeve_Savard]

those explanation were perfect for me buddy...

they got it to work...

thanks...

cheers.

[ramnath k]

- Problem not solved as such. I installed ESXi 7.0

[root@localhost:~] ls -lha /vmfs/devices/disks/
total 790622193
drwxr-xr-x 2 root root 512 Nov 21 09:39 .
drwxr-xr-x 16 root root 512 Nov 21 09:39 ..
-rw------- 1 root root 750.0G Nov 21 09:39 naa.6000c29f18895dbbb2f909fa191e35d8
-rw------- 1 root root 100.0M Nov 21 09:39 naa.6000c29f18895dbbb2f909fa191e35d8:1
-rw------- 1 root root 500.0M Nov 21 09:39 naa.6000c29f18895dbbb2f909fa191e35d8:5
-rw------- 1 root root 500.0M Nov 21 09:39 naa.6000c29f18895dbbb2f909fa191e35d8:6
-rw------- 1 root root 2.9G Nov 21 09:39 naa.6000c29f18895dbbb2f909fa191e35d8:7
lrwxrwxrwx 1 root root 36 Nov 21 09:39 vml.02000000006000c29f18895dbbb2f909fa191e35d8564d77617265 -> naa.6000c29f18895dbbb2f909fa191e35d8
lrwxrwxrwx 1 root root 38 Nov 21 09:39 vml.02000000006000c29f18895dbbb2f909fa191e35d8564d77617265:1 -> naa.6000c29f18895dbbb2f909fa191e35d8:1
lrwxrwxrwx 1 root root 38 Nov 21 09:39 vml.02000000006000c29f18895dbbb2f909fa191e35d8564d77617265:5 -> naa.6000c29f18895dbbb2f909fa191e35d8:5
lrwxrwxrwx 1 root root 38 Nov 21 09:39 vml.02000000006000c29f18895dbbb2f909fa191e35d8564d77617265:6 -> naa.6000c29f18895dbbb2f909fa191e35d8:6
lrwxrwxrwx 1 root root 38 Nov 21 09:39 vml.02000000006000c29f18895dbbb2f909fa191e35d8564d77617265:7 -> naa.6000c29f18895dbbb2f909fa191e35d8:7
[root@localhost:~]

- getinfo about the 750GB partition

[root@localhost:~] partedUtil getptbl /vmfs/devices/disks/naa.6000c29f18895dbbb2f909fa191e35d8
gpt
97906 255 63 1572864000
1 64 204863 C12A7328F81F11D2BA4B00A0C93EC93B systemPartition 128
5 208896 1232895 EBD0A0A2B9E5443387C068B6B72699C7 linuxNative 0
6 1234944 2258943 EBD0A0A2B9E5443387C068B6B72699C7 linuxNative 0
7 2260992 8388574 4EB2EA3978554790A79EFAE495E21F8D vmfsl 0
[root@localhost:~]

- try setting to msdos...same error

[root@localhost:~] partedUtil setptbl /vmfs/devices/disks/naa.6000c29f18895dbbb2f909fa191e35d8 msdos
msdos
0 0 0 0
Error: Read-only file system during write on /dev/disks/naa.6000c29f18895dbbb2f909fa191e35d8
WriteNewPtable: error removing gpt signatures from (/dev/disks/naa.6000c29f18895dbbb2f909fa191e35d8)

[root@localhost:~]

[Mallenge]

Mysteriously I'm facing the same issues on my home lab, after upgrading my machine to Big Sur and the Fusion to 12.1.0.

The ESXi servers are now working fine, but the FreeNAS 11.3U5 I have no. No matter which configuration I use (BIOS, UEFI, SCSI, SAS) the disk looks like "read only".

Looks like it's happening with more than just test stuff...

[KyleBrown]

I can't figure this one out, so here is a screenshot of the output of ls -lha /vmfs/devices/disks/.

[PaulRobert1]

Hello, Thank you very much Cookies04, your procedure worked like a charm.

[Crunchy-Nugget]

 Thank you. This helped. My disks were taken from another device that had a RAID controller. The disks were initialized with the RAID controller prior to removal to erase them, but I guess it doesn't work.

[m_manar]

Thank you !!!!!!!!

[m_manar]

 i logged in only to say: thank you! 

[Mattneed]

 I know this is old as hell but even in 2022 you the MVP thanks for this!!!

[Texan]

This post is a bit old but I found it while researching the same error message. My R540 had the exact same message and it ended up being that the newly created array that I was attempted to create a new datastore on was not finished doing a full initialization by the controller. During the array config I opted to do a full initialization instead of a quick. Once I waited it out, the VMFS datastore creation went fine. Even though the iDrac had the array marked as "online" and ready it would fail with the error message.

[yeric79]

hi,

i have the same problem with an esxi 6.7 and a local ssd Samsung 840 Pro.

esxi 6.7 FSS: 2350: Failed to create FS

(this is a test computer)

[aamirgondal]

hi, I want to ask have you resolve your issue?

i am having the same problem

thanks

 

[Mallenge]

Hello! Yes, I have. Take a look at my post on this same thread about the MTU and this has a big chance of solving your problem. 

[Mallenge]

Hey, everybody! I figured out the problem with my home lab, at least. Could be worth a look to yours, as well. 

My main problem was with the iSCS communication between the hosts and the FreeNAS. I was continuously getting timeout errors between them and when I reduced the MTU from 9000 to standard 1500, the timeouts stopped and the volumes were correctly created without issues. 

Therefore, my tip is just this: check the protocols with the communication between the hosts and the storage. 

[zachmi23]

I can't believe it, but this worked for me! I went into each VMKernel Adapter and sure enough, the iSCSI adapters had an MTU of 9000 (presumably to allow for jumbo packets) so I changed them to 1500 and the datastore created without issue!! Cheers!!

[davdjross]

Thanks for this!

I'd just updated my homelab with a 2.5gbe switch and stupidly assumed 9000MTU would be what should be set at rather than 1500. Clearly not. Set it back to 1500MTU and the datastore was created immediately.