Warning: WD Red WD60EFRX-68L0BN1

[Chris Moore]

You have a pool of 60 6TB WD Red Pro drives?

At work, that is just one of many servers.

You have another pool with 64 4TB WD Red's and an unspecified number of Seagate drives?

Also at work, that system has 80 drives, 62 are WD Red (not pro) drives and the last 16 are Seagate Constellation drives.

Sorry if I didn't make it clear. I work in a facility where one of my job roles is to manage storage servers. We are expecting to receive two more servers with 60 drives each in the near future. The contract allowed up to 60 days for delivery, so they should be here by the end of September. Anyhow, one is going to have 10TB drives and the other will have 12TB drives. I plan to put FreeNAS on those to see how it runs, but if I have any trouble with it, I will have to use Linux instead. These systems for work must be rock solid and FreeNAS has had some trouble with the larger drives, so I will be monitoring them closely during the burn-in process.

My primary server at home has 36 drives but it is broken up into four different pools and my backup server has 12 drives in a single pool.

 

[tobiasbp]

These systems for work must be rock solid and FreeNAS has had some trouble with the larger drives, so I will be monitoring them closely during the burn-in process.

Can you tell me more about this? I'd really like to know. I'm in the process of upgrading 6TB drives to 12TB ones.

 

[DaveY]

Have you checked the firmware to see if they're all the same? I'm more inclined to think it's related to firmware than the submodel number (part after the dash) All my drives are WD60EFRX-68L0BN1 and I've never had a single Scrub error. Last successful scrub was Aug 6. The drives have been running constantly for over a year. Another thing you can do to rule out the drive model is to swap places with a drive where you have a WD60EFRX-68MYMN1 just to rule out controller/position

My Drives:
Model Family: Western Digital Red
Device Model: WDC WD60EFRX-68L0BN1
Firmware Version: 82.00A82

 

[Chris Moore]

Can you tell me more about this? I'd really like to know. I'm in the process of upgrading 6TB drives to 12TB ones.

I went looking for the posts so I could point at them, but I couln't find them. Sometimes the forum search tool doesn't do as good a job as I wish it did.
There were a variety of posts from several users that were having problems with drives getting too many errors and being dropped out of the pool. The drives were 10 and 12 TB models but I don't remember all the details of the situation. One of the posters installed Linux using ZFS on Linux (ZoL) and had no problems with the hardware. It was theorized that there was a firmware or driver issue that was at fault, but I don't think any answer was discovered. This may still be an issue with the larger drives. I suppose I will find out when I start testing those new servers.

 

[Earthling]

To add to the mystery of these errors... while I never got them before and only had one or two occasional zpool errors that ZFS promptly fixed (running raid-z2), all six of my WD60EFRX-68MYMN1 drives (02/2015 manufacture, 03/2015 purchase) have begun throwing "pre-fail" and "old-age" status codes simultaneously, while only the sixth drive is throwing genuine SMART error codes during short and long smartctl scans (long scans were scheduled monthly, scrubs weekly).

I missed that this was happening for a couple weeks as I didn't have emails configured after the FreeNAS 10.x -> 11.1.x upgrade and was also generally using the new UI which apparently doesn't show SMART or zpool errors yet like the old one. Also, vacation, and other travel engagements. Anyway, once I noticed the errors, I promptly contacted Western Digital and they eventually agreed to send me 6x new RMA drives (which, ironically, will be WD60EFRX-68L0BN1 revisions). It should also be noted that, apparently, WD only warranties the RMA drives for 3 months and they're likely to be certified refurbs rather than new manufacture. This worries me somewhat and I'm considering upgrading to Seagate Exos or Ironwolf 12TBs.

 

[Borja Marcos]

A very, very silly question. Did you get all the drives from the same vendor at the same time?

Back in the early 90's I suffered a series of disk failures that could only be attributed to a sloppily handled package at a store. I purchased one drive and installed everything only to have it fail in a week. After taking it back to the store and replacing it I had another failure after several days and so on. I think it was four drives or so. The fifth "replacement" was actually my money back. When I asked about the origin of the drives they confirmed that all of them had arrived in the same shipment.

And I had another comparable problem with a stupid dealer that insisted on packaging them improperly. So we installed everything and some drives failed after a month or so. I had to resort to heavyhanded measures like issuing an order to our warehouse staff: any package containing a single poorly packaged hard drive was to be rejected as "improperly packaged" no matter what was the ratio between hard disks and other stuff. Ie, a package with $10,000 dollars in other stuff and a single poorly packaged hard disk was rejected on the spot. It took four or five packages and some phone calls for them to learn.

I offered them the alternative to pay for the costs of a failed hard disk after working for a month or so at the customer's premises. Rebuilding everything from backup, travel expenses, and a compensation to the customer.

Finally they begun to pack the disks properly and those mysterious failures ended.

 

[Earthling]

A very, very silly question. Did you get all the drives from the same vendor at the same time? [...]

I picked them up in March 2015 (5x from NewEgg and 1x from Amazon).

 

[Chris Moore]

It should also be noted that, apparently, WD only warranties the RMA drives for 3 months and they're likely to be certified refurbs rather than new manufacture.

The drives I have gotten lately from WD under warranty checked out to have the same amount of warranty remaining as the drive they were replacing.

 

[Earthling]

The drives I have gotten lately from WD under warranty checked out to have the same amount of warranty remaining as the drive they were replacing.

I will call them again to ask specifically about that but I was told they would be refurbished drives with a three month warranty. The person that told me that was a level one tech, so may have been mistaken. As of now, I've only received three of the six (two are refurb with a black label, one looks to be new manufacture with expected WD Red label) as UPS didn't collect them all into same delivery truck. So that part, at least, has checked out.

I've got a few days of resilvering ahead of me but at least I'm running raidz2.

 

[caveman]

Hey all.

Running WD EX4100 and just found a disk error. drives are WD60EFRX-68L0BN1 like what seems to be a common thread note here.

I am planning on RMA the drive, but have a question. Is it possible to upgrade to the 6TB RED Pro drive and swap just one drive for the 6TB RED or will there be issues related to drive speed and other factors that might corrupt my RAID? Or would I need to swap all drives to upgrade to the RED Pro drives.

Thanks

 

[tobiasbp]

Is it possible to upgrade to the 6TB RED Pro drive and swap just one drive for the 6TB RED ....

Yes, that is what I have done.

 

[tobiasbp]

Have you checked the firmware to see if they're all the same?

Looking through my notes, all the drives throwing errors have/had firmware v. 82.00A82. This is the firmware used on all of my WD60EFRX-68MYMN1 and WD60EFRX-68L0BN1 drives.

Another thing you can do to rule out the drive model is to swap places with a drive where you have a WD60EFRX-68MYMN1 just to rule out controller/position

Please read my original thread for more information.

 

[Dacesilian]

I've got 4 drives in 2-way mirror and also got errors from scrub. Yes, I have the same firmware as mentioned.
Device Model: WDC WD60EFRX-68L0BN1
Firmware Version: 82.00A82

 

[Sidewinder]

I just finished building my replacement NAS. It has 6x WD60EFRX-68L0BN1 with the same firmware and I have not received any SCSI errors during scrubs or otherwise. [Edit 2] This is after going through a burn-in of S.M.A.R.T tests and badblocks. Also, I am not going through any backplane.

Code:

root@artemis[~]# for x (da0 da1 da2 da3 da4 da5); do smartctl -a /dev/$x | egrep 'Device Model|Firmware'; done
Device Model:     WDC WD60EFRX-68L0BN1
Firmware Version: 82.00A82
Device Model:     WDC WD60EFRX-68L0BN1
Firmware Version: 82.00A82
Device Model:     WDC WD60EFRX-68L0BN1
Firmware Version: 82.00A82
Device Model:     WDC WD60EFRX-68L0BN1
Firmware Version: 82.00A82
Device Model:     WDC WD60EFRX-68L0BN1
Firmware Version: 82.00A82
Device Model:     WDC WD60EFRX-68L0BN1
Firmware Version: 82.00A82

The complete info on my build is in the signature.

Edit: One thing to note (though I don't think it would make a difference) is that I haven't yet copied my data from the old NAS to the new one. So my scrubs (thus far) are only with a small amount of data in the pool.

Edit 2: Just received an IOC Fault 0x40002622 during a reboot. I'm investigating.

 

[tobiasbp]

Just for the hell of it, I have made a new zpool of WD60EFRX-68L0BN1 drives just to see if I could confirm the issues. I could: On the 1st scrub I get errors. This is on the same machine and same slots were i run a bunch of WD Golds with out issue.

 

[raito88]

Just for the hell of it, I have made a new zpool of WD60EFRX-68L0BN1 drives just to see if I could confirm the issues. I could: On the 1st scrub I get errors. This is on the same machine and same slots were i run a bunch of WD Golds with out issue.

I have also issue Freenas Version 11.2 with this HDD:
WD NAS RED PRO - 2TB - 7200rpm
Device Model: WDC WD2002FFSX-68PF8N0
Firmware Version: 81.00A81

P.S.
With preivous Freenas v10 - No problem with this HDD