Please create an account to participate in the Slashdot moderation system

 


Forgot your password?
Close
typodupeerror
×
Data Storage

SanDisk Extreme SSDs Keep Abruptly Failing (theverge.com) 59

Posted by BeauHD from the buyer-beware dept.
According to Ars Technica, some SanDisk Extreme SSDs are wiping people's data. While SanDisk told Ars that a firmware fix is coming "soon," owners with 2TB drives are out of luck. From the report: An Ars reader tipped us (thanks!) to online discussions filled with panicked and disappointed users detailing experiences with recently purchased Extreme V2 and Extreme Pro V2 portable SSDs. Most users seemed to be using a 4TB model, but there were also complaints from owners of 2TB drives.

Until now, there has been little public response from SanDisk, which has mostly referred online users to open a support ticket with SanDisk's technical support team. Questions about refunds have been left unanswered. When Ars contacted SanDisk about the issue, a company representative said: "Western Digital is aware of reports indicating some customers have experienced an issue with 4TB SanDisk Extreme and/or Extreme Pro portable SSDs (SDSSDE61-4T00 and SDSSDE81-4T00 respectively). We have resolved the issue and will publish a firmware update to our website soon. Customers with questions or who are experiencing issues should contact our Customer Support team for assistance."

SanDisk didn't answer our questions about refunds, whether or not the firmware would address issues with the 2TB models, what caused the issue, or when exactly this firmware fix will come. Some Reddit users have suggested that SanDisk has dragged its feet on the monthlong saga, with ian__ claiming they needed to collect "data to prove to SanDisk that it actually is more than a fluke." SanDisk's brief response to Ars' questions fails to clarify what's been going on behind the scenes.
This discussion has been archived. No new comments can be posted.

SanDisk Extreme SSDs Keep Abruptly Failing More

SanDisk Extreme SSDs Keep Abruptly Failing

Comments Filter:

  • This really is a rogue-like disk! (Score:5, Funny)

    by gweihir ( 88907 ) on Tuesday May 23, 2023 @05:41AM (#63544519)

    Store the wrong data once, get permadeath! The user just did not understand what "Extreme" really meant.

    I am somewhat surprised how long this took. Apparently the lessons from the crap OCZ and others used to sell are wearing off.

    • I've been having more and more failures with Sandisk SD cards to the point that I will no longer consider buying them. Guess it's Samsung from here until they get lazy

      • Re: (Score:2)

        by gweihir ( 88907 )

        Interesting. Another respected name down the drain. They probably let the usual no-clue bean-counters make decisions.

        So far no problems here with several Samsung, NVME and SATA. One caveat: For example a 990 pro NVME goes down to around 550MB sustained write speed after a while. That is with the naked SSD. The cooler is still in the mail, hence I cannot yet say whether that is temperature or something else. It did not seem very warm to the touch though. At least that speed was then sustained for an entire o

        • Re:This really is a rogue-like disk! (Score:5, Informative)

          by arglebargle_xiv ( 2212710 ) on Tuesday May 23, 2023 @06:57AM (#63544681)

          Intel had the same problem with their advertised-as-superior-resilience SSDs some years ago, there was a firmware bug that turned any size SSD into an 8MB storage device. After denying the problem for months, then promising a firmware fix for at least some models that didn't actually fix it, then denying it a bit more, they quietly discontinued that entire product line and pretended it had never existed.

          That was the last Intel storage device we ever bought.

          • That was the last Intel storage device we ever bought.

            In defence of Intel, if you treated all storage companies with the same approach then you wouldn't ever be able to buy any storage device ever again. They have all had their turds, and they have all pretended it wasn't them who shat them out.

        • Re: (Score:2)

          by tlhIngan ( 30335 )

          Toshiba bought up OCZ which made very unreliable SSDs back in the day - because OCZ cheaped out by running the firmware in a mode that required the SSD have power failure protection capacitors, but of course they didn't put them on. Thus it's possible to corrupt them by simply losing power.

          The controllers could run in a slower mode which didn't require the backup capacitors, but that mode was slower.

          It was an odd acquisition since OCZ was known to be makers of unreliable SSDs and someone to avoid at all cos

          • Re: (Score:2)

            by gweihir ( 88907 )

            Possibly. The other point with OCZ is really cheap FLASH memory. I tried to boot 4 older OCZ ones, 3 of which worked fine when I put them in storage, all completely dead. I would at least have expected them to give me an error, but simply nothing. Fortunately, I never trusted them and was well aware they could fail on me at any time and hence no data-loss.

            Incidentally, I have a data point on Toshiba being crap as well, just for HDDs. I recently got two of their external 4TB USB drives for backups. After put

          • OCZ was hit and miss. I still have one of their more problematic models (vertex 2) running well over a decade after the purchase.

            • Re: (Score:2)

              by gweihir ( 88907 )

              I have 3 dead (of 3) and another one from them dead, all bought seperately. My guess is they bought mixed FLASH and much of it was bad, but sometimes they put in good memory and then the disks kept working. May also be you have a really unproblematic workload or got lucky with the firmware version.

          • because OCZ cheaped out by running the firmware in a mode that required the SSD have power failure protection capacitors

            No. OCZ had some real fundamental flaws in their firmware. Their drives failed without any help of power outages. The general industry had a problem with power failure protection, and that's a given since you otherwise run your SSD slower than molasses on a cold day or spend money on capacitors that virtually no one put in outside of their enterprise class drives.

            It was several years before manufacturers in general designed consumer drives to be properly resilient to power failure. But OCZ needed no such he

        • Samsung isn't exactly squeaky clean either. The 840 and 840 EVO issues still have not been properly fixed, and there are premature / excessive wear issues with the 980 Pro and 990 Pro
          That and "Pro" means jack nowadays since they switched from MLC (970 Pro, 860 Pro) to TLC. Wonder how soon until even the "Pro" switches to QLC.

        • Thermal throttling is definitely a possibility; but having the speed fall substantially after some heavy writes to a lower sustained speed that is maintained until the drive fills sounds precisely like a drive running out of pseudo-SLC cache and showing you what its NAND is actually capable of. I'm not sure if Samsung has official details, 3rd party benchmarks suggest that the cache is in the vicinity of 220-240GB when a 2TB drive is empty, dynamically resized as it fills up.

          If your workloads are primari
      • Knocking on wood, I havent had any issues yet. I wonder if the article is referring to the external ones. I have a couple 1TB models running on usb 3.3 gen2 ports.
        • The word "portable" is normally used to refer to external usb attached drives.

          • Re: (Score:2)

            by e3m4n ( 947977 )
            its reddit, the thread morphed into SD cards and microSD. I have two of the externals that are supposedly ruggedized. One is mounted on the PS5 for storing and running ps4 games without a hit to its performance. I havent noticed anything ever since I fixed a windows issue where it was laggy until I went in and enabled write-back cache and performance. Now its definitely pushing the 3.2gen2 threshold and responds quickly. You just cant hot plug the cable without first ejecting via software.

            • Yeah I mixed things up a bit here, I don't have any Sandisk SSDs so I can't comment on those, but the quality of their SD cards has been plummeting. So maybe the problem is dev and/or QA and it's endemic.

      • Re:This really is a rogue-like disk! (Score:5, Interesting)

        by UnknowingFool ( 672806 ) on Tuesday May 23, 2023 @09:15AM (#63545013)
        One of the problems that Linus Tech Tips and others have noted is that SSD manufacturers have been switching out components [youtube.com] during the lifetime of SSD models. The initial versions (especially those sent out for review) have the best components while subsequent versions have what he deemed as less capable components (slower flash, slower controllers, slower cache). While this component substitution happens in manufacturing due to multiple factors like supply chain constraints, a particular complaint is that manufacturers were not labeling these new variants as different models or revision numbers publicly. A member of the public is not going to know that a model that did well in review is not exactly the same and will not perform as close to the review numbers.

        • I have seen this with other types of products. For instance, a completely different USB driver is needed, despite no change in the model or part number. It's fine if you're on Windows, but it's a major headache if you're in an embedded system. Mostly this happens with consumer oriented products, some that don't even have real data sheets.

      • I'm not sure why you would switch to Samsung, they have their own issues.

        https://www.theverge.com/2023/... [theverge.com]

        • Re: (Score:2)

          by gweihir ( 88907 )

          I smell a rat.

          The 980 pro and the 990 pro do not have a "disk health" SMART attribute (I have a 2TB model of both, no issues). This "disk health" score is produced by the _tool_.

          From the screenshot given in the reference, the SSD has 100% of spares available and no media or data integrity errors or error log entries at all. It also has no critical warnings. "Low health" seems to be entirely due to the "percentage used" attribute and this looks like it is a display or math error because any real problems wo

      • Re: (Score:2)

        by stikves ( 127823 )

        Samsung has their share of problems:
        https://arstechnica.com/gadget... [arstechnica.com]
        "Samsung 990 Pro SSD firmware update should halt—but not reverse—rapid wear-out"

        I would go with SK Hynix at this point.

        (Half expecting a futher reply in this chain :) )

  • "Resolved" is wrong! (Score:5, Insightful)

    by Entrope ( 68843 ) on Tuesday May 23, 2023 @05:45AM (#63544535) Homepage

    They have not "resolved" the issue until users can fix their drives. They may have found a fix for the issue, but until they make that available for all their users, they have not resolved it.

  • Sandisk forgot the adjective. Can you help?
    SanDisk Extreme _______ SSD

    • Dunno about the blank but the SSD seems to mean "shitty storage device". I think Intel use the same naming (see my earlier post about their self-bricking device 8MB bug).

    • Re: (Score:2)

      by gweihir ( 88907 )

      Hmmm. "Russian Roulette?" Or should that be "Freedom Roulette" these days?

    • Re: (Score:2)

      by dgatwood ( 11270 )

      After "Extreme", it needs to be a noun. I would suggest "risk".

  • WD nvme inside (Score:5, Interesting)

    by Anonymous Coward on Tuesday May 23, 2023 @07:12AM (#63544715)

    I got this sandisk extreme 4TB as a gift. As i have no use for it as external storage I opened it up and realized it has normal NVMe drive inside - Western Digital SN730E.

    Installed it into my main PC - use it for temporary storage and games. And have no issues with the drive itself.

    The interesting thing is that i tried to put some other NVMe drive into the enclosure and it wouldn't work. Seems to be software locked.

    • Re: (Score:1)

      by Anonymous Coward
      Putting a drive that needs considerable cooling into an enclosure that has no cooling? No wonder the thing has problems when it is used for longer write and read operations.

    • That isn't uncommon in the slightest. Samsung T5 external SSD is a Samsung mSATA drive in an enclosure. Crucial X8 is a NVMe drive. It's actually harder to find external SSDs which aren't internal SSDs with a USB controller attached. I mean it stands to reason, normal external HDDs were exactly the same.

      • What else would you want than external drive = internal drive plus USB interface? That keeps development cost low and development faster.
  • Customer hostility is not what I look for from a data storage company.

    • But who do I buy from? Intel had it's problem and ignored the customer. Samsung is customer hostile across all their products. WD (SanDisk's parent company) has been doing underhanded shit to customers for a while now. Seagate still pretends like they don't have reliability problems and blame the user for every failure.

      It's almost now a minimum requirement for a storage customer to have shit customer support.

  • This IS [hackaday.com] your great-grandfather's persistent storage medium.

    It's a bit low on capacity, but at least you can do component-level repairs.

  • In the hope this helps even a single reader....

    A backup is about having more than one copy of important data. An additional 1TB external disk is like $50, easily within budget of most of users.

    I personally use drives from different manufacturers (my paranoia about hardware defects doesn't seem so absurd now).

  • This is why backups are so important, even for regular folks.

    If you're not backing up regularly your sitting on a pile of dynamite with a lit fuse. There is no IF just WHEN you will lose data.

    Large spinning disk media pluggable USB drives are almost a dime a dozen so there's no excuse to not have one.

  • SanDisk Extreme SSDs Keep Abruptly Failing

    I thought that was part of the "Extreme" part -- can't make an omelette w/o sometimes destroying all your data, and all that ... :-)

  • Ah, hell - I bought one a few months ago for my laptop and get the same thing - massive r/w errors and then it goes off the SATA bus until being powered down for a while.
    I finally kicked it out of my ZFS mirror until I can replace it.
    FWIW, mount /boot/efi nofail if you use zfs boot.

  • I wont ever spend that in-ordinate amount of money on a 4TB SSD vs using a cheap as chips HDD which wont wipe itself as well as covering my ass by also using optical media as well as LTO tape in my archival process.

    I still feel 64GB flash drives are too big for lobbing about the place and carrying in pockets lol.

    Anyway, this issue has happened before with multiple manufacturers, including sandisk, who ended up using the Phison S11 flash controller which was known for doing exactly this, after a reboot etc i

Slashdot Top Deals

One good suit is worth a thousand resumes.

Close