Start a Conversation

Unsolved

This post is more than 5 years old

S

4166

October 10th, 2013 16:00

SCSI-3 PR error on WIN2012 hyperv cluster and VNX 5500

Hi All

We have a greenfield site being setup which consists of  UCS b200 Blades, VNX5500  (OE 5.32.206), Win2012 Hyper-V cluster, Powerpath 5.7Sp1

We are using the VIC on the UCS Blades with 4 vHBA's.  These have been set up so 2 (4 paths) are in a BOOT storage group per host and the other 2 are in a Shared storage group for the Clustered volumes.

Running the Cluster validation test fails with a message that the Cluster is unable to obtain a SCSI-3 Persistent Reservation - from everything I have read, the VNX should support SCSI-3 out of the box - is that correct ?

Initiators on the VNX are set to ALUA (mode 4), Clariion/VNX

Any ideas on what the issue is would be most appreciated (the site wants to go live on Sunday !!! )

Thanks


steph

2 Intern

 • 

4K Posts

October 10th, 2013 18:00

VNX supports SCSI-3 PR with ALUA mode.

Have you tried the physical HBA directly? If it works, maybe we can narrow down the problem.

2 Intern

 • 

1.1K Posts

October 10th, 2013 19:00

On Windows, this test validates that the cluster storage uses the more recent (SCSI-3 standard) Persistent Reserve commands (which are different from the older SCSI-2 standard reserve/release commands). The Persistent Reserve commands avoid SCSI bus resets, which means they are much less disruptive than the older reserve/release commands. Therefore, a failover cluster can be more responsive in a variety of situations, as compared to a cluster running an earlier version of the operating system. In addition, disks are never left in an unprotected state, which lowers the risk of volume corruption.

On VNX/CLARiiON, you can also check a file in the SPCollect called SPA_hostside.txt or SPB_hostside.txt file.  Search for ActiveRes in that file to see which LUN has a reservation set.  You will see an output like this where you see the target information.

In this reservation information, the reservation is a SCSI-3 Persistent group reservation. Note the differences (Persisting is set to True, ClearOnReset is set to False, ResType is set to 5.

          =====Persistent Reservation Info=====
         Persisting       : TRUE  ActiveRes          : TRUE
          ResScope         : 0x0   PeerClearNeeded    : NO
          ClearOnReset     : FALSE PResvCmdInProgress : NO
          ResType          : 0x5   ResGenNumber       : 0x10d
          PeerLogoutNeeded : NO    ClearPerResNeeded  : NO
          ResKey : 0x2             
          ResWWN : 20000000C96D6FA9:10000000C96D6FA9
          =====Persistent Reservation Command=====
          ServiceAction   : 0x0   ReservationType : 0x4
          ReservationScope: 0x0 ReservationKey : 0x8367f11d
          HostWWN : 20000000C96D6FA9:10000000C96D6FA9
          DluWWN  : 60060160EA802900:280C204138F6DF11
          Persist : TRUE ClearOnReset: TRUE

In some cases, you need to clear persistent reservations when the host which put the reservations on the devices is no longer there. To do that, I suggest you open a SR and let the EMC support do this for you.

51 Posts

October 10th, 2013 22:00

So Hi KEvin


Thats been a great help.

Ive dug out the SPcollects and here is the data i get

          Persistent Reservation Info:

             Persisting          : FALSE ActiveRes          : TRUE

             ResScope            : 0x0   ResType            : 0x5

             ClearOnReset        : FALSE PResvCmdInProgress : NO

             PeerClearNeeded     : NO    ResGenNumber       : 0x314F

             PeerLogoutNeeded    : NO    ClearPerResNeeded  : NO

             ResKey  :  0AF97AE50001734D

             ResWWN  :  20000025B50C01EF:20000025B50C114F

Persistent Reservation Command:

             ServiceAction   : 0x0 

             ReservationScope: 0x0 ReservationType : 0x4

             ReservationKey             :00000000845EE782

             ServiceActionReservationKey:00000000845EE782

             HostWWN :  20000025B50C01DF:20000025B50C119F

             DluWWN  :  60060160D9F03500:D072EFD4AB30E311

             Persist : FALSE ClearOnReset: TRUE

So any ideas why its not working ??

34 Posts

October 22nd, 2013 17:00

Are the virtualized servers failing CVT or physical nodes themselves?

51 Posts

October 23rd, 2013 16:00

The physical hyper-v cluster failed CVT

Steph

4.5K Posts

March 12th, 2014 12:00

Was there ever a resolution to this issue?

glen

No Events found!

Top