Problem with RAID setup

S

spitz66

n00b
Joined
Jan 29, 2010
Messages
8
I have a RAID-5 array that consists of five 1.5TB disks (WD15EADS).
The controller is Areca ARC-1220 and TLER is activated on all disks.

Sometimes the array works just fine and I get descent speed in both read/write but quite often there is serious problem with performance, both on read/write and access time.

I just cant figure out what's going on and I have had this problem on both XP (x86) and Windows 7 (x86) with both scsiport and storport driver. Latest Areca firmware (1.48) as well.
All disks seems fine and don't drop out of the array.
I have tried just about everything I can think of with different stripe-size, cache-mode, LBA64, 4K, SATA mode and so on. :(

The Areca card sits in a PCIe x16 slot on a DFI Lanparty JR T2RS motherboard.
More details of the system below.
Any ideas? :confused:

Volume Set Information
Volume Set Name ARC-1220-VOL#00
Raid Set Name Raid Set # 00
Volume Capacity 6000.0GB
SCSI Ch/Id/Lun 0/0/0
Raid Level Raid 5
Stripe Size 128KBytes
Block Size 512Bytes
Member Disks 5
Cache Mode Write Back
Tagged Queuing Enabled

IDE Channels
Ch01 Raid Set # 00 1500.3GB WDC WD15EADS-00P8B0
Ch02 Raid Set # 00 1500.3GB WDC WD15EADS-00P8B0
Ch03 Raid Set # 00 1500.3GB WDC WD15EADS-00P8B0
Ch04 Raid Set # 00 1500.3GB WDC WD15EADS-00R6B0
Ch05 Raid Set # 00 1500.3GB WDC WD15EADS-00R6B0
Ch06 N.A. N.A. N.A.
Ch07 N.A. N.A. N.A.
Ch08 N.A. N.A. N.A.

IDE Drive Information
IDE Channel 1
Model Name WDC WD15EADS-00P8B0
Serial Number WD-WMAVU0163258
Firmware Rev. 01.00A01
Disk Capacity 1500.3GB
Current SATA Mode SATA300+NCQ(Depth32)
Supported SATA Mode SATA300+NCQ(Depth32)
Disk APM Support No
Device State Normal
Timeout Count 0
Media Error Count 0
SMART Read Error Rate 200(51)
SMART Spinup Time 185(21)
SMART Reallocation Count 200(140)
SMART Seek Error Rate 200(0)
SMART Spinup Retries 100(0)
SMART Calibration Retries 100(0)
 
Did you let the array complete rebuild itself? Use some kind of utility to see what the status of the array is i.e., healthy, rebuilding, paused, etc.
 
1. Try disabling NCQ on the controller [most likely cause].
2. Are you sharing OS partition and media partition on the same volume? Should perform better if OS was separate drive (or RAID 1 set).
3. SMART Reallocation Count is non-zero; you might have bad sectors.
4. You're using WD green drives; they slow their rotational speed to reduce heat.

Did you see this thread?
http://www.allquests.com/question/1241461/Areca-ARC-1220-Seeking-Stalling-Read-Issues.html
 
Thanks for the info,however:

1. Tried that without any difference
2. OS is located on another disk (SSD).
3. This looks the same on all disks but I really don't know how it works. Worth digging into?
4. Yes but then this stalling and staggering wouldn't happen randomly in the middle of an filecopy or so? Furthermore I ran the WDIDLE tool to prevent this or am I wrong?

I have read that but maybe I missed something of interest, quite long thread..;)
 
I'm right now in the process of running an extended test for each HDD in the array with the program HD sentinel.
This software seems to be one of the few that can identify and run tests with all HDD still attached to the RAID card
Are there any downsides with this procedure so that I should disconnect each HDD and test with for example Western Digitals own software instead?
 
Extended test went ok for all disks no errors found.

However when I performed a random seektest there were some strange results.

Sometimes the test went ok with an average seek time of ~9ms and sometimes the testresult was not so good with an average seek time of ~600ms. :eek:

The same pattern can be seen for all HDD and not a specific drive.
I attach som screenshots below on the result.

I could really need some advice/tips/tricks here....! :confused:

HDD2_fast.jpg


HDD2_slowseek.jpg
 
Green Drives are just dogs. I noticed the same type of problems when I used them as well.
 
danman said:
Green Drives are just dogs. I noticed the same type of problems when I used them as well.
Click to expand...

So I just have to live with it then or buy some new drives? :(

On the other hand I've heard others who are happy with their drives and it will be an expensive lesson to learn to buy new ones.

Maybe it's the controller choking up for som reason?
Guess i'll have to test the disks from the motherboards SATA connections to be sure then.
 
If the array is not vital, or in production, then try recreating the issue with a smaller array - 3 drives would work.

If you do get the issue once again with the original set of drives, pickup 3 replacement drives, and see if you can reproduce the issue with the replacement drives.

You won't be the first one to make a mistake with hard drive selection... I have done it, and I kept this link cause the guy sounds like what would happen to me!

Surviving The Perfect Storm… The Data Protection Strategy Paid Off!

Happy Thanksgiving!

L8ter...
 
This guy from another thread had to enable write back on his RAID controller, and now he seems happy with the green drives. But I see you already did this....

I feel like you're missing something with the computer, but I have no idea what. I also wonder if your whole computer slows down when the drives do. Is there any chance the slowdown problem is caused by something unrelated to the array?
 
Just about anything is possible for the moment... ;)

I have now deleted the RAID-5 and created pass-through disks. Still the same hickups!
At least for the WD15EADS-00P8B0 disks. Haven't had time to verify all yet.
 
I have 3 WD15EADS-00P8B0 drives in my software raid5 in linux, and i have had performance issue since i bought them over 1 year ago, and i recently figured out that it is a problem with this model. They get hickups all the time. i haven't found any solution to the problem yet and i will have to boy new drives because these are totaly crap. They get hickups even if they are not in a raid config also.
I have mailed WDC but they dont even answer my email. I will never buy a WD disk anymore if they cant even answer the mail.
Are going to buy 3 new drives from another manufacture if i cant get new drives on waranty.
 
Ok, a little update here...

I was lucky to be able to RMA the disks to the dealer (although they were more than an year old) and just pay the difference to some new 2TB Hitachi drives.

So now everything is up and running normal with good performance! :)

Never more WD (at least not "consumer drives" in RAID environment...)
 
You must log in or register to reply here.
Back
Top