RocketRaid 2320 and Segate Drives (question and issue)

Hey guys,

I've got an ASUS P5KC board with an Intel chipset -- but more importantly I have a PCI-X HighPoint RocketRaid 2320 controller with 8-channels.

I've got the raid setup in two arrays, both are RAID 0.

I've got a few partitions and the drives, one has Windows Vista 64-bit and the other has Windows XP 32-bit... both are running without any significant errors as I can tell.

I have three Seagate DB35.3 Series™ SATA 3Gb/s 320-GB Hard Drive (http://www.seagate.com/ww/v/index.jsp?vgnextoid=b3327ea70fafd010VgnVCM100000dd04090aRCRD&locale=en-US), and two Barracuda 7200.10 SATA 3.0Gb/s 320-GB Hard Drive (http://www.seagate.com/ww/v/index.jsp?vgnextoid=2d1099f4fa74c010VgnVCM100000dd04090aRCRD&locale=en-US) drives... so this breaks down into the following on the controller.

Channel 1
- ST3320820SCE

Channel 2
- ST3320620AS

Channel 3
- nothing

Channel 4
- ST3320820SCE

Channel 5
- nothing

Channel 6
- ST3320820SCE

Channel 7
- nothing

Channel 8
- ST3320620AS

Now I have heard some weird noises at times from the very beginning of having this machine (about 1.5-2 years ago) coming from I believe one of the drives -- it sounds like a really high pitched beeping sound sometimes when the drive is really in use (such as defrag, etc). It happens more frequent at times than others, I guess just depends on data load.

Earlier today on the XP system I was looking at the event log and saw the following...
System Event Log
WARNING TYPE
An error was detected on device \Device\Harddisk1\D during a paging operation.

There was about 100 of these, occuring every few seconds.


I also (after seeing these errors) decided to install the HighPoint RAID Management Console to check it out. Under device management I see the devices, and all drives have "Read Ahead", "Write Cache", and "NCQ" enabled.

When I go to view S.M.A.R.T. details on the drives, all of them "PASS" except for the one on channel 8, which is a ST3320620AS -- it fails on the metric "Reallocation Sector Ct" -- not entire sure what it means, but below are the details...

The threshold for this metric is "36", and the value & the worst are "1" -- so it failed.

This is the only metric on this drive that has failed, and this is the only drive that has a S.M.A.R.T. error. There hasn't been any data loss as far as I can tell. The only thing I do know of are the errors in event viewer, and the noise I mentioned earlier.

What exactly does this mean, and do you know of any reason why it may be?

This metric on the other drives is at "100" on all of them.



Another thing is dealing with the "Read Ahead", "Write Cache", and "NCQ" -- do I need to have all of those items enabled on all drives, or are there reasons why I should disable some? (note, this is a home box used for development and gaming).

Thanks,
Travis

Yep, I meant PCI-E.

Also, the thing is that this has been making this sound for a long time, every since I got the machine -- and more than likely the S.M.A.R.T. report has been the same as well, and there has been no data loss.

So really not sure.

I think if I do re-do the array's I'll make it either RAID 1 or RAID 5... but something new on the POST of the Raid Controller when I first start, it takes about 30 seconds to recognize the arrays -- is that normal?

You originally said both arrays were RAID 0.

If it is infact a RAID 1 then if one drive failed you would not get a bluescreen or any loss of data. If it is RAID 0 which it sounds like it is then you'd have to restore from a backup or take the drive to an expensive data recovery place.