I hope somebody can help me out here, I am running out of ideas. I have a DS4300 connected to IBM TotalStorage SAN24 something fiber switches. I have two IBM xServer 3650's connected using Qlogic single port HBA's, two per host. The hosts are running Windows Server 2008 R2 and I am using IBM Storage Manger 10.83 though I have other versions if required. The issue I am having is that no matter what I try to do with the SAN, the fiber switches, the Qlogic cards, MPIO, DSM's, RDAC etc. every time I run the Cluster Validation Wizard it tells me that the LUN I presented has a persistent reservation on it. When I try to add storage to the cluster it says that there are no suitable disks available. I have run cluster node <nodename> /clearpr:<disk number>, I have run diskpart to clean the drive, I have tried with the disk online, offline formatted, unallocated etc. Nothing works. The disks appear just fine on both hosts, I can online and offline them to my hearts content, no errors, no issues. Just can't add them to the cluster.
What am I doing wrong? I have seen multiple threads about DSM/SDDSM/MPIO/RDAC etc. But to be honest none of them have been really easy to follow how to fix this issue. It sounds like my problem is something to do with the fact that RDAC is not supported by Windows any longer though I do not know how to fix it or the fixes I have tried do not work.
I should note that this SAN was previously hooked up to these hosts and worked just fine with Windows Server 2008 R2 Failover Clustering. The difference now is that I tore it all down and when I set it back up I updated the firmware/esm/bios/nvram on everything I possibly could. The DS4300, the fiber switches, and the Qlogic HBA's and Storage Manager are all as up to date as I can get them. So, did I update too far and updated myself out of a working SAN for Windows Server 2008 R2?
Any help would be greatly appreciated, I am going insane here trying to get this working.
This topic has been locked.
5 replies Latest Post - 2013-05-04T11:31:07Z by Novikov_Alexander
Pinned topic DS4300 + Windows Server 2008 R2 = unable to cluster
Answered question This question has been answered.
Unanswered question This question has not been answered yet.
Updated on 2013-03-20T16:03:00Z at 2013-03-20T16:03:00Z by SystemAdmin
jbrannon 270005F7TA2 Posts
Novikov_Alexander 100000HTD51404 Posts
Shark800 270000VD763 PostsACCEPTED ANSWER
Re: DS4300 + Windows Server 2008 R2 = unable to cluster2012-12-18T19:35:18Z in response to jbrannonJesse,
I'm having the same issue on 7946 AC1's (5 node cluster) , running Win2k8 R2 SP1, MPIO 1.03.305.608 x64bit, que depths, and MPIO policies set appropriate (Least Queue Depth-DSM acknowledged), Emulex 42C2069's storport:126.96.36.199. The back end are two 1742's fully populated, and IBM/Brocade model 2005 FC switches. All active zone sets are configured appropriate.
From what I can see the DS4000 subsystem is supportive SCSI3 commands which Win2k8 R2 SP1 utilizes in clustering instead of the legacy SCSI-2. I'm assuming it is? as it's referenced in a couple readme release files, and an IBM document for GPFS-Persistent Reservations along with your reference in having a previous configuration with Win2k8 clustering working.
I have found a few kb's referencing persistent reservations, but have not implemented yet. I'm curious if you've had any luck or direction since your original post. I have also verified from the shell of the DS4400's a/b controllers that reserveId:0xffff resv3rdId:0xffff on all LUN id's (qty: 77) have no persistent reservation being held from the subsystem acknowledged.
SystemAdmin 110000D4XK4779 Posts
Novikov_Alexander 100000HTD51404 PostsACCEPTED ANSWER
Re: DS4300 + Windows Server 2008 R2 = unable to cluster2013-05-04T11:31:07Z in response to SystemAdmin
could you post DSA logs from both nodes, FC zone configurations, DS Support Data and screenshot with error.