ssd failed? how should prevent future failures?

[iMBP15]

I have a macbook pro 15" 2014 and one day I was streaming to my chromecast watching a movie and when it finished I grabbed the laptop to search something up to realize it had froze during the movie. I held the power button down until it shut down and started it back up again. black screen with folder question mark. researched what can be done, tried:

"command+r" disk utility couldn't discover the drive and trying to reinstall osx no drive popped up
"D" error VDH002 possible problem with hard drive
"cmd+option+p+r" nothing

took the bottom off and removed and reinserted the ssd. still nothing.
finally grabbed another macbook took it apart and stuck the ssd in and nothing happen, but the drive from that laptop booted right up on my main.

ordered a new 512gb for $340 off ebay hopefully I dont have any issues with that order but my question is why the hell did it fail???? some research suggested heat as the culprit so my other question is whether anyone has tried appling a heatsink to the ssd. I've found these copper shim "heatsinks" im thinking of applying with thermal paste or might order these interesting micro porous ceramic heatsinks I found off a website which are supposed to be 30% more efficient than metal ones. they're made for raspberry pi chips but thought I would give it a try

can anything be done with the old m.2 ssuax ssd? can I sell it to someone for some value? thoughts about the heatsink on the ssd?

 

[Samuelsan2001]

I have a macbook pro 15" 2014 and one day I was streaming to my chromecast watching a movie and when it finished I grabbed the laptop to search something up to realize it had froze during the movie. I held the power button down until it shut down and started it back up again. black screen with folder question mark. researched what can be done, tried:

"command+r" disk utility couldn't discover the drive and trying to reinstall osx no drive popped up
"D" error VDH002 possible problem with hard drive
"cmd+option+p+r" nothing

took the bottom off and removed and reinserted the ssd. still nothing.
finally grabbed another macbook took it apart and stuck the ssd in and nothing happen, but the drive from that laptop booted right up on my main.

ordered a new 512gb for $340 off ebay hopefully I dont have any issues with that order but my question is why the hell did it fail???? some research suggested heat as the culprit so my other question is whether anyone has tried appling a heatsink to the ssd. I've found these copper shim "heatsinks" im thinking of applying with thermal paste or might order these interesting micro porous ceramic heatsinks I found off a website which are supposed to be 30% more efficient than metal ones. they're made for raspberry pi chips but thought I would give it a try

can anything be done with the old m.2 ssuax ssd? can I sell it to someone for some value? thoughts about the heatsink on the ssd?

Nothing to be done ssd's are generally very reliable but like any electronic equipment they can fail at any time for any number of reasons. You have just been unlucky that's all simple as that no heat sinks or thermal paste or anything will change your SSD life, they will mostly last longer than the computer they are in by years.

 

[Newtons Apple]

I suspect your first SSD was defective and there was little you could do to save it. Your new drive should last longer that you want to use it.

 

[Weaselboy]

Yup... I agree with the other comments OP. There is really nothing you did to cause this. It sounds like you just got a dud.

 

[iMBP15]

I did push it to the limit like every day... torrenting, gaming, video editing easily when through 10tb read/write a day... and the heat that came off the laptop felt like it could cook eggs. you sure that wasnt the issue?

all of you said it couldnt be heat but it wouldnt hurt the ssd if I did go ahead and apply a heat sink right?? hardware that runs cooler in general lasts longer so I figure what the hell. the macbook has a terrible air flow in it.
[doublepost=1488399244][/doublepost]

Yup... I agree with the other comments OP. There is really nothing you did to cause this. It sounds like you just got a dud.

it worked for 2 years I really feel like if it was a dud it would have failed much sooner. :/
[doublepost=1488399278][/doublepost]

I suspect your first SSD was defective and there was little you could do to save it. Your new drive should last longer that you want to use it.

Im buying the second used so hopefully no problems

 

[Newtons Apple]

I did push it to the limit like every day... torrenting, gaming, video editing easily when through 10tb read/write a day... and the heat that came off the laptop felt like it could cook eggs. you sure that wasnt the issue?

all of you said it couldnt be heat but it wouldnt hurt the ssd if I did go ahead and apply a heat sink right?? hardware that runs cooler in general lasts longer so I figure what the hell. the macbook has a terrible air flow in it.
[doublepost=1488399244][/doublepost]
it worked for 2 years I really feel like if it was a dud it would have failed much sooner. :/
[doublepost=1488399278][/doublepost]
Im buying the second used so hopefully no problems

You are buying a used hard drive?

Hope the original owner did not do the same and passed it on to you!

Heat sink might help but these SSD units work under worst conditions than you will find in your lappy.

 

[iMBP15]

Nothing to be done ssd's are generally very reliable but like any electronic equipment they can fail at any time for any number of reasons. You have just been unlucky that's all simple as that no heat sinks or thermal paste or anything will change your SSD life, they will mostly last longer than the computer they are in by years.

Extreme heat and trauma are well known reasons what a hard drive would fail but considering it did last 2 years I dont think it was just simple bam done.
[doublepost=1488399484][/doublepost]

You are buying a used hard drive?

yeah is that a bad idea?

 

[jerryk]

yeah is that a bad idea?

No way I would every buy a used drive.

What does Apple want for a new one?

 

[Newtons Apple]

yeah is that a bad idea?

No you should be just fine!

 

[Sanpete]

yeah is that a bad idea?

If you can get one used from a reputable seller who can tell you how much it's been used, you should be fine, but I don't know if you'll save enough to make up for the lack of warranty.

 

[ZapNZs]

I did push it to the limit like every day... torrenting, gaming, video editing easily when through 10tb read/write a day... and the heat that came off the laptop felt like it could cook eggs. you sure that wasnt the issue?

all of you said it couldnt be heat but it wouldnt hurt the ssd if I did go ahead and apply a heat sink right?? hardware that runs cooler in general lasts longer so I figure what the hell. the macbook has a terrible air flow in it.
[doublepost=1488399244][/doublepost]
it worked for 2 years I really feel like if it was a dud it would have failed much sooner. :/
[doublepost=1488399278][/doublepost]
Im buying the second used so hopefully no problems

My knowledge here is limited, so bear with me...

I assume your SSD is Samsung-made? That particular SSD has a likely endurance of at least several petabytes, and the closest thing to a retail version has, IIRC, a 400 or 500 TB/W warranty. However, you noted that you were writing 10 terabytes daily...that insanely high amount (filling and fully emptying the 512 GB SSD 20 times each day) could have exhausted the drive, and it is going to rapidly wear even high-endurance MLC SSDs. However, the few SSDs I have had that have become exhausted showed (SMART) pre-fail warning signs and one "died" into a 'read-only' mode (which I believe was by design after a certain amount of writes had occurred.)

I have been told that, with SSDs, a sudden failure like that often indicates a defect with the individual product that the SSD/SSD+OS cannot compensate for (such as a critical failure in a circuit from a manufacturing defect) or damage caused by power loss during a write cycle (and I assume Apple and/or Samsung has ensured the SSD has PLP or an equivalent?)

I use a SMART app that monitors HDDs (as well as SSDs), but one of SMART's limitations is arguably predicting sudden, catastrophic failure (as the Google drive study indicated that there may be no warning signs until sudden death, although these failures do appear more common within the first several months of use), and so it seems very possible that SMART could not have predicted this failure.

Considering that you are writing 10 TB a day, if you are not already, have you considered something along the lines of a ThunderBolt RAID enclosure with 4 high-endurance HDDs?
https://www.amazon.com/Akitio-Thund...16&sr=1-1&keywords=thunderbolt+raid+enclosure
https://www.bhphotovideo.com/c/product/985498-REG/hgst_0s03601_4tb_ultrastar_internal_hd.html
https://www.bhphotovideo.com/c/product/1162542-REG/hgst_0f23267_ultrastar_8tb_7200_sata6_64mb.html

 

[Weaselboy]

I have been told that, with SSDs, a sudden failure like that often indicates a defect with the individual product....

Exactly, and this is why I made the comment OP got a dud. There are couple tests I have seen like this one that just hammered an SSD until it died, and they don't usually just stop working all at once.

 

[ZapNZs]

Exactly, and this is why I made the comment OP got a dud. There are couple tests I have seen like this one that just hammered an SSD until it died, and they don't usually just stop working all at once.

Sorry to go a little OT here...do you know if OS X monitors SMART metrics on its own and has the ability to alert the User to a prefail flag during normal usage?

The OP noted he writes 10 TB a day...over two years, if my math is correct that equates to like 7 petabytes and such a high number made me curious about this type of warning. Further, do you know if the Apple-Samsung SSD referenced here has its own 'read-only-death-mode'? On an older Micron SSD, I saw the rise in reallocated sector count as they did in that test (but it was a 3rd party App and not OS X that flagged this [as well as a wear leveling count that indicated the service life was about over, so the drive was obviously replaced immediately.]) Just out of my own curiosity, I'm interested what (if anything) the system or SSD would do once the OEM SSD reaches exhaustion. Can it tell the User this without having to use a 3rd party App?

 

[Weaselboy]

Sorry to go a little OT here...do you know if OS X monitors SMART metrics on its own and has the ability to alert the User to a prefail flag during normal usage?

The OP noted he writes 10 TB a day...over two years, if my math is correct that equates to like 7 petabytes. Further, do you know if the Apple-Samsung SSD referenced here has its own 'read-only-death-mode'? On an older Micron SSD, I saw the rise in reallocated sector count as they did in that test (but it was a 3rd party App and not OS X that flagged this [as well as a very low wear leveling count, so the drive was obviously replaced immediately.]) Just out of my own curiosity, I'm interested what (if anything) the system or SSD would do once the OEM SSD reaches exhaustion. Can it tell the User this without having to use a 3rd party App?

You can see SMART status in Disk Utility, but interestingly, my new 2016 TouchBar MBP shows SMART is not supported, so I don't know any way you could tell how things are going with write cycles on this thing. My 2014 rMBP did show the SMART status in that same spot, but no detailed info. You would have to use something like the open source smartmontools/smartctl package to see detailed SMART info and that would start to show error rates and dead NAND cells marked off.

I have no idea if the Apple OEM drives have that read only mode you mentioned and I have never seen any flash storage impending death indicators other than the normal disk write error messages you would get.

 

[ZapNZs]

You can see SMART status in Disk Utility, but interestingly, my new 2016 TouchBar MBP shows SMART is not supported, so I don't know any way you could tell how things are going with write cycles on this thing. My 2014 rMBP did show the SMART status in that same spot, but no detailed info. You would have to use something like the open source smartmontools/smartctl package to see detailed SMART info and that would start to show error rates and dead NAND cells marked off.

I have no idea if the Apple OEM drives have that read only mode you mentioned and I have never seen any flash storage impending death indicators other than the normal disk write error messages you would get.

That's interesting how the 2016 doesn't show it as supported. I wonder if Apple has implemented their own protocol for monitoring/alerting with the redesign?

I'm still using El Cap, but with El Cap, all I can verify is that SMART is "Verified" via Disk Utility. For example, this drive shows as Verified, I can format back/forth to ExFAT/MBR to HFS+/GPT, and it passes when running First Aid...


...But OS X doesn't tell me the drive has failed or throw any sort of warning, as far as I can tell





Does that mean the "Verified" only indicates SMART capability, as opposed to status?

 

[Samuelsan2001]

I did push it to the limit like every day... torrenting, gaming, video editing easily when through 10tb read/write a day... and the heat that came off the laptop felt like it could cook eggs. you sure that wasnt the issue?

all of you said it couldnt be heat but it wouldnt hurt the ssd if I did go ahead and apply a heat sink right?? hardware that runs cooler in general lasts longer so I figure what the hell. the macbook has a terrible air flow in it.
[doublepost=1488399244][/doublepost]
it worked for 2 years I really feel like if it was a dud it would have failed much sooner. :/
[doublepost=1488399278][/doublepost]
Im buying the second used so hopefully no problems

Read amounts don't wear the drive at all it's only the writes that will count and there is no way you were writing 10TB a day with that usage. Heat would kill the GPU and cpu way before it killed a drive. It just failed it happens. You are making something of nothing.

 

[leman]

my question is why the hell did it fail????

The most treasonable thing is to write it down to "luck". Stuff breaks, and stuff fails. Thats why one should always have backups and (if possible) warranty.

If you are looking for a recipe to prevent the SSD from failing, well, just don't use the computer. Although it will still fail eventually from old age.

 

[Weaselboy]

Does that mean the "Verified" only indicates SMART capability, as opposed to status?

No... I have seen that DU readout say "Failed" also, so it does work. Odd you have that bad reading with Drive DX though.

 

[NT1440]

Extreme heat and trauma are well known reasons what a hard drive would fail but considering it did last 2 years I dont think it was just simple bam done.

Extreme heat, yes. But your laptop is not generating anywhere close what electronics would consider "extreme" so that point is moot.

 

[Ma2k5]

yeah is that a bad idea?

To buy and use a used hard drive for me, would be like washing and re-using a used condom. The thought of it makes me cringe.

 

[ZapNZs]

No... I have seen that DU readout say "Failed" also, so it does work. Odd you have that bad reading Drive DX though.

My first thought was that DriveDX reading was wrong (because I've seen different SMART apps sometimes have slightly different dispositions on the same drive, which I always figured was a deviation between how they interpret the raw figures?), so I tried CrystalDiskInfo on 10, and then WD's own App on XP (edit: originally said 7), which both showed the drive was bad also. I tried zeroing it for good measure, which made no difference, and then tried writing and reading data, where some errors occurred, and both sustained transfer speeds were well below the 90-120 MB/s that specific drive should have been doing using a 25 GB data file I use to test all my drives. Until then I just used the DriveDX program now and then to take a look at how all my drives were (I wasn't sure of the value of the App but I liked the interface and having an on-the-fly interpretation of raw values, even if qualifying quantifiable data meant some degree of subjectivity...), but after that I set it to load in the background at startup and refresh every 12 hours, as I wasn't sure what to make of not being notified by OS X. There wasn't a risk of data loss since I wouldn't store unique data on just one drive, but I still want to know if a drive is going bad.

I had a similar experience with a WD 750GB 2.5-inch Black, where WD's App and DriveDX flagged it but OS X did not.


With pending sectors, I then zeroed it and then kept using that drive to see if it may have been a false report. Shortly after putting it back into use, DX again flagged it, and eventually the drive became unusable. (This was also with El Cap, but on a different MBP.)


Could the threshold for what the DriveDX, CrystalDiskInfo, and Western Digital apps consider "bad" differ from what OS X considers "bad"?

 

[Weaselboy]

Could the threshold for what the DriveDX, CrystalDiskInfo, and Western Digital apps consider "bad" differ from what OS X considers "bad"?

I just don't know. I know many of these tools are just a GUI for the open source smartctl package, but I have no idea if there is a standard metric for what level of errors changes "verified" to "failed".

If I had a drive like that showing those SMART errors I would toss it in the trash. It seems very common on the forums to have people with obviously bad drives and DU still shows nothing wrong. My sense is SMART is just another tool/indicator and not the definitive pass/fail test like we would hope.

https://www.smartmontools.org

 

[JPNFRK7]

ordered a new 512gb for $340 off ebay hopefully I dont have any issues

Isnt this pretty expensive? Im looking at 512GB SSDs on Amazon for around $150 brand new. Am I missing something?

 

[maerz001]

Isnt this pretty expensive? Im looking at 512GB SSDs on Amazon for around $150 brand new. Am I missing something?

Yes. The MBP SSDs have a proprietary connector. So the usual cheap 2,5" won't fit

 

[BiscottiGelato]

This seems to be a rare problem, as I don't see widespread complaints on SSD failure.
But mine did fail. Was running fine for the most part for the past 3 years. Then today the computer froze. I thought it's another one of those Xcode 9 stunts... The laptop auto restarted and the kernel panic report contained the message

"Could not recover SATA HDD after 5 attempts"

I used it as normal and an hour or so later it happened again. It froze, restarted and then gave me a kernel panic report. I decided to look the above message up, and 30 minutes later the computer froze again, and on restart it gave me the folder icon signifying that it can no longer find the OS. Took it to the Apple Store and they confirm that the system is not seeing the drive...

Late 2013 13" rMBP. Samsung SSAUX 256GB SSD.