VMware Cloud Community
andrevechiatto
Enthusiast
Enthusiast
Jump to solution

ESXI 5.0 with Cluster Microsoft Windows 2003 Lost RDM after reboot host

Whenever I need to do maintenance on a ESXi host that has a cluster microsoft 2003 after the host is restarted all RDM disks Cluster Microsoft loses access to RDM disks and displayed with size 0 KB. Only it is possible to restore the cluster off the second Microsoft cluster node and removing and adding the RDM disks. Anyone ever had this problem?

Reply
0 Kudos
1 Solution

Accepted Solutions
a_p_
Leadership
Leadership
Jump to solution

You'll find this information in the MSCS setup guide (see http://kb.vmware.com/kb/1004617)

André

View solution in original post

Reply
0 Kudos
17 Replies
a_p_
Leadership
Leadership
Jump to solution

Just to clarify. Is there anything you need to do manually, before you can move the resources to the restarted cluster node?

It is not unusual for a restarted MSCS cluster node that it doesn't see the resources before they have been active at least once on that node.

André

Reply
0 Kudos
linotelera
Hot Shot
Hot Shot
Jump to solution

What type of RDM are you using? Virtual or Physical?

Before maintenance have you switched all cluster service to failover VM?

Reply
0 Kudos
linotelera
Hot Shot
Hot Shot
Jump to solution

[ot] Sorry for duplicate answer... We are replying at the same time... But a.p. win 🙂

Reply
0 Kudos
andrevechiatto
Enthusiast
Enthusiast
Jump to solution

I move the resources from the active node to the passive node. I switch off the active vm and restart the host. After the host restart the RDM disks appear as 0KB the VM that was turned off

Reply
0 Kudos
andrevechiatto
Enthusiast
Enthusiast
Jump to solution

RDM Physical - Cluster CAB

Yes I switched all cluster service to failover VM

Reply
0 Kudos
a_p_
Leadership
Leadership
Jump to solution

I move the resources from the active node to the passive node. I switch off the active vm and restart the host. After the host restart the RDM disks appear as 0KB the VM that was turned off

The question is, are you able to failover resources to the restarted MSCS cluster node at this point?

MSCS is a shared-nothing cluster, where the active node doesn't allow the passive node to access its owned resources.

André

Reply
0 Kudos
a_p_
Leadership
Leadership
Jump to solution

linotelera wrote:

[ot] Sorry for duplicate answer... We are replying at the same time... But a.p. win 🙂

No need to apologize! All help is welcome, and (hopefully) finding a solution makes us all winners, doesn't it.

André

Reply
0 Kudos
andrevechiatto
Enthusiast
Enthusiast
Jump to solution

Yes I able to failover resources before restarted MSCS cluster node.

Reply
0 Kudos
a_p_
Leadership
Leadership
Jump to solution

Are you able to failover cluster resources after restarting the MSCS node? If not, are there any error messages?

André

Reply
0 Kudos
andrevechiatto
Enthusiast
Enthusiast
Jump to solution

Not possible power-on VM again because RDM disk appear with 0 KB

Reply
0 Kudos
roneng
Enthusiast
Enthusiast
Jump to solution

Hi

I might be completely off here, but take a look at that

http://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC&externalId=101610...

Another thing that comes to mind is that you might be using an old esxi 5.0 release that has some bugs with MSCS , or unsupported storage configuration.

good luck.

Reply
0 Kudos
a_p_
Leadership
Leadership
Jump to solution

Can you please attach the vmware.log as well as the .vmx files for the two cluster nodes to a reply post (use "Attach files" below the reply editor window).

André

Reply
0 Kudos
andrevechiatto
Enthusiast
Enthusiast
Jump to solution

Node 1  - bsbrsp3331

node 2  - bsbrsp3334

Reply
0 Kudos
a_p_
Leadership
Leadership
Jump to solution

From a first look at the files, it seems that the virtual disks' scsi adapters do not match. Make sure the shared virtual disks are mapped to the same virtual scsi port on both VMs. What you may also need to verify is that the RAW LUNs are presented to both hosts using the same Host LUN ID.

André

andrevechiatto
Enthusiast
Enthusiast
Jump to solution

Is requirements that shared virtual disks are mapped to the same virtual scsi port on both VMs? What's Vmware KB ou doc?

Reply
0 Kudos
a_p_
Leadership
Leadership
Jump to solution

You'll find this information in the MSCS setup guide (see http://kb.vmware.com/kb/1004617)

André

Reply
0 Kudos
andrevechiatto
Enthusiast
Enthusiast
Jump to solution

The problem is different virtual scsi port. Thanks your help.

Reply
0 Kudos