"Western Digital admits 2TB-6TB WD Red NAS drives use shingled magnetic recording" from Blocks & Files

[Patrick M. Hausen]

My distributor accepted to RMA my recent order of 4 WD Red 3 TB EFAX. I could order WD Red 4 TB EFRX instead. They have about 100 of them still on stock. I would have preferred Toshiba N300, but these are hard to get currently. 2 weeks or so. I ordered 5 EFRX so I have one spare. Let's hope they'll last until they get replaced by larger drives in a couple of years.

ALSO Germany, in case anyone is interested.

Kind regards
Patrick

 

[colmconn]

On WD Red NAS Drives - Western Digital Corporate Blog

On WD Red NAS Drives - Data - Western Digital Corporate Blog

blog.westerndigital.com

 

[Yorick]

Right. If they fixed the firmware bug, and were very upfront in their data sheets about what drives are DM-SMR, I’d even follow their argument.

But going “resilver / raid rebuild failing is not a bug for a 8 drive small NAS enclosure; and no, we do not tell our customers whether drives are DM-SMR”, that is not a good look.

 

[maclark7029]

Just replaced a failed 2014 WD20EFRX with a new WD40EFAX at ada5. Now getting these errors that I've never seen before. Related to this SMR speed issue?

Apr 21 08:41:48 server siisch7: Timeout on slot 30
Apr 21 08:41:48 server siisch7: siis_timeout is 00000000 ss 70000000 rs 70000000 es 00000000 sts 801e0000 serr 00000000
Apr 21 08:41:48 server siisch7: ... waiting for slots 30000000
Apr 21 08:41:49 server siisch7: Timeout on slot 29
Apr 21 08:41:49 server siisch7: siis_timeout is 00000000 ss 70000000 rs 70000000 es 00000000 sts 801e0000 serr 00000000
Apr 21 08:41:49 server siisch7: ... waiting for slots 10000000
Apr 21 10:32:39 server (ada5:siisch7:0:0:0): WRITE_FPDMA_QUEUED. ACB: 61 08 e0 7c c9 40 70 00 00 00 00 00
Apr 21 10:32:39 server (ada5:siisch7:0:0:0): CAM status: ATA Status Error
Apr 21 10:32:39 server (ada5:siisch7:0:0:0): ATA status: 41 (DRDY ERR), error: 10 (IDNF )
Apr 21 10:32:39 server (ada5:siisch7:0:0:0): RES: 41 10 e0 7c c9 00 70 00 00 00 00
Apr 21 10:32:39 server (ada5:siisch7:0:0:0): Retrying command
Apr 21 13:51:49 server siisch7: Timeout on slot 30
Apr 21 13:51:49 server siisch7: siis_timeout is 00040000 ss 70000000 rs 70000000 es 00000000 sts 801e0000 serr 00000000
Apr 21 13:51:49 server siisch7: ... waiting for slots 30000000
Apr 21 13:51:49 server siisch7: Timeout on slot 29
Apr 21 13:51:49 server siisch7: siis_timeout is 00040000 ss 70000000 rs 70000000 es 00000000 sts 801e0000 serr 00000000
Apr 21 13:51:49 server siisch7: ... waiting for slots 10000000
Apr 21 13:51:59 server siisch7: Timeout on slot 28
Apr 21 13:51:59 server siisch7: siis_timeout is 00040000 ss 70000000 rs 70000000 es 00000000 sts 801e0000 serr 00000000
Apr 21 13:51:59 server (ada5:siisch7:0:0:0): FLUSHCACHE48. ACB: ea 00 00 00 00 40 00 00 00 00 00 00
Apr 21 13:51:59 server (ada5:siisch7:0:0:0): CAM status: Command timeout
Apr 21 13:51:59 server (ada5:siisch7:0:0:0): Retrying command
Apr 21 13:51:59 server (ada5:siisch7:0:0:0): DSM TRIM. ACB: 06 01 00 00 00 40 00 00 00 00 01 00
Apr 21 13:51:59 server (ada5:siisch7:0:0:0): CAM status: Command timeout
Apr 21 13:51:59 server (ada5:siisch7:0:0:0): Retrying command
Apr 21 13:51:59 server (ada5:siisch7:0:0:0): FLUSHCACHE48. ACB: ea 00 00 00 00 40 00 00 00 00 00 00
Apr 21 13:51:59 server (ada5:siisch7:0:0:0): CAM status: Command timeout
Apr 21 13:51:59 server (ada5:siisch7:0:0:0): Retrying command
Apr 21 14:12:26 server (ada5:siisch7:0:0:0): WRITE_FPDMA_QUEUED. ACB: 61 08 08 74 e9 40 70 00 00 00 00 00
Apr 21 14:12:26 server (ada5:siisch7:0:0:0): CAM status: ATA Status Error
Apr 21 14:12:26 server (ada5:siisch7:0:0:0): ATA status: 41 (DRDY ERR), error: 10 (IDNF )
Apr 21 14:12:26 server (ada5:siisch7:0:0:0): RES: 41 10 08 74 e9 00 70 00 00 00 00
Apr 21 14:12:26 server (ada5:siisch7:0:0:0): Retrying command
Apr 21 19:05:22 server (ada5:siisch7:0:0:0): WRITE_FPDMA_QUEUED. ACB: 61 10 b8 f2 1e 40 71 00 00 00 00 00
Apr 21 19:05:22 server (ada5:siisch7:0:0:0): CAM status: ATA Status Error
Apr 21 19:05:22 server (ada5:siisch7:0:0:0): ATA status: 41 (DRDY ERR), error: 10 (IDNF )
Apr 21 19:05:22 server (ada5:siisch7:0:0:0): RES: 41 10 b8 f2 1e 00 71 00 00 00 00
Apr 21 19:05:22 server (ada5:siisch7:0:0:0): Retrying command

 

[proto]

What about 3TB?

Actually I own:
- 6 3tb WD RED WD30EFRX-68EUZN0 -- in a zraid2 pool
- 2 3tb WD RED WD30EFRX-68N32N0 -- still unused, untested, out of pool.

They are quite old, but they are "different" in smartctl output, like in 4TB series.

smartctl -l gplog,0x30,9 /dev/daX

Code:

--- WD30EFRX-68EUZN0
smartctl 7.0 2018-12-30 r4883 [FreeBSD 11.3-RELEASE-p7 amd64] (local build)
Copyright (C) 2002-18, Bruce Allen, Christian Franke, www.smartmontools.org

General Purpose Log 0x30 does not exist (override with '-T permissive' option)

--- WD30EFRX-68N32N0
smartctl 7.0 2018-12-30 r4883 [FreeBSD 11.3-RELEASE-p7 amd64] (local build)
Copyright (C) 2002-18, Bruce Allen, Christian Franke, www.smartmontools.org

General Purpose Log 0x30 has only 9 sectors, output skipped

Attached:
smartctl -x /dev/daX smartctl --identify=wb /dev/daX

 

[Yorick]

@maclark7029 You are resilvering? You are seeing the dreaded IDNF errors, the firmware bug that WD insists is not a firmware bug. With other DM-SMR drives, resilver would eventually finish, after 5 days or so. This one, it’ll fail out of the vdev.

There might be ways around this - turn off write cache, a couple other tweaks, and waiting for even longer - and, you want no part of this. If you can, return the drive, and get something else.

 

[Johnnie Black]

WD30EFRX

These are all CMR, only the WD30EFAX is SMR.

 

[proto]

These are all CMR, only the WD30EFAX is SMR.

Sure they are. I'd like to see if there are further differences between newest 3TB EFRX and mine.
I kindly ask @Tempted and @Patrick M. Hausen if they can share their smartctl as soon as the drives arrive.

Just for reference.

I must admit that as soon as I read that ticket published on the smartmontools TRAC I was quite scared ... now I am relieved.

 

[Patrick M. Hausen]

No problem, I will post. Just note that I ordered 4 TB EFRX for my returned 3 TB EFAX - reason again being availability.

 

[Tempted]

...I kindly ask @Tempted and @Patrick M. Hausen if they can share their smartctl as soon as the drives arrive.

Gonna be at least 3 weeks until I get the other components to commence the build, but happy to once it's running.

Drives coming today though :)

 

[rudds]

Is anyone going to bother seeking replacement from WD for an EFAX drive?

I have 4x6TB EFRXs in my FreeNAS box and bought a a fifth last July from Amazon as a cold spare for when one of these drives eventually fails, but only discovered last week that drive turned out to be an EFAX. Obviously far too late to send back to Amazon and I doubt my credit card can help either, but I'm not optimistic about convincing WD to do anything at this point other than tell me to get over it. Since I bought this drive for the sole purpose of resilvering it into my existing array, which it's explicitly incapable of doing, I'm stuck with a $200 paperweight if they won't do anything for me.

It's almost hard to believe the arrogance and deceit that led to this situation, but here we are. WD will never get another dime of my money if they don't plan to do anything to address this.

 

[Yorick]

I have 4x6TB EFRXs in my FreeNAS box and bought a a fifth last July from Amazon as a cold spare for when one of these drives eventually fails, but only discovered last week that drive turned out to be an EFAX. Since I bought this drive for the sole purpose of resilvering it into my existing array, which it's explicitly incapable of doing, I'm stuck with a $200 paperweight

Their blog post says that wd pro are for these kind of workloads, and that anyone with concerns should get in touch with their support.

Reading between the lines, if that drive Is still in original packaging, they might let you return it if you can show you have bought a wd pro as replacement. Worth a try to see whether they’d help out in that way.

 

[rudds]

Their blog post says that wd pro are for these kind of workloads, and that anyone with concerns should get in touch with their support.

Reading between the lines, if that drive Is still in original packaging, they might let you return it if you can show you have bought a wd pro as replacement. Worth a try to see whether they’d help out in that way.

I understand what you're saying, but WD claiming that putting a Red into a four-drive FreeNAS system already populated with Reds is "a workload far exceeding its intended use" is so absurd it's infuriating. That post is nothing but marketing doublespeak papering over the fact that they cut costs and delivered an inferior product that can't do what it used to do, then lied about it, then told longtime customers it was their fault for expecting to get what they paid for. I'm not sure I've ever been so angry at a company, let alone one whose products I've spent thousands of dollars on over the years.

 

[Yorick]

@rudds Oh I get what you’re saying. I wasn’t commenting on the merits of their argument. I was trying to suggest a pragmatic approach that might cost you less than scrapping the drive altogether.

And, given your understandable anger, that approach may not be open to you, if WD had ever even offered it, which we don’t know yet. That leaves you with buying a drive from another company, and eBaying the one you have.

 

[Cantrepeat]

Was there a point/date at which they changed drives to this not so good product? Or is it all 2/4/6 REDs suffer from this fault?

I just upgraded my FreeNAS mini to 4x4TB WD40EFRX in Feb 2020.

On Amazon, this is WDs exact listing.

• Specifically designed for use in NAS systems with up to 8 bays
• Supports up to 180 TB/yr workload rate Rate is defined as amount of user data transferred to or from drive. Workload Rate is annualized (TB transferred times (8760/recorded power-on hrs)). Workload Rate will vary on hardware/software components
• NASware firmware for compatibility
• Small and home office NAS systems in a 24/7 environment
• 3-year manufacturer's limited warranty

I feel bamboozled by WD.

 

[Patrick M. Hausen]

Was there a point/point at which they changed drives to this no so good product? Or is it all 2/4/6 REDs suffer from this fault?
I just upgraded my FreeNAS mini to 4x4TB WD40EFRX in Feb 2020.

From what we know EFRX are fine, EFAX are SMR ...

 

[elorimer]

Newegg has the 4TB EFRX for $90 shipped today. On Amazon you can get a certified refurbished one for $145!

 

[Cantrepeat]

Newegg has the 4TB EFRX for $90 shipped today. On Amazon you can get a certified refurbished one for $145!

Got a link? It comes up $118 for me.

Also, amazon comes up $109 for new EFRX - https://www.amazon.com/gp/product/B00EHBERSE/ref=ppx_yo_dt_b_search_asin_title?ie=UTF8&psc=1

 

[elorimer]

Promo:

Newegg.com - Get It Now! $89.99 WD Red 4TB NAS 3.5" Hard Disk Drive

 

[Cantrepeat]

Thanks,

Code - EMCDHFE24