I have only spotty details; it appears a drive failure occurred and the firmware decided to fault some of the other drives too, instead of going into failover mode or starting a rebuild on the hot spares. Some "sananigans" performed by the storage group was able to recover most of the data including my 4.8 million files/250Gb, the rest from backup.
Could be buggy firmware; i believe it is years old. End user decided to buy an enterprise SAN, hook new devices up to it year after year, then not commit to maintaining the firmware 8(
Gear is Dell/EMC CX700
I am modifying my original advice.
Quote:
If you put all your eggs into one basket for storage, make sure the basket maker is Hitachi Data Systems
A little tidbit from a coworker regarding marketing/cost versus the "technically correct" solutions. Hitachi makes SATA drives which Hitachi Data Systems will not install into their enterprise SAN storage due to their MTBF ratings; however these drives are provided by Dell/EMC for installation in Dell/EMC kit. One little "incident" evaporates any cost savings. I can believe that this "philosophy" can easily find its way further into their engineering decisions.
We've had 2 SAN issues with 2 different EMC SANs in two weeks; (issues meaning failed drives not handled correctly)
All our Hitachi arrays have been running years with no issues.
edit: ZDnet has an excellent article on why SATA drives will be the death of RAID5
http://blogs.zdnet.com/storage/?p=162
