hba puts disk into closed status

[pgroce]

DL380 G6 has an hba with 9 disks. At random times the HBA puts the status of the disk into a closed state. Its an ibm 2145 san. We have to reboot the server for the disk to work again. Is there any history of the proliant servers having hba problems like this?

There were some events in the windows log

All paths have failed. \Device\MPIODisk9 will be removed.

A fail-over on \Device\MPIODisk9 occurred.

The server was unable to allocate from the system nonpaged pool because the pool was empty

The device, \Device\Harddisk9\DR9, is not ready for access yet.

The operating system returned error 21(The device is not ready.) to SQL Server during a write at offset 0x00000985b30000 in file 'm:\data\tempdb.mdf'.  (repeated many times until reboot)

We are not using MPIO its all handled by the hba.