Jump to content
Search In
  • More options...
Find results that contain...
Find results in...

Western Digital's Red 2 - 6TB NAS drives apparently aren't good for NAS use?

2 hours ago, leadeater said:

image.png.f319dbed1912cb2cff844dd3af3f8511.png

He meant major model or series aka "Red". Literally no one for home usage is buying drives based on these model numbers. No one. You only do this when you're looking for something ridiculously specific like I was chasing specs for WD UltraStar drives with HelioSeal and without instant secure wipe function but with 4Kn support. And you had to dig through bunch of specs to find the exact model number on which you then look for these exact models in stores. Because if you just search for "WD UltraStar" bunch of websites don't even show you all of them or they do all of them and then you wonder why one drive of same series and capacity costs almost 70€ more. It's that instant secure wipe thing for example. 4Kn also made it more expensive. And often these tiny details are not mentioned in online shops, I had to rely on vendor specs. But when vendor specs only list you useless things like SATA 6 Gb/s, you of course pick the one that's cheaper. It's been bunch of times that drives literally had no noteworthy differences which is why no one gives a damn about model numbers. But in this case, there was a huge difference inside. Not just firmware with tiny changes but whole different platter recording method and slower write speeds. That's a big deal. If people were warned ahead of time, they'd be careful about this and if spec sheet just had one extra actually useful like that just said "Recording method" and one would have CMR and another SMR and we wouldn't be discussing this at all.

AMD Ryzen 7 5800X | ASUS Strix X570-E | G.Skill 32GB 3733MHz CL16 | PALIT RTX 3080 10GB GamingPro | Samsung 850 Pro 2TB | Seagate Barracuda 8TB | Sound Blaster AE-9 MUSES 01/03 | Altec Lansing MX5021 Nichicon FineGold/MUSES 8920

Link to post
Share on other sites
30 minutes ago, RejZoR said:

That's a big deal. If people were warned ahead of time, they'd be careful about this and if spec sheet just had one extra actually useful like that just said "Recording method" and one would have CMR and another SMR and we wouldn't be discussing this at all.

And the other side of that is the disk was designed for the task and supposedly tested to meet that and passed so mentioning the change wasn't required and only those people that do check spec sheets would of seen it, rest would have purchased either way.

 

Knowing Reds changed to DM-SMR ahead of time wasn't going to let know that it will have issues with ZFS for example. We are discussing a reported issue with the current revision of the disk yes, like I keep saying it has not been proven DM-SMR is the root cause of the problem nor the technology itself fundamental to not being suited for NAS usage, yet people are saying both.

 

If WD wants to change WD Red over to DM-SMR and the end product is fit for purpose the fact that DM-SMR is being used is actually inconsequential. It only matters to you and others right now because you believe that is the cause of problems.

 

Had there been no issues it could of been years before anyone noticed. You are retrospectively applying an alertness to a problem you could not possibly be aware of.

 

Your huge difference really may not be at all, considering the fact that for a lot of operations the DM-SMR disk is significantly faster too due to the large cache and fast PMR zone.

seagate_archive_8tb_sata_4k_randomtransf

Large cache drives destroy PMR disks with small caches, even the most expensive 7200RPM Near-Line. 

 

WD Red is not a pure SMR only disk, it utilizes both technologies.

Link to post
Share on other sites
2 hours ago, leadeater said:

If it has no actual bearing on the suitability of the product and it doing what it is being sold to do you are going to have a very hard time arguing any legal case. You don't get told the brand and spec information of the drive motor and actuator either. You don't even get told with the controller chip is for that matter, SSDs tell you that.

 

Wanting something listed and it needing to be are quite different things. Withholding non critical information to protect either competitive information or to protect product image is actually something you are allowed to do. Information that is currently supplied is detailed as it is, wanting more, needing more, requiring more entail quite different things.

Then they lose me as a customer. That is the thing. Complain all you like about me being wrong. Or those customers being wrong. If there is no relationship with the seller, the buyer might walk.

'Customers might argue SMR should not be used in a NAS and it is the cause of the faults, they might be wrong' (To paraphrase)

So? Two wrongs don't make a right. The consumers have a failing product and find out a material change in the design, a functional difference in the processing of it's performance or function, is different. That is wrong on the case of WD.

 

As said, like with SSDs, if I had a Samsung EVO 860 xxxzzyA and an EVO 860 xxxzzyB and found out the Bs had no RAM cache, instead using extra/better designed FLASH cache, I'd be annoyed *if* those drives also failed/froze/hit bottlenecks at different times to the As.

 

If I was told in the spec sheet "Cache, 32mb RAM A and 32mb FLASH B" I'd not care, I'd go in with knowledge of my purchase.

 

Here in the UK, we have "chip shop condiment". It's vinegar flavour (chemically identical, but other source of acid), but not vinegar (grape/malt/cider sourced), thus they don't call it the same thing. Customers don't care, because they are not being wilfully deceived.

Link to post
Share on other sites
2 hours ago, Mihle said:

Toshiba also have desktop HDDs that is SMR, but the spec sheet does not mention it.

I never said I don't want SMR. Again... lol. I said companies not advertising specs/technology of the product may lose customers and is a poor move. I look up the Toshiba drives I get, I also don't use NAS yet. But I do try to find out if it's SMR/cache sizes/etc. If I find a Toshiba drive I don't like the reviews/spec to, I won't buy it. Same for the Samsung drives.

 

But somehow, people not liking the WD drives is wrong of them... even if these drives are failing in NAS?

Link to post
Share on other sites
25 minutes ago, TechyBen said:

Then they lose me as a customer. That is the thing. Complain all you like about me being wrong. Or those customers being wrong. If there is no relationship with the seller, the buyer might walk.

'Customers might argue SMR should not be used in a NAS and it is the cause of the faults, they might be wrong' (To paraphrase)

So? Two wrongs don't make a right. The consumers have a failing product and find out a material change in the design, a functional difference in the processing of it's performance or function, is different. That is wrong on the case of WD.

Then you are walking for ill-informed reasons so it's not two wrongs at all, it's you just not buying something because you have a perception about something which isn't correct.

 

Now if you are saying you don't want to buy a current WD Red because of reported issues and are not satisfied they are fixed, great excellent reasoning. If you are in a situation where you have actually purchased one and it's not working for your needs and you wish to return it, do so, and you also do not wish to purchase any more WD Red onging also great reasoning.

 

What's not great reasoning is taking that further and placing the blame on something you don't actually know is the cause.

 

You don't have to make ill-informed assumptions and throw around blame needlessly, forever tainting your opinion of a technology that may have nothing to do with the issue.

 

You don't have to assume anything at all, that's the point. If you don't know what the issue is and lack the evidence then simply do not blame it. Just wait out the evidence because I can guarantee there will be people(s) testing this right now.

 

It's not a material change if it's not a material change, why can't you just wait on evidence. You can still choose to not buy the disks and advise others do not while waiting. 

Link to post
Share on other sites
3 hours ago, TechyBen said:

 

But somehow, people not liking the WD drives is wrong of them... even if these drives are failing in NAS?

I never said that, I am on your side in that.

Its just that you talked like Toshiba did disclose it in the spec sheet always, while they arent. I was just trying to be devils advocate.
Seagate surfer said Ironwolf are not SMR, and Toshiba NAS drives are not SMR, both do desktop drives with SMR without saying it anywhere, so to me it looks like they are in the same boat. While WD is worse because they do it to the WD Reds.
Unless Seagate does the same thing as WD, I will go Seagate over WD in the future.

“Remember to look up at the stars and not down at your feet. Try to make sense of what you see and wonder about what makes the universe exist. Be curious. And however difficult life may seem, there is always something you can do and succeed at. 
It matters that you don't just give up.”

-Stephen Hawking

Link to post
Share on other sites
27 minutes ago, Mihle said:

I never said that, I am on your side in that.

Its just that you talked like Toshiba did disclose it in the spec sheet always, while they arent. I was just trying to be devils advocate.
Seagate surfer said Ironwolf are not SMR, and Toshiba NAS drives are not SMR, both do desktop drives with SMR without saying it anywhere, so to me it looks like they are in the same boat. While WD is worse because they do it to the WD Reds.
Unless Seagate does the same thing as WD, I will go Seagate over WD in the future.

Well, Toshiba's standard desktop drives are the P300s.   So, it's would most likely be mixed up in there.   I believe one of the links clarified that.

 

https://blocksandfiles.com/2020/04/16/toshiba-desktop-disk-drives-undocumented-shingle-magnetic-recording/ Found it, was in another thread.

Home AI/Game Development, and Music Editing Workstation:

Bloodshed and the Fenris-Wolf:  https://pcpartpicker.com/list/Fx3sNP

 

Mobile Workstation:

Ryu Hayabusa:  https://pcpartpicker.com/list/37bVz7

 

Link to post
Share on other sites

Skipped the last page and a bit as it felt like things where a bit circular so this is more of a general broad comment.

 

Depending on quite a few factors that would really need a full legal court case to decide on customers in the EU for which the downsides of SMR specifically would be an issue could still have WD and the others bent over a barrel on the issue. If your product, either as a brand or as a general product class has a perception of certain features/capabilities/e.t.c in the public consciousness of the customers buying that product and you don't meet all those expectations you are required to state that it doesn't meet those expectations somwhere, depending on how major an expectation it is there's also i believe occasions where you can't bury it in the small print it has to be pretty clearly stated in a more front leading way than would be the case with small print. Now weather NAs drives as a whole or WD Red as a brand has created a perception that there won't be SMR is way above my pay grade. But in potentia, they could still be in some real trouble in europe.

Link to post
Share on other sites
4 hours ago, Mihle said:

I never said that, I am on your side in that.

Its just that you talked like Toshiba did disclose it in the spec sheet always, while they arent. I was just trying to be devils advocate.
Seagate surfer said Ironwolf are not SMR, and Toshiba NAS drives are not SMR, both do desktop drives with SMR without saying it anywhere, so to me it looks like they are in the same boat. While WD is worse because they do it to the WD Reds.
Unless Seagate does the same thing as WD, I will go Seagate over WD in the future.

Nah. Me liking Toshiba and Samsung is historical, and if they follow suit, I won't buy those products (but do realise they are massive companies... so if Sony Music make DRM breaking cds, I may still buy Sony headphones ;) , but won't buy Sony music/cds etc).

 

I'll go back to a company if I can trust it. But really. I don't want the hassle, so I avoid those companies that give me hassle.

Link to post
Share on other sites
2 hours ago, CarlBar said:

Skipped the last page and a bit as it felt like things where a bit circular so this is more of a general broad comment.

 

Depending on quite a few factors that would really need a full legal court case to decide on customers in the EU for which the downsides of SMR specifically would be an issue could still have WD and the others bent over a barrel on the issue. If your product, either as a brand or as a general product class has a perception of certain features/capabilities/e.t.c in the public consciousness of the customers buying that product and you don't meet all those expectations you are required to state that it doesn't meet those expectations somwhere, depending on how major an expectation it is there's also i believe occasions where you can't bury it in the small print it has to be pretty clearly stated in a more front leading way than would be the case with small print. Now weather NAs drives as a whole or WD Red as a brand has created a perception that there won't be SMR is way above my pay grade. But in potentia, they could still be in some real trouble in europe.

I don't even think it requires a downside. Sometimes it does, other times it's just if the customer was mislead, or if the customer was not told, of certain specifics. See the AMD case, which they did list specs, but made marketing claims they could not back up (IIRC which was the only reason they lost/settled out of court for that one).

Link to post
Share on other sites
7 hours ago, leadeater said:

Then you are walking for ill-informed reasons so it's not two wrongs at all, it's you just not buying something because you have a perception about something which isn't correct.

 

Now if you are saying you don't want to buy a current WD Red because of reported issues and are not satisfied they are fixed, great excellent reasoning. If you are in a situation where you have actually purchased one and it's not working for your needs and you wish to return it, do so, and you also do not wish to purchase any more WD Red onging also great reasoning.

 

What's not great reasoning is taking that further and placing the blame on something you don't actually know is the cause.

 

You don't have to make ill-informed assumptions and throw around blame needlessly, forever tainting your opinion of a technology that may have nothing to do with the issue.

 

You don't have to assume anything at all, that's the point. If you don't know what the issue is and lack the evidence then simply do not blame it. Just wait out the evidence because I can guarantee there will be people(s) testing this right now.

 

It's not a material change if it's not a material change, why can't you just wait on evidence. You can still choose to not buy the disks and advise others do not while waiting. 

?

Company removes listings of specs for fear customers won't buy them. I care less what the other customers do. I don't like a company removing the spec, or starting to hide their practices.

 

An example, theoretical. Phone companies stop listing battery MAH because it's misleading. A phone with a fast but power efficient chip (using better tech, say 4nm instead of 14nm) would last longer on a smaller battery, than a wasteful chip on a bigger battery... So they stop listing it.

 

That's fine, right, we only care about "talk time" and "screen on time" and "standby time", and those are still listed? I might be ok with that. If they also drop those metrics, because dirty witch hinter consumers are stupid, I'm out and not buying it. Why? Because I'll not know if they swapped for smaller/worse parts mid release cycle (happens), and tries to swindle me/consumers. They might have a good reason for doing it. It will have unintended consequences though. Besides, *it's pointless* hiding specs from consumers, buyers can make their own list by buying, checking/disassembling/xraying products and (if nondestructive examination) returning. :D

 

 

So, WD not listing something, makes me reluctant to buy their products. Just as I might not buy Samsung TVs as they don't list certain features/display tech, and (IIRC along with other companies) mix in tech and product lines, and you might get OLED/backlight/zones/etc different to the similar model that looks the same. Without seeing/using it in person to know I like it, and not seeing specs online, I would skip.

Quote

You don't have to assume anything at all, that's the point. If you don't know what the issue is and lack the evidence then simply do not blame it. Just wait out the evidence because I can guarantee there will be people(s) testing this right now.

 

It's not a material change if it's not a material change, why can't you just wait on evidence. You can still choose to not buy the disks and advise others do not while waiting.

PS, where did I blame SMR for making these drives fail in NAS boxes? 🤷‍♂️ It is a change though. SMR is not PMR. 🤦‍♂️

Link to post
Share on other sites
2 hours ago, TechyBen said:

Company removes listings of specs for fear customers won't buy them

WD never listed anything at all, for them you are wanting to add something that was never before under the belief it would somehow have informed you before the fact that it would impact you. Like I keep saying and will keep saying, DM-SMR usage could be utterly inconsequential and you wanting to know doesn't change anything. A company is perfectly within their right to not tell you this if it has no impact on you and the intended usage of the product.

 

As I've also said I'd like it listed, but I'm not going to stand around and try and claim that it being on the spec sheet would have done anything at all. You have no idea how DM-SMR impacts the disk at all, you don't know how it was designed. You don't know how large the PMR zone is in the disk and that is likely commercially sensitive information. There are things you just do not need to know and don't actually have a right to demand to know.

 

You are still attributing malice to the act of using DM-SMR, you are still saying with complete confidence it actually affects you and you are still saying it would affect your purchase choice on that ill-informed information. 

 

Again DM-SMR to you may as well be irrelevant, if the product functions correctly, has expected performance knowing DM-SMR is used is just an informative nice to have. If WD thinks it's a marketable technology that would increase sales it'll be slapped all over the box and shouted to high heaven, but it's not so it isn't.

 

Edit:

Also to be really clear you cannot claim you have been mislead when your argument is premised on these user claims and reports that DM-SMR is the cause of the problem. This is why you think there is a problem, this is why you think it even matters at all. As such you and everyone else are misleading themselves and I'm just telling you not to do that. You want to make some kind of consumer being mislead claim you will need evidence, which you currently lack.

 

You can ignore everything I'm saying and just trundle down the path you are but without evidence behind what you are saying it's just baseless complaining which holds water like a sieve.

Link to post
Share on other sites
6 hours ago, leadeater said:
Spoiler

 

WD never listed anything at all, for them you are wanting to add something that was never before under the belief it would somehow have informed you before the fact that it would impact you. Like I keep saying and will keep saying, DM-SMR usage could be utterly inconsequential and you wanting to know doesn't change anything. A company is perfectly within their right to not tell you this if it has no impact on you and the intended usage of the product.

 

As I've also said I'd like it listed, but I'm not going to stand around and try and claim that it being on the spec sheet would have done anything at all. You have no idea how DM-SMR impacts the disk at all, you don't know how it was designed. You don't know how large the PMR zone is in the disk and that is likely commercially sensitive information. There are things you just do not need to know and don't actually have a right to demand to know.

 

You are still attributing malice to the act of using DM-SMR, you are still saying with complete confidence it actually affects you and you are still saying it would affect your purchase choice on that ill-informed information. 

 

Again DM-SMR to you may as well be irrelevant, if the product functions correctly, has expected performance knowing DM-SMR is used is just an informative nice to have. If WD thinks it's a marketable technology that would increase sales it'll be slapped all over the box and shouted to high heaven, but it's not so it isn't.

 

Edit:

Also to be really clear you cannot claim you have been mislead when your argument is premised on these user claims and reports that DM-SMR is the cause of the problem. This is why you think there is a problem, this is why you think it even matters at all. As such you and everyone else are misleading themselves and I'm just telling you not to do that. You want to make some kind of consumer being mislead claim you will need evidence, which you currently lack.

 

You can ignore everything I'm saying and just trundle down the path you are but without evidence behind what you are saying it's just baseless complaining which holds water like a sieve.

 

 

They removed platter numbers.

 

Quote

that was never before under the belief it would somehow have informed you before the fact that it would impact you.

This is not correct. SMR will impact me in certain workloads.

Quote

DM-SMR usage could be utterly inconsequential and you wanting to know doesn't change anything.

*Could be*. Could also not be. I can test this, and thus the argument is invalid. 🤷‍♂️

 

Quote

 You have no idea how DM-SMR impacts the disk at all, you don't know how it was designed.

Yep. But I wish to make informed choices. WD not allowing me to, will mean I go else where (I can look up reviews, but it takes longer than checking a spec sheet with other companies). Hence why I went for a Samsung SSD over Cruicial, burnt once with a lemon there, not going back for a while.

 

Quote

You are still attributing malice to the act of using DM-SMR

No. To the fact of hiding that they are using it.

 

Quote

Again DM-SMR to you may as well be irrelevant, if the product functions correctly, has expected performance knowing DM-SMR is used is just an informative nice to have. If WD thinks it's a marketable technology that would increase sales it'll be slapped all over the box and shouted to high heaven, but it's not so it isn't.

As a customer, I'm not giving them the choice to not tell me.

 

Quote

Also to be really clear you cannot claim you have been mislead when your argument is premised on these user claims and reports that DM-SMR is the cause of the problem. 

If it is or if it isn't is inconsequential. The drives are failing. Currently the only difference seen is SMR. If it's the firmware/seek time and the NAS controller is failing, it's still a material change the consumer could not have accounted for *because they were not told*.

 

Quote

You can ignore everything I'm saying and just trundle down the path you are but without evidence behind what you are saying it's just baseless complaining which holds water like a sieve.

Baseless? Baseless that I request a company tell me what they are doing or what they sell me? ;) Fault or no fault, I ask for hands open deals, not closed fists (as a sign language expression would put it).

Link to post
Share on other sites
6 hours ago, leadeater said:

 

Edit:

Also to be really clear you cannot claim you have been mislead when your argument is premised on these user claims and reports that DM-SMR is the cause of the problem. This is why you think there is a problem, this is why you think it even matters at all. As such you and everyone else are misleading themselves and I'm just telling you not to do that. You want to make some kind of consumer being mislead claim you will need evidence, which you currently lack.

 

You can ignore everything I'm saying and just trundle down the path you are but without evidence behind what you are saying it's just baseless complaining which holds water like a sieve.

This,  why does this need to be said and why are we seeing a lot of it recently.    I get the need to be concerned and that's fair enough, as consumers when things go wrong we should be wary of why.  However we don't know shit about this, hell,  we don't even know how big the problem is, for all we know there are only 15 cases and the confounding factors could literally be hundreds of things other than SMR.   

 

Has anyone got any stats on how many cases there are and what the actually performance difference is between SMR and CMR (in any use case)?    There is no point in trying to demand manufacturers list things unless you know if that thing has any effect on end use. 

QuicK and DirtY. Read the CoC it's like a guide on how not to be moron.  Also I don't have an issue with the VS series.

Sometimes I miss contractions like n't on the end of words like wouldn't, couldn't and shouldn't.    Please don't be a dick,  make allowances when reading my posts.

Link to post
Share on other sites
1 minute ago, TechyBen said:

They removed platter numbers.

 

This is not correct. SMR will impact me in certain workloads.

 

Which ones and by how much? 

 

1 minute ago, TechyBen said:

 

*Could be*. I can test this, and thus the argument is invalid. 🤷‍♂️

Show me the test.

 

I put it to you that the only benchmarks and reviews on HDD's out there are userbenchmark and maybe the odd crystal diskmark that people upload.   None of them can tell you the difference between SMR and CMR so there is no way to actually put a figure on it that would make knowing the feature a benefit.   This is the problem, no one knew there was a performance issue and these drives have been in circulation for a bit now.  We only know about the whole issue because someones silvering process failed and they were clued in enough to deduce the disk was SMR.   Other than that we have no other information.   There is a distinct problem with trying to workout what eh cause of the problem iis when you  only have one or two media articles and the word of a few people that there are issues with hte disk itself.  Independent testing needs to be carried out to find out what the problem is and what the solution is.

 

 

So far all I see is a feature that gives us more storage (at a performance penalty that is yet to be determined) has become a dirty word because of the internet insatiable desire not to think anything through before demanding it knows everything.

 

.

 

 

 

QuicK and DirtY. Read the CoC it's like a guide on how not to be moron.  Also I don't have an issue with the VS series.

Sometimes I miss contractions like n't on the end of words like wouldn't, couldn't and shouldn't.    Please don't be a dick,  make allowances when reading my posts.

Link to post
Share on other sites
23 minutes ago, mr moose said:
Spoiler

 

Which ones and by how much? 

 

Show me the test.

 

I put it to you that the only benchmarks and reviews on HDD's out there are userbenchmark and maybe the odd crystal diskmark that people upload.   None of them can tell you the difference between SMR and CMR so there is no way to actually put a figure on it that would make knowing the feature a benefit.   This is the problem, no one knew there was a performance issue and these drives have been in circulation for a bit now.  We only know about the whole issue because someones silvering process failed and they were clued in enough to deduce the disk was SMR.   Other than that we have no other information.   There is a distinct problem with trying to workout what eh cause of the problem iis when you  only have one or two media articles and the word of a few people that there are issues with hte disk itself.  Independent testing needs to be carried out to find out what the problem is and what the solution is.

 

 

 

So far all I see is a feature that gives us more storage (at a performance penalty that is yet to be determined)

Spoiler

has become a dirty word because of the internet insatiable desire not to think anything through before demanding it knows everything.

 

.

 

 

 

And at that. I'm out. You admit there is a performance penalty, but not mentioned by the manufacturers, but above you claim there is no difference ("prove it" then say "there is a penalty")! As said, Samsung QVO is quad cell SSD, vs their older 3 cell, and other companies have platters/arms/vibration/power capacitor power off safety/etc/etc and are open about it. Here they were not. That's as much as needs to be said.

 

PS, thread from June last year, as an example of them *not* performing the same. Granted, they were not failing, but they were throwing errors, as the users were not expecting SMR, thus the latency was a problem (I assume/guess due to mixed disk performance, because *the consumer was not notified before purchasing*). https://www.ixsystems.com/community/threads/smr-resilver.77964/

Link to post
Share on other sites
1 minute ago, TechyBen said:

And at that. I'm out. You admit there is a performance penalty, but not mentioned by the manufacturers, but above you claim there is no difference ("prove it" then say "there is a penalty")! As said, Samsung QVO is quad cell SSD, vs their older 3 cell, and other companies have platters/arms/vibration/power capacitor power off safety/etc/etc and are open about it. Here they were not. That's as much as needs to be said.

 

There is a performance penalty on paper as per how it was designed.  But that was 6 years ago,  technology has changed so much that there may not be a penalty anymore.  And if there is it may not be big enough to warrant any mention.  There are literally hundreds of different parts and features of each part that have an effect on performance, why does SMR warrant a specific mention over any of the other parts?  we have just as much information on them as we do SMR.

 

Do you see how many may's and and maybe's there are in all this?  it cannot be treated as an absolute as there is just not enough information to do that.  Please find a website or something that has performed benchmarks on various drives and concluded that today's SMR tech is indeed slower or the cause of certain issues.  Do that and you absolutely can legitimately call it a problem that needs  resolution,  like having it listed on drives that are advertised for specific uses like raid arrays or for boot drives.   But until you have that information, you are literally claiming consumer wide problem because some people have had an issue and a lot of maybes surrounding one aspect of those products.

 

 

 

 

 

 

QuicK and DirtY. Read the CoC it's like a guide on how not to be moron.  Also I don't have an issue with the VS series.

Sometimes I miss contractions like n't on the end of words like wouldn't, couldn't and shouldn't.    Please don't be a dick,  make allowances when reading my posts.

Link to post
Share on other sites
1 hour ago, TechyBen said:

If it is or if it isn't is inconsequential. The drives are failing. Currently the only difference seen is SMR. If it's the firmware/seek time and the NAS controller is failing, it's still a material change the consumer could not have accounted for *because they were not told*.

You don't need to account for it if it's not actually the problem and is unrelated, correlation is not causation and you're being a great sample of this. Show me that causal evidence link otherwise it doesn't matter.

 

Just like it doesn't matter what the platter is made out of, or the composition of the magnetic film applied to it. Did you know there are difference for both and these have changed over the Years?

 

You also know that the number of platters have changed over the years, even on PMR disks, so do you think the number of these matter? Does a 4 platter 1TB disk perform better than a 2 platter 1TB disk or a 1 platter 1TB disk? If you don't know the answer to this then saying it needs to be on the spec sheet just because it's different is a not a good argument, you have no idea if it's even relevant or why you need to know, or even if it has an impact at all.

 

1 hour ago, TechyBen said:

No. To the fact of hiding that they are using it.

It's not hiding if it's irrelevant and you don't need to know. You haven't proven you need to know, you haven't evidenced how it impacts you at all.

 

1 hour ago, TechyBen said:

*Could be*. Could also not be. I can test this, and thus the argument is invalid. 🤷‍♂️

Then EVERYTHING you have been saying is also invalid, so are we done? You haven't tested anything, you have no evidence and everything you say is just coulds. It could be space ghosts? No it is space ghosts causing this problem. It could be so I demand WD tell me if WD Reds contain space ghosts!

 

How about like I said wait for evidence around the issue so you can be better informed about the problem so you can know what information to ask for and why you need to ask for it.

Link to post
Share on other sites
1 hour ago, TechyBen said:

PS, thread from June last year, as an example of them *not* performing the same. Granted, they were not failing, but they were throwing errors, as the users were not expecting SMR, thus the latency was a problem (I assume/guess due to mixed disk performance, because *the consumer was not notified before purchasing*). https://www.ixsystems.com/community/threads/smr-resilver.77964/

No, and everyone saying anything like this is wrong, the source of the information never said this either. I actually have coming up 25 years experience in RAID, I know what I'm talking about. The person who reported the problem and is using ZFS also knows what he is talking about too.

 

The disks not being an exact match is not the issue. It's not the issue. The issue is literally spelled out in the source articles.

 

Quote

In the case of ZFS, resilvering isn’t a block level “end to end” scan/refill, but jumps all over the drive as every file’s parity is rebuilt. This seems to trigger a further problem on the WD40EFAXs where a query to check a sector that hasn’t been written to yet causes the drive to internally log a “Sector ID not found (IDNF)” error and throws a hard IO error from the interface to the host system.

 

Quote

Because I don’t have any Seagate SMR drives, I can’t test the hypothesis that the IDNF issue is a WD firmware bug rather than a generic SMR issue. But throwing an error like that isn’t the kind of thing I’d associate with SMR as such – I’d simply expect throughput to turn to shit.

 

 

Edit:

Also these new disks will only have a potential performance problem if you are adding it to an existing ZFS pool to rebuild a failed disk or adding it to a traditional RAID array that is largely full. After the disk has been rebuilt in to an array or added into an existing array, like disk revisions or not, these new revisions of disks will perform better than the old ones for the target market they are for, small NAS systems of less than 8 disks.

 

If you try and fill the entire array from empty to full all in one go you also may encounter a performance problem, however unless you have 10Gb network connection it would be impossible to trigger this situation as your sustained throughput is too low.

Link to post
Share on other sites
19 minutes ago, leadeater said:
Spoiler

 

No, and everyone saying anything like this is wrong, the source of the information never said this either. I actually have coming up 25 years experience in RAID, I know what I'm talking about. The person who reported the problem and is using ZFS also knows what he is talking about too.

 

The disks not being an exact match is not the issue. It's not the issue. The issue is literally spelled out in the source articles.

 

 

 

 

Edit:

Also these new disks will only have a potential performance problem if you are adding it to an existing ZFS pool to rebuild a failed disk or adding it to a traditional RAID array that is largely full. After the disk has been rebuilt in to an array or added into an existing array, like disk revisions or not, these new revisions of disks will perform better than the old ones for the target market they are for, small NAS systems of less than 8 disks.

 

If you try and fill the entire array from empty to full all in one go you also may encounter a performance problem, however unless you have 10Gb network connection it would be impossible to trigger this situation as you're sustain throughput is too low.

 

 

And. Did you read those posts in that forum? "I wish to keep to PMR drives" because they already had PMR drives. You are saying they don't have the right to make that decision to ask what type of drive it is and decide to buy/not/redesign their NAS/RAID/ZFS/failure redundancy on that? and not changing tech/design/spec = better knowledge of the hardware = (generally) less "new adoption" failures errors. This could be a normal firmware bug. But those *opting* to buy like for like to avoid new firmware bugs (mainly from new features) have the choice to chose.

 

They don't if companies withould that info till it fails in use.

Link to post
Share on other sites

from what i understood smr does bad in raid configurations yet raid is often used in nas, if you ask me that kind of info needs to be disclosed as it can prevent a lot of problems with failures.

Link to post
Share on other sites
7 minutes ago, TechyBen said:

And. Did you read those posts in that forum? "I wish to keep to PMR drives" because they already had PMR drives. You are saying they don't have the right to make that decision to ask what type of drive it is and decide to buy/not/redesign their NAS/RAID/ZFS/failure redundancy on that?

He can like what ever he wants, they are saying this because they believe DM-SMR is the source of the problem. And yet again you do not need to redesign the NAS because of this, these disks can be used along side the old one and it will operate fine. Yes there is an issue with the new revisions of the disk, what exactly is the cause of the problem is unknown.

 

WD has replaced the old revision with the new, unless you can find a source of the old revision you have to buy this one.

 

Do you even know how many revisions of the WD Red line there has been? It's not 2.

Link to post
Share on other sites
25 minutes ago, leadeater said:

He can like what ever he wants, they are saying this because they believe DM-SMR is the source of the problem. And yet again you do not need to redesign the NAS because of this, these disks can be used along side the old one and it will operate fine. Yes there is an issue with the new revisions of the disk, what exactly is the cause of the problem is unknown.

 

WD has replaced the old revision with the new, unless you can find a source of the old revision you have to buy this one.

 

Do you even know how many revisions of the WD Red line there has been? It's not 2.

?

?

They asked for a PMR drive. That is what they were asking for. That is what they wished to purchase. I hope you never serve me at a restaurant. 🤣

 

Quote

Do you even know how many revisions of the WD Red line there has been? It's not 2.

I would test every product before rolling out. Because I don't trust them or the products they release to perform as expected (note "expected", I still expect them to perform some way, just assume they will change their mind without noting it in a spec sheet). Their spec sheets are not reliable/avaliable. So I'd probably go to a competitor. What is your defence against them again?

 

Unless you have access to some additional data, such as SMR total PMR cache zone size, or consistent throughput rate etc. AFAIK Samsung (and others) don't supply this for SSDs, but constant throughput rates can be seen in available benchmarks (theoretical example "450mpbs over 50gb till the 1 bit 4 bit cell cache is used up, then 350mbps after that" and/or "450mbps sustained on an empty drive or with overprovisioning, 350mbps when full disk"). We don't have those for  these WD SMR drives because *we were not told they were SMR*. But we are told SSDs do/don't have cache (though some companies try to hide if it has a ram cache or not, but as said, disassembling it can find out!).

 

For those who need certain performance/behaviour type, they can just buy/test/refund or buy/disassemble (to see what platters/cache chips are on the PCB for HDD/SSD) and post reviews (at a cost mind :( ) to test. So hiding the specs is pointless from a marketing/business practice and only hurts those customers if it's cheaper/more convenient/more reliable to go to their competitor. WD wins a large bulk low effort consumer space, but loses out to those who need the reliability/openness. Note, "openness" need not have a fault. These could be 1000% reliable, yet not being told, that is where people feel uneasy.

Link to post
Share on other sites
8 minutes ago, TechyBen said:

They asked for a PMR drive. That is what they were asking for. That is what they wished to purchase. I hope you never serve me at a restaurant. 🤣

He asked for it because he doesn't know what the problem is and like yourself are applying bad logic and assumptions to the situation. You can wish what ever you like, just don't give me incorrect and faulty reasoning as to why you want something.

 

If you don't like this being pointed out then just try not to use it in your arguments with me. 

 

If you want DM-SMR listed on the spec sheet cool, just say that. Avoid bringing anything up like this as to you why want it, otherwise I'm going to repeat to you again why that reasoning is flawed.

 

You can avoid this circular debate by avoiding unevidenced reasoning.

 

I want it on the spec sheet, end of sentence.

Link to post
Share on other sites
3 minutes ago, leadeater said:

He asked for it because he doesn't know what the problem is and like yourself are applying bad logic and assumptions to the situation. You can wish what ever you like, just don't give me incorrect and faulty reasoning as to why you want something.

 

If you don't like this being pointed out then just try and use it in your arguments with me. 

 

If you want DM-SMR listed on the spec sheet cool, just saying that. Avoid bringing anything up like this as to you why want it, otherwise I'm going to repeat to you again why that reasoning is flawed.

 

You can avoid this circular debate by avoiding unevidenced reasoning.

 

I want it on the spec sheet, end of sentence.

🧐 He asked for a PMR drive. He can be told "yes or no". Right? I have that one question. Can he ask WD "Is this drive PMR or SMR"?

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now


×