Could someone provide a guideline if it is still posible to implement Microsoft Failover clustering across the boxes with vMotion, DRS capability?
This was definetely possible before with RDM virtual mode but it seems discontinued after Windows Server 2003. The scenario I am looking is as below;
- 2 Node MS Failover cluster implementing SQL Clustering across two nodes.
- Windows Server 2012 R2.
- SQL Server 2014
- ESXi 5.x.
Thanks in advance.
You gotta upgrade to vSphere 6
Thanks but I find it very odd that Windows Server 2003 with MSCS is supported with RDM virtual compatibility mode and VMware pulled this feature after 2008 and above just to bring it back in with vSphere 6.0? Article suggest this is the first time VMware supported this feature, which I do not necessarily agree.
This was and is officially supported on Windows Server 2003 and I know for a fact I used to use this feature.
Any explaination on why VMware would not support this after 2003?
To be honest, I do not remember of it being ever supported. However, I could be wrong and would let others chime in.
For reference, I pulled MSCS setup guide for VI 3 (ESX 3.x) and it clearly states that vMotion for clustered VMs is not supported:
It can be seen on Failover Clustering guide that
Windows Server 2003 was/is still supported for CAB in RDM virtual compatibility mode. On a seperate KB article it states that vMotion across hosts with RDM physical compatibility mode is supported.
This made me think that although VMware pulled the support of CAB on RDM non-pass-through they provided the warm vMotion on physical mode RDM but it does not seem to be working.
We have a 2008 MSCS cluster setup with Physical mode RDM's within ESXi 5.0. I was looking over the whitepaper I used to setup the cluster, and as vGuy says, it does state that Migration with vMotion of clustered virtual machines is a setup limitation of the MSCS configuration. Have a look at this table that shows the different clustering solutions.
I think the reason why virtual mode RDM support was dropped for CAB Windows Server 2008+ is because Microsoft changed the way cluster members use SCSI-3 persistent reservations since 2008+. Part of the reason why it has to be a physical mode RDM is being explained here:
vSphere 5.5, RDMs and Microsoft Clustering - CormacHogan.com
MSCS is probably the only real use case for passthru RDMs these days. We need to use passthru RDMs in MSCS as these allow all SCSI commands to be passed all the way down to the disks. In the case of MSCS, it uses SCSI reservations on disks to gain quorum/ownership/control of services in the event of a failure/fail-over. If we didn’t use these passthru RDMs and used VMDKs or virtual mode/non-passthru RDMs instead, the SCSI reservation would get translated to a file lock. This would be okay if we were running MSCS using VMs on the same ESXi host (we call this a CIB configuration, short for Clustering in a Box) but it wouldn’t allow MSCS to work correctly across ESXi hosts.