6

This is not about DRIVE failure. It's about drive CONFIG failure.

I bought 3 brand new drives for my server, because existing ones have worked for over 4 years and one of them is failing (shows ECC ERROR or DEGRADE). I'm always able to rebuild array with existing drives, but replacing them all is my immediate goal.

I checked specs of these old drives, they are WD2003FYYS, 512 bytes per sector. I thought it would be proper to also acquire 2TB drives with 512n format, so I now have Seagate ST2000NM0055.

THE PROBLEM

As seen on the following picture, all three new drives behave exactly the same - 3ware /LSI 9750-8i does not want to detect these properly. Just for kicks I inserted one desktop drive, WD10EZEX with Advanced Format (4K sectors) and it showed status of OK. All 8 drives are in hot-swap bays, so I did a lot of swapping, also tried different bays - no change. Controller's event log does not even show "drive connected" event, as it did for WD10EZEX Drive config failure

All Seagates are brand new and have no partition on them (WD10EZEX had one). They work properly in my workstation, connected to eSATA port. CrystalDiskInfo says they're all OK. (lang enter image description here

QUESTION

Has anybody encountered this type of behavior? How do I deal with this? I've read docs of 9750-8i but didn't find anything about drive config failure, let alone what to do in this case. It's not like I'm a newbie when it comes to various RAIDs and I had 5 different 3ware controllers to this day, but in the case at hand I'm stumped. Please help.

Kitet
  • 378
  • 2
  • 12

1 Answers1

0

I had a similar issue when replacing what appeared to be a failing drive on a 3Ware 9650SE controller.

When replacing the drive with a Seagate ST2000DM008-2FR102 drive, I was seeing "config failure", which I couldn't clear.

Eventually, we reseated all the SATA connections, and the problem went away. So it is possible for this to be caused by a slightly loose / dirty SATA cable

Michael Firth
  • 131
  • 1
  • 8