On my system (running Windows 7 x64) I have an Asus Z87 Deluxe motherboard, with a built-in Intel RAID controller; my config includes a Kingston 240-GB SSD disk for the OS and two Western Digital 4-TB HDDs for data. I purposefully bought two data disks in order to mirror them, because I want to minimize the risk of data loss on this system. All disks are 6-GB SATA.
I configured the two disks as a RAID 1 array using the onboard Intel controller, and the system has been running with this config for a while; however, it has always beens awfully slow, and I struggled with it for almost a year, before finally breaking the RAID and reconfiguring the controller for standard disk access, in order to be able to troubleshoot that incredible slowness (the WD diagnostic tool is unable to talk with RAIDed disk, it needs direct access to the physical disks).
As soon as I broke the RAID, lots of disk errors started popping up in the system event log, and the WD diagnostic tool immediately discovered one of the disks was faulty; it's still under warranty, thus I called WD for a replacement which is presently shipping. The other disk is fine, thus no data loss occurred.
However, there's a big problem here: the system has been running for almost a year with a faulty disk, and not only the Intel RAID controller completely failed to diagnose this, but it also actually worsened the situation by hiding the disk errors from the OS, which had no way to know a disk was failing.
Over the course of this troublesome year, I tried all the released version of the RST software (which also has plenty of other bugs), up to the latest one (13.2.4.1000); no one of them changed this behaviour or brought any performance improvement (which I was assuming was caused by the RAID config or controller, and not by a faulty disk, of which I had no knowledge at all).
Why did the controller fail to diagnose a faulty disk?
And how can I make sure it actually warns me if a disk is failing, if I configure it for RAID mode again?