LUN can not add without formatting

[MartinWest]

Hello

Details:

2 x ESXi 4.1 HOST (HP DL380G7)

1 x HP MSA2012fc

The MSA are directly attached over one FC HBA each Server.

I have two LUN's on the MSA (LUN_0 = 1 TB, LUN_1 = 1.6 TB)

On the two ESXi 4.1 HOST, I have access to both LUN's.

My Problem:

I have updated the first HOST to ESXi 5.0.

After the installation, I have two Datastore in the Configuration. Local Datastore and the LUN_0 Datastore.

I jumped to the Storage Adapter and did a Rescan. On the FC HBA I could see three Path:

1. Enclosure

2. LUN_0

3. LUN_1

I jumped back to Storage and press also Rescan. Nothing happen.

I press Add Storage. ESXi 5.0 shows me my LUN_1, and I click on Next.

Now appears to me that I should reformat the LUN. I can choose anything else.

I can not do this! On my LUN_1 are my productive VM's.

Has anybody a solution for me?

Beste wishes

Martin

[kastlr]

Hi Martin,

while updating to ESXi 5.0, did you remove connectivity to the shared LUNs as recommended by VMware in the following article?

Installing ESXi 5.0 best practices

If not you should check if both disks still contain a valid partition table.

Recovering a lost partitiontable on a VMFS volume (1002281)

Kind regards

Ralf

[Virtualinfra]

• 
  

1.Log in to the vSphere Client and select the server from the inventory panel.

2.Click the Configuration tab and click Storage in the Hardware panel.

3.Click Add Storage.

4.Select the Disk/LUN storage type and click Next.

5.From the list of LUNs, select the LUN that has a datastore name displayed in the VMFS Label column and click Next.

Note: The name present in the VMFS Label column indicates that the LUN is a copy that contains a copy of an existing VMFS datastore.

6.Under Mount Options, these options are displayed:

     Keep Existing Signature  - Persistently mount the LUN (for example, mount LUN across reboots)

     Assign a New Signature  - Resignature the LUN

     Format the disk  - Re-format the LUN.

Note: This option will delete any existing data on the LUN.

7.Select the desired option for your volume.

8.In the Ready to Complete page, review the datastore configuration information and click Finish

In the 6th step select keep existing signature- which will not formated the disk.

Refer the below KB for more information

http://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC&externalId=101138...

Award points for the helpful and correct answer by clicking the below tab

[MartinWest]

Hi

Thank you for answering.

My Problem is, that I can not select  this two options:

Keep Existing Signature - Persistently mount the LUN (for example, mount LUN across reboots)

Assign a New Signature - Resignature the LUN

These are greyed out.

Only "Format the Disk" is avaiable.

Regards

Martin

[Virtualinfra]

Did you try that from vcenter or connecting to esxi host via vsphere client.

If you have tried from vcenter, please try with connecting hte esxi host directly via vsphere client.

Also try with the esxcli command to mount the disk, will if that works fine,

Or you have try with the other option with the another update stating recovering inaccessible  vmfs datastore,

[MartinWest]

Hallo Ralf

Thank you for answering!

Best Practices for installing ESXi 5.0, are written that you have to detach the fibre before installing.

This is only recomended to be secure that you not use the HBA as primary boot partition.

think we can exclude that.

Recovering a lost partitiontable?

Do you think I should do this on a productive system?

I have access to the LUN over the second Server, without any problems.

Grüsse

Martin

[Virtualinfra]

Ok try this.

With storage admin help. ask him to just unpresent the LUN(remove visiblity to the host, make sure he is not to deleting the volume or remove the volume) which your not able to add to the esxi host.

Do a rescaning.

request storage admin to present the lun

do a recaning.

connect the esxi directly via vsphere client and add the lun with option keep signature, if its not greyout.

[kastlr]

Hallo Martin,

you should at least check if the partition table is still present.

If the partition table was wiped out, your second server will lose the VMFS datastore after a rescan or reboot operation.

And two weeks ago I was involved in a similar incident on a large customer.

They installed two ESXi 5 Servers in their clustered environment and for unknown reason the partition tables of ALL used VMFS datastores were cleared.

VMware was able to recreate them for the majority of disk with the following article.

Recovering a lost partition table on a VMFS volume (1002281)

We did analyze the enterprise storage array and SAN and didn’t find anything there.

VMware is still working on their analysis.

Viel Glück,

Ralf

[bsomers · Accepted solution]

Your solution could be as simple as the LUNs need to be mounted on the other hosts.

Putty into the host(s) not showing the Luns and type

esxcfg-volume -l , this will list any usable LUNs that are NOT mounted, if no LUNs are displayed it means all your LUNs have been mounted.

If you see any LUNs listed from the command above, type this to mount each LUN one at a time

esxcfg-volume -m "vmfs_label_name" , the lable name should be the friendly name you gave the LUN when originally formatted.

I had this same problem with one of my ESXi hosts and this took a care of it without formatting.

Hope it helps

BS

[MartinWest]

Hi BS

Thank you very much!

It works.

All the best and merry christmas and a happy new year!

Martin

[CMCC]

I'm having this situation with a new ESXi 5 host.

When I run the esxcfg-volume -M I can mount the lun, but everytime it reboots, it's lost again...and this command supposed to be to mount the volume without performing a resignaturing of that volume and the volume is mounted when the ESX host is rebooted.

What else can I do? I already reinstall the ESXi server but the same.

The lun is vmfs3

Thanks in advance.

[bsomers]

Well you need to make sure you are using a capital M in the command. A lowercase m does just that mounts the LUN until a reboot while an uppercase M is persistant accross reboots.

[sa_ostad]

It works for me :

esxcfg -M "DataStore Label"