asr woes.

[Dan Rue]

Recently I had a machine running an Adaptec (asr) raid 5 array lose one
drive and subsequently crash.  After dealing with the failure, I have
brought the machine in question into testing and have discovered that I
receive read errors in degraded mode.  Lots of em: 

Oct 26 05:37:44 whiteandnerdy kernel: g_vfs_done():da0s1e[READ(offset=3725403922432, length=16384)] error = 5
Oct 26 05:37:44 whiteandnerdy kernel: g_vfs_done():da0s1e[READ(offset=3725403889664, length=16384)] error = 5
Oct 26 05:37:44 whiteandnerdy kernel: g_vfs_done():da0s1e[READ(offset=3725404168192, length=16384)] error = 5
Oct 26 05:37:44 whiteandnerdy kernel: g_vfs_done():da0s1e[READ(offset=3723171880960, length=16384)] error = 5

The card is an Adaptec 2100S:
asr0: <Adaptec Caching SCSI RAID> mem 0xfc000000-0xfdffffff irq 24 at device 1.0 on pci2
asr0: ADAPTEC 2110S FW Rev. 380E, 1 channel, 256 CCBs, Protocol I2O

The system is FreeBSD 6.1-RELEASE with an SMP kernel (i386).

Note that I havn't seen any write errors.  Could this be a hardware
problem or is it more likely an asr bug?

Also related, it would be really nice if asr-utils worked (I've tried
everything) :-/

Dan