Running multiple drives in JBOD mode under heavy I/O resulted in SCSI error

Problem

Running multiple drives in JBOD mode under heavy I/O resulted in the following SCSI error messages: 

Aug 26 10:58:42 asl142 kernel: megaraid: aborting-2408937 cmd=2a 
Aug 26 10:58:42 asl142 kernel: megaraid abort: scsi cmd:2408937, do now own
Aug 26 10:58:42 asl142 kernel: megaraid: aborting-2408938 cmd=2a 
Aug 26 10:58:42 asl142 kernel: megaraid abort: scsi cmd:2408938, do now own
Aug 26 10:58:42 asl142 kernel: megaraid: aborting-2408946 cmd=2a 
Aug 26 10:58:42 asl142 kernel: megaraid abort: scsi cmd:2408946, do now own

Configuration

Omega 8220SR, MegaRAID 320-1LP (FW 1L37, BIOS G119), 6 Seagate Cheetah 10K.7

Fedora Core 3/64, 2.6.11-13_FC3, megaraid_mbox v. 2.20.4.5

Solution

This is a driver bug which currently has no fix. Under normal operating environment, the disks will be configured in RAID array(s) which does not trigger the SCSI error.