Western Digital gets sued for sneaking SMR disks into its NAS channel

[Pickles von Brine]

Quote

All three of the surviving conventional hard drive vendors—Toshiba, Western Digital, and Seagate—have gotten caught sneaking disks featuring Shingled Magnetic Recording technology into unexpected places recently. But Western Digital has been the most brazen of the three, and it's been singled out for a class action lawsuit in response.

 

Although all three major manufacturers quietly added SMR disks to their desktop hard drive line-up, Western Digital is the only one so far to slip them into its NAS (Network Attached Storage) stack. NAS drives are expected to perform well in RAID and other multiple disk arrays, whether ZFS pools or consumer devices like Synology or Netgear NAS appliances...

This was bound happen. As to why they decided it was a great idea to sneak SMR onto NAS drives I am not quite sure... Money?  Regardless, I wonder how many people though they were fine, went to rebuild only to have their raid take a dump. Being a former data recovery engineer, I am not sure how much this would affect data recovery as well.. Maybe it doesn't have much bearing but having a failed raid that is in the middle of a rebuild doesn't end well in a recovery scenario. Either way, I digress. This is just shady in my eyes and they are likely to loose a bunch of money. 

Source

Edited May 30, 2020 by Pickles - Lord of the Jar
Added source

[rawrdaysgoby]

RIP, 

[mr moose]

I predict this will fail. They would have to prove that SMR was a problem and that it was marketed or advertised as something other than SMR first.  I can't see either of those happening given the testing done so far shows minimal if any issues with DM SMR.

[wanderingfool2]

24 minutes ago, mr moose said:

I predict this will fail. They would have to prove that SMR was a problem and that it was marketed or advertised as something other than SMR first.  I can't see either of those happening given the testing done so far shows minimal if any issues with DM SMR.

This will be an interesting case to at least follow I think.  I have a feeling that this may end up failing (if they only take the SMR approach to it).  I actually suspect the issues that people had with their SMR drives failing during raid rebuilds might have been due to firmware issues (maybe related to their SMR implementation).

 

I think for a NAS grade drive, SMR is unacceptable and shouldn't exist in it...but all that WD would need to show was that SMR has benefits in NAS harddrives.  (I think SMR drives actually had a better read speed or something similar to that).  If they could prove that, then they could likely just use the defense that they were targeting that performance metric

[wanderingfool2]

10 minutes ago, valdyrgramr said:

I think the most interesting part is how they plan to prove it in a court of law.   This goes beyond your average techies knowledge, and most court systems in the US are made up of non-techies.   You then have a well known company who can easily bs a court, and has the money for a pro-bsing legal team.   On top of that, the court would likely question the background of the people suing this company.  That's literally why I love following these class action lawsuits in America.

Seagate's statements on NAS grade drives could be a large factor, and given that provable performance hit taken during rebuilds and that the change occurred so silently doesn't bode well for WD.  The fact is rebuilding on RAID's are a major feature (and given how badly this affected the performance in a crucial area of RAID, I think WD should be held accountable)...with that said, I think it will be all about WD trying to up playing the other benefits and claim it was a decision made.

 

Either way, I hope WD fails at this.

[leadeater]

45 minutes ago, wanderingfool2 said:

The fact is rebuilding on RAID's are a major feature (and given how badly this affected the performance in a crucial area of RAID

Well the thing is for the one person so far to actually do proper testing to see how this was affected it actually wasn't by a much. For a 90%-95% full array with mixed drives a rather under resourced lower spec NAS had rebuild times 60% slower and the high spec NAS has rebuild times 40% slower. Now where this will, and why this will fail in court, is that the actual times are not unreasonable and 10TB+ disks that are CMR/PMR have longer rebuild times than 6TB DM-SMR disks (smaller size even quicker) so it's going to fail the customer impact assessment.

 

It would have as much chance at success as suing because WD Red (FRX) has slower rebuild times than WD Red Pro because they sold 'NAS rated disks' with only ~5400 RPM and I the customer deem that to not be 'NAS rated'. Or suing because a 10TB disk takes longer to rebuild than a 6TB disk. Any and all technical arguments are going to be so easily countered and brushed away by a mountain of counter technical evidence and it'll just stall the case until the court just dismisses it.

 

Literally the only case they have here is the non-disclosure of DM-SMR, anything else is irrelevant and will fail. Opinions on DM-SMR and it's NAS suitability just don't matter, for this.

[mr moose]

1 hour ago, valdyrgramr said:

Not really.   This is in sue happy USA with non-techie courts.   It comes down to who the better bser is.

Your right, this is exactly what happened when the courts decided that glyphosate was cancerous in the face of evidence from just about every university, the FDA, the EPA and independent biologists from all over the world.  Not too mention all the testing and basic data on the subject and the fact there was evidence the plaintiff's lawyer had been trying to backmail Monsanto. Woo hoo justice...

 

 

1 hour ago, wanderingfool2 said:

This will be an interesting case to at least follow I think.  I have a feeling that this may end up failing (if they only take the SMR approach to it).  I actually suspect the issues that people had with their SMR drives failing during raid rebuilds might have been due to firmware issues (maybe related to their SMR implementation).

 

I think for a NAS grade drive, SMR is unacceptable and shouldn't exist in it...but all that WD would need to show was that SMR has benefits in NAS harddrives.  (I think SMR drives actually had a better read speed or something similar to that).  If they could prove that, then they could likely just use the defense that they were targeting that performance metric

 

Personally, the video's I watched where they tested the drives in various NAS boxes doing most of the common tasks we expect of a NAS, for me the rebuild time was a worthwhile concession for the extra storage per dollar.  Even the SMR drives int he consumer cuda's do not have enough of a real world impact to deter me from buying them knowingly.   Not everyone will feel the same granted, but given these drives have been in circulation for a while now and people are only just starting to realize after a blog from someone who had theirs fail,  I dare say majority of users were not impacted enough to be able to prove it in court. 

[Mihle]

3 hours ago, mr moose said:

I predict this will fail. They would have to prove that SMR was a problem and that it was marketed or advertised as something other than SMR first.  I can't see either of those happening given the testing done so far shows minimal if any issues with DM SMR.

 

9 days rebuild on SMR drive while it only took 17 hours or whatever it was on CMR drives on ZFS. The exstra time is very bad, makes it much higher chance that another drive will fail under the rebuild.

 

Also, some Synology users have had rebuild outright fail. (It wouldn't have if the NAS knew it was SMR but it didn't, at least not at the time)

[Mihle]

1 hour ago, mr moose said:

Your right, this is exactly what happened when the courts decided that glyphosate was cancerous in the face of evidence from just about every university, the FDA, the EPA and independent biologists from all over the world.  Not too mention all the testing and basic data on the subject and the fact there was evidence the plaintiff's lawyer had been trying to backmail Monsanto. Woo hoo justice...

 

 

 

Personally, the video's I watched where they tested the drives in various NAS boxes doing most of the common tasks we expect of a NAS, for me the rebuild time was a worthwhile concession for the extra storage per dollar.  Even the SMR drives int he consumer cuda's do not have enough of a real world impact to deter me from buying them knowingly.   Not everyone will feel the same granted, but given these drives have been in circulation for a while now and people are only just starting to realize after a blog from someone who had theirs fail,  I dare say majority of users were not impacted enough to be able to prove it in court. 

Most videos out there that try to test with an empty NAS, and then ofc you don't have problems. It's something else if the NAS is actually in use.

[leadeater]

7 minutes ago, Mihle said:

9 days rebuild on SMR drive while it only took 17 hours or whatever it was on CMR drives on ZFS.

This impact is unique to ZFS, where WD actually targets these disks to (SMB NAS) don't really have a problem.

 

8 minutes ago, Mihle said:

Also, some Synology users have had rebuild outright fail. (It wouldn't have if the NAS knew it was SMR but it didn't, at least not at the time)

Other tests show no problems with that at all so who knows why theirs failed, rebuilds can fail it's just very rare.

[leadeater]

3 minutes ago, Mihle said:

Most videos out there that try to test with an empty NAS, and then ofc you don't have problems. It's something else if the NAS is actually in use.

 

1 hour ago, leadeater said:

For a 90%-95% full array with mixed drives a rather under resourced lower spec NAS had rebuild times 60% slower and the high spec NAS has rebuild times 40% slower

 

[mr moose]

18 minutes ago, Mihle said:

 

9 days rebuild on SMR drive while it only took 17 hours or whatever it was on CMR drives on ZFS. The exstra time is very bad, makes it much higher chance that another drive will fail under the rebuild.

 

Also, some Synology users have had rebuild outright fail. (It wouldn't have if the NAS knew it was SMR but it didn't, at least not at the time)

As leadeater pointed out we all know/knew there was a ZFS problem, you can find a more in depth and better tested set of videos in the original SMR thread in the news section.

[leadeater]

1 hour ago, mr moose said:

As leadeater pointed out we all know/knew there was a ZFS problem, you can find a more in depth and better tested set of videos in the original SMR thread in the news section.

That's the problem with Reddit, it's typically a good place to see examples of confirmation bias. An issue present in one thing doesn't make it present in all things and user reports of a problem in a limited scale doesn't mean it's a wide scale issue. Definitive statements need to be evidenced which nobody is doing over on Reddit, in particular at the initial time of reports.

 

Somebody, preferably multiple, actually needs to take the time and effort to document testing of a potential problem so we know if it is a problem and under what situations. Is it any configuration with X software or Y configuration with X software etc. To what degree is the issue, what is the impact etc.

 

For this legal case there is a reason it is a complaint under advertising laws, at lot of the technical aspects don't make or break the case and flawed assumptions of consumers don't either. It's going to come down to whether or not listing DM-SMR on the product sheet is materially relevant to consumers, this is where it gets tricky and where I think all the 'lawyering' is going to happen, and by virtue leaving that out mislead or deceived consumers. Argument crafting is important here because if done well there is actually a decent chance of winning.

 

The problem I see is that ZFS is the only use case and software I have seen where the impact goes below the bar of acceptable product behavior and consumer expectation, within this there are two points of argument. First is that if everything else does not have an issue then does the fault actually lie with ZFS and is not a product fault with the hard drive. Second is that if I were WD I would be asking for evidence from the plaintiffs in regards to their hardware configuration and use case. I know full well, WD probably does too, that most of the people using ZFS and complaining on Reddit are above the stated recommendation by WD of 8 disks for WD Red so these consumers knowingly went outside the products stated recommendations and design.

 

So you have an issue that as far as the evidence I have seen is only relevant to ZFS and also a high probably consumers operating the product outside of the manufactures recommendations, as a not lawyer I'd have to say care needs to be taken here.

 

Another issue I see is around whether or not it was materially relevant therefore required to disclose the use of DM-SMR, had it played out that there was no performance difference or better (in the cases where it is for the new disk revision) and no software or hardware had an issue with it is it therefore required under law to put on the product sheet, I don't think that would be the case. This is not a question of opinion whether people think DM-SMR is good or not or if they think it would have or always will have a performance impact this it the "If you cannot to detect it and has no impact on consumer usage is it legally required to notify it's usage". Consumers not being able to tell isn't always a defense either, far from it, but neither is not being told something about a product misleading or deceptive behavior.

 

Affected users can and should have just used WD support and returned the product as unfit for purpose and got replacements, WD were offering this and to upgraded product lines at not cost. The rest can be dealt with using standard market influence, stop buying WD Red and recommend others not to. It's a competitive market with other good options in the same price category, better ones I might add, so just buy those. WD will either notice this and be impacted by it and change the product or it won't impact them and they won't.

 

Given that cases like these tend to favor the consumer I think I'd give it a 60/40 shot at winning.

[mr moose]

8 minutes ago, leadeater said:

Consumers not being able to tell isn't always a defense either, far from it, but neither is not being told something about a product misleading or deceptive behavior.

 

 

I still think this would be a big argument in court.  Part of the case will be proving there were adverse effects from the products directly related to DM SMR,  if they did not experience those effects it is going to be very difficult to prove they existed when no measurement or benchmark was presented in any of the advertising material with reference to the performance difference between DM SMR and any other form of MR. 

 

 

Oh, and I agree completely with reddit being an echo chamber for confirmation bias.  The bit that annoys me the most about reddit is that everyone there thinks they are smart and that they have their finger on the pulse. As a result they don't listen to anyone else or research beyond what satisfies their immediate grievance.  

 

 

[Kilrah]

The SMR aspect won't be the determinant one for the court, they'll just see that a drive advertised for a purpose was not fit for said purpose, the tech details are secondary.

[Master Disaster]

5 hours ago, mr moose said:

I predict this will fail. They would have to prove that SMR was a problem and that it was marketed or advertised as something other than SMR first.  I can't see either of those happening given the testing done so far shows minimal if any issues with DM SMR.

Depends on the wording of the suit. If they try to sue for false advertising they'll probably lose, if they go for misleading customers they'll probably win.

[leadeater]

9 minutes ago, Kilrah said:

The SMR aspect won't be the determinant one for the court, they'll just see that a drive advertised for a purpose was not fit for said purpose, the tech details are secondary.

And WD's counter to that is ZFS is not all NAS's so not working with ZFS doesn't make it not fit for purpose. If I buy a QNAP and throw these disks in it it will work fine, if it mix the disks that also works fine. I can do the same with a Netgear NAS, Seagate NAS, Synology NAS etc. The disks are still worse than every other option on the market but a not fit for purpose argument is going to be hard because DM-SMR purposefulness is nothing but opinion based.

[Master Disaster]

Turns out SMR is actually demonstrably bad for NAS drives.

[leadeater]

Just now, Master Disaster said:

Turns out SMR is actually demonstrably bad for NAS drives.

*Sigh*, ONLY for ZFS.

[Master Disaster]

Just now, leadeater said:

*Sigh*, ONLY for ZFS.

Fair.

[jagdtigger]

2 minutes ago, leadeater said:

*Sigh*, ONLY for ZFS.

Which is still NAS application..... There is no such thing as "NAS drive except ZFS".

[Kilrah]

7 minutes ago, leadeater said:

And WD's counter to that is ZFS is not all NAS's so not working with ZFS doesn't make it not fit for purpose.

But then the advertising material must let you know that this drive is not suitable for ZFS, whether by mentioning either just that or that it's SMR. At this point mentioning it's SMR would probably be best since saying "not suitable for ZFS" might bite them later if another FS has similar issues.

[leadeater]

Just now, jagdtigger said:

Which is still NAS application..... There is no such thing as "NAS drive except ZFS".

Well then add ZFS to the defendant list because I can show ZFS is not fit for purpose as a NAS using a single use case under an single hardware configuration.... 

[jagdtigger]

Just now, leadeater said:

Well then add ZFS to the defendant list because I can show ZFS is not fit for purpose as a NAS using a single use case under an single hardware configuration.... 

FreeNAS, and they actually sell neat boxes that fit for consumers just like any other NAS box. Busted......

[leadeater]

30 minutes ago, Kilrah said:

But then the advertising material must let you know that this drive is not suitable for ZFS, whether by mentioning that it's SMR or just that it isn't.

That's one of the problems, hindsight. I actually don't think WD tested the disk under ZFS so had no idea how bad it was for it. Sure that's just bad product testing but it's also not like WD target WD Red to that user base compared to the off the shelf NAS buyers and the NAS vendors that OEM WD Reds in their products. I think all testing went in to that and only that honestly.

 

That's why the consumer being mislead is the most likely argument to win, because the manufacture not being aware that it would affect the consumer or not isn't a defense for that.