I recently built a new server and have three Supermicro CSE-M35T-1B drive cages and two Supermicro AOC-SASLP-MV8 cards.  The issue seems to be with one drive in one slot, no matter what I do.  I've gotten the same error with different drives and changed out cables.  The only thing I haven't swapped out is the SAS card.  No matter what I do, /dev/md7 will not format.  I have run a preclear on all drives before trying to format, however, I get the following when I try:

 

Oct 12 12:34:01 ZEUS emhttp: shcmd (267): set -o pipefail ; mkreiserfs -q /dev/md7 |& logger
Oct 12 12:34:01 ZEUS logger: mkreiserfs 3.6.21 (2009 www.namesys.com)
Oct 12 12:34:01 ZEUS logger: 
Oct 12 12:34:01 ZEUS logger: 
Oct 12 12:34:01 ZEUS logger: The problem has occurred looks like a hardware problem. If you have
Oct 12 12:34:01 ZEUS logger: bad blocks, we advise you to get a new hard drive, because once you
Oct 12 12:34:01 ZEUS logger: get one bad block that the disk drive internals cannot hide from
Oct 12 12:34:01 ZEUS logger: your sight,the chances of getting more are generally said to become
Oct 12 12:34:01 ZEUS logger: much higher (precise statistics are unknown to us), and this disk
Oct 12 12:34:01 ZEUS logger: drive is probably not expensive enough for you to you to risk your
Oct 12 12:34:01 ZEUS logger: time and data on it. If you don't want to follow that follow that
Oct 12 12:34:01 ZEUS logger: advice then if you have just a few bad blocks, try writing to the
Oct 12 12:34:01 ZEUS logger: bad blocks and see if the drive remaps the bad blocks (that means
Oct 12 12:34:01 ZEUS logger: it takes a block it has in reserve and allocates it for use for
Oct 12 12:34:01 ZEUS logger: of that block number). If it cannot remap the block, use badblock
Oct 12 12:34:01 ZEUS logger: option (-B) with reiserfs utils to handle this block correctly.
Oct 12 12:34:01 ZEUS kernel: quiet_error: 23 callbacks suppressed
Oct 12 12:34:01 ZEUS kernel: Buffer I/O error on device md7, logical block 0
Oct 12 12:34:01 ZEUS kernel: Buffer I/O error on device md7, logical block 1
Oct 12 12:34:01 ZEUS kernel: Buffer I/O error on device md7, logical block 2
Oct 12 12:34:01 ZEUS kernel: Buffer I/O error on device md7, logical block 3
Oct 12 12:34:01 ZEUS kernel: Buffer I/O error on device md7, logical block 4
Oct 12 12:34:01 ZEUS kernel: Buffer I/O error on device md7, logical block 5
Oct 12 12:34:01 ZEUS kernel: Buffer I/O error on device md7, logical block 6
Oct 12 12:34:01 ZEUS kernel: Buffer I/O error on device md7, logical block 7
Oct 12 12:34:01 ZEUS kernel: Buffer I/O error on device md7, logical block 8
Oct 12 12:34:01 ZEUS kernel: Buffer I/O error on device md7, logical block 9
Oct 12 12:34:01 ZEUS logger: 
Oct 12 12:34:01 ZEUS logger: bread: Cannot read the block (0): (Input/output error).
Oct 12 12:34:01 ZEUS logger: 
Oct 12 12:34:02 ZEUS emhttp: _shcmd: shcmd (267): exit status: -122
Oct 12 12:34:02 ZEUS emhttp: shcmd (268): :>/etc/samba/smb-shares.conf
Oct 12 12:34:02 ZEUS emhttp: shcmd (269): cp /etc/netatalk/AppleVolumes.default- /etc/netatalk/AppleVolumes.default
Oct 12 12:34:02 ZEUS emhttp: Restart SMB...
Oct 12 12:34:02 ZEUS emhttp: shcmd (270): killall -HUP smbd
Oct 12 12:34:02 ZEUS emhttp: shcmd (271): ps axc | grep -q rpc.mountd
Oct 12 12:34:02 ZEUS emhttp: _shcmd: shcmd (271): exit status: 1
Oct 12 12:34:02 ZEUS emhttp: Restart AFP...
Oct 12 12:34:02 ZEUS emhttp: shcmd (272): killall -HUP afpd
Oct 12 12:34:02 ZEUS emhttp: shcmd (273): /usr/local/sbin/emhttp_event svcs_restarted
Oct 12 12:34:02 ZEUS emhttp_event: svcs_restarted

 

Any ideas?

If the issue's always with drives in the same slot, then clearly you have either a bad slot in the cage or a bad port on the controller.

 

You've tried swapping cables -- now try connecting a different port to that slot.  If it works, you have a bad slot on your controller;  if not, the slot is bad -- RMA the cage.

 

Attach a SMART report.