RE4 WD5003ABYX Drives dropping from RAID with Adaptec 3805 [ Edited ]

[Lampiarz]

My four brand new RE4 WD5003ABYX (FW 01.01S01) drops randomly from RAID at Adaptec 3805 (the newest firmware).

It drops at 1,5Gb/s mode too.

I replaced cables, and did test with fresh operating system (win7 32 bit). Nothing helps.

How it can be? RE means Raid Edition. I had the same problem with desktop drives so I switched to enterprise disks.

What can I do now, please help?

Best regards Chris

[Nesvic · Accepted solution]

Maybe you should try contacting WD's Technical Support about this. You can do so either by phone or email.

To Contact WD for Technical Support
http://support.wdc.com/contact/index.asp?lang=en

[Lampiarz]

Nesvic - thanks for your advice, I did as you said.

I have also tested cache mode on/ff. NCQ on/off, RAID 0, 1, 5. It still happens :-(

Let's hope WD support will help me :-)

[fzabkar]

Perhaps the drives' R/W TLER timeouts are set to higher values than the RAID controller's own error recovery timeouts??? If so, then this would mean that your drives will still be retrying bad sectors after the RAID controller has given up waiting for them.

WD has a WDTLER utility to set these timeout values. Alternatively you may be able to reconfigure your RAID.

[Lampiarz]

I discover TLER in my RE4 is set to 7 sec. My old ES.2 is set to 7 sec and it worked fine for years.

My three Barracudas desktop drives ST3500418AS (FW CC35 and CC37) with TLER set to 0 sec also works fine.

Two very old Barracudas 7200.7 ST380817AS never dropped from RAID with this Adaptec.

One IBM-ESXS MAX3073RC (73GB, 15k) as SAS drives is working perfectly.

I think I can say that my controler is not the problem here.

Can you imagine how I was suprised seeing RE4 dropping all the time (one drive/hour).

Seller refused me to back WD drives. Am I doomed  with them?

I hope not :-)

Some light in tunnel : http://wdc.custhelp.com/app/answers/detail/a_id/1493/session/L3RpbWUvMTMxMTUwMjQ5Ny9zaWQvRU1vQU5Nems...

Seems other drives got similar problem and WD resolved that.

[Lampiarz]

In the meantime, I tried to check out IntelliPark status with wdidle3.exe. Tools reported that this feature is disabled on all disks. So it is not the cause of error.

[datatomb]

6 drives in Raid5EE configuration, at least one drive drops from Raid after a few days. WD20EARS-00M does not behave so. Adaptec says, 3TB are no problem for 51245, maybe a firmware problem of WD30EZRX-00M ?

[fzabkar]

Have you confirmed whether the WD30EZRX supports TLER or ERC?

[datatomb]

How can I check this ?

[fzabkar]

I expect that it should be safe to run WDTLER without any arguments. That should report the settings of any Read and Write Timeouts.

Otherwise you could use CrystalDiskInfo to capture the drive's Identify Device information block. These data contain the drive's settings, including whether or not it supports ERC.

http://crystalmark.info/software/CrystalDiskInfo/index-e.html

First go to Edit -> Copy Option and select Identify Device. Then go to Edit -> Copy. This will copy the disk information to the clipboard. Then open up NotePad or whichever word processor ships with your version of Windows, and choose Edit -> Paste (or Ctrl-V) to input the clipboard data.

If bit #3 of byte 0x19D is a 1, then the drive supports ERC.

See section 7.16.7.71 of the following document:

Working Draft AT Attachment 8 - ATA/ATAPI Command Set (ATA8-ACS):
http://www.t13.org/documents/UploadedDocuments/docs2008/D1699r6a-ATA8-ACS.pdf