MPIO policy settings for Windows cluster

Hi experts,

What is the correct load balance policy for windows cluster? I currently run with SQST, ALB on but I got many error messages from time to time. It is very slow(300~700ms per read) during backup and it caused disk missing once. Please help.(Please see attachment for full text)

------

In system event log:

HP MPIO DSM for EVA4x00/6x00/8x00 family of Disk Arrays is attempting an operation on \Device\MPIODisk2. The Type is noted in the dump data.

An unrecoverable path failure occurred on SCSI address (3.0.4.3). Disk 6001438005DED1EC0000D00000F80000 is still accessible over redundant path(s).

The DSM has completed remove processing for path (SCSI address (3.0.4.3)) to multipath capable disk 6001438005DED1EC0000D00000F80000.

------

In sql server errorlog:

2011-08-11 17:11:06.97 spid11s SQL Server has encountered 1 occurrence(s) of I/O requests taking longer than 15 seconds to complete on file [j:\TCPDATA2\TCPDATA10.ndf] in database [TCP] (5). The OS file handle is 0x0000000000001FF0. The offset of the latest long I/O is: 0x00003054d6e000

2011-08-11 17:11:34.83 Backup Error: 3203, Severity: 16, State: 1.

2011-08-11 17:11:34.83 Backup Read on "J:\TCPLOG3\TCPLOG3.ldf" failed: 1167(The device is not connected.)

2011-08-11 17:11:34.83 Backup Error: 3203, Severity: 16, State: 1.

2011-08-11 17:11:34.83 Backup Read on "j:\TCPDATA2\TCPDATA7.ndf" failed: 1167(The device is not connected.)

2011-08-11 17:11:34.83 spid1836 Internal I/O request 0x000000B22589CAA0: Op: ReadDatabase, pBuffer: 0x000001027E360000, Size: 458752, PageNumber: 8:1929880, SOS: Internal: 0xC000009D, InternalHigh: 0x0, Offset: 0xAE530000, OffsetHigh: 0x3, m_buf: 0x000001027E360000, m_len: 458752, m_actualBytes: 0, m_errcode: 1167, File: j:\TCPDATA2\TCPDATA7.ndf

------

In cluster log:

00001ac0.0000342c::2011/08/11-09:11:39.712 ERR [RES] Physical Disk <Cluster Disk 4>: IsAlive sanity check failed!, pending IO completed with status 1167.

00001ac0.0000342c::2011/08/11-09:11:39.712 ERR [RES] Physical Disk <Cluster Disk 4>: IsAlive sanity check failed!, pending IO completed with status 1167.

00001ac0.0000342c::2011/08/11-09:11:39.712 WARN [RHS] Resource Cluster Disk 4 IsAlive has indicated failure.

00000bf8.00002a94::2011/08/11-09:11:39.712 INFO [RCM] HandleMonitorReply: FAILURENOTIFICATION for 'Cluster Disk 4', gen(0) result 1.

00000bf8.00002a94::2011/08/11-09:11:39.712 INFO [RCM] TransitionToState(Cluster Disk 4) Online-->ProcessingFailure.

00000bf8.00002a94::2011/08/11-09:11:39.712 ERR [RCM] rcm::RcmResource::HandleFailure: (Cluster Disk 4)

00000bf8.00002a94::2011/08/11-09:11:39.712 INFO [RCM] resource Cluster Disk 4: failure count: 1, restartAction: 2.

00000bf8.00002a94::2011/08/11-09:11:39.712 INFO [RCM] Will restart resource in 500 milliseconds.

00000bf8.00002a94::2011/08/11-09:11:39.712 INFO [RCM] TransitionToState(Cluster Disk 4) ProcessingFailure-->[WaitingToTerminate to DelayRestartingResource].

00000bf8.00002a94::2011/08/11-09:11:39.712 INFO [RCM] rcm::RcmGroup::UpdateStateIfChanged: (SQL Server (MSSQLSERVER), Online --> Pending)

00000bf8.00002a94::2011/08/11-09:11:39.712 INFO [RCM] TransitionToState(SQL Server) Online-->[WaitingToTerminate to OnlineCallIssued].

00000bf8.00002a94::2011/08/11-09:11:39.712 INFO [RCM] TransitionToState(SQL Server Agent) Online-->[WaitingToTerminate to OnlineCallIssued].

00000bf8.00002a94::2011/08/11-09:11:39.712 INFO [RCM] TransitionToState(SQL Server Agent) [WaitingToTerminate to OnlineCallIssued]-->[Terminating to OnlineCallIssued].

00001624.00003ab0::2011/08/11-09:11:41.194 INFO [RES] SAP Resource <SAP TCP 00 Instance>: LooksAlive request.

00000bf8.000034dc::2011/08/11-09:11:43.799 INFO [RCM] HandleMonitorReply: TERMINATERESOURCE for 'SQL Server Agent', gen(0) result 0.

00000bf8.000034dc::2011/08/11-09:11:43.799 INFO [RCM] Restarting resource 'SQL Server Agent'.

00000bf8.000034dc::2011/08/11-09:11:43.799 INFO [RCM] TransitionToState(SQL Server) [WaitingToTerminate to OnlineCallIssued]-->[Terminating to OnlineCallIssued].

00001bb4.0000085c::2011/08/11-09:11:43.815 INFO [RES] SQL Server <SQL Server>: [sqsrvres] SvcTerminateProcess: Terminated the sqlserver process (processID = 2610) explicitly.