Complete SSD failure: Dell and HPE release firmware against 40K hour bug

Discussion in 'Frontpage news' started by Hilbert Hagedoorn, Mar 25, 2020.

  1. Hilbert Hagedoorn

    Hilbert Hagedoorn Don Vito Corleone Staff Member

    Messages:
    38,580
    Likes Received:
    7,148
    GPU:
    AMD | NVIDIA
  2. cryohellinc

    cryohellinc Ancient Guru

    Messages:
    3,236
    Likes Received:
    2,480
    GPU:
    RX 5700 XT/GTX 1060
    "bug"
     
    DG21 and Neo Cyrus like this.
  3. Aura89

    Aura89 Ancient Guru

    Messages:
    7,876
    Likes Received:
    1,043
    GPU:
    -
    "troll"
     
    insp1re2600 likes this.
  4. MBTP

    MBTP Member Guru

    Messages:
    141
    Likes Received:
    10
    GPU:
    Sapphire RX590
    Planed obsolescence now is called "bug".
    No one knows, no one will ever know... and things will eventually be forgotten..
     

  5. HeavyHemi

    HeavyHemi Ancient Guru

    Messages:
    6,963
    Likes Received:
    959
    GPU:
    GTX1080Ti
    Except for the number of hours, isn't this precisely the same as the last firmware bug?
     
  6. cryohellinc

    cryohellinc Ancient Guru

    Messages:
    3,236
    Likes Received:
    2,480
    GPU:
    RX 5700 XT/GTX 1060
    Oh, but of course it was a bug in the firmware and companies involved would never ever do anything such as that.
     
    DG21 likes this.
  7. HeavyHemi

    HeavyHemi Ancient Guru

    Messages:
    6,963
    Likes Received:
    959
    GPU:
    GTX1080Ti
    Simple logic would tell you they would not design a bug where the failure mode was obvious and identical for 100% of the product. That's like playing Russian Roulette with no empty cylinder.
    That observation is irrelevant to what any other company or entity may have done. Which you know, is irrelevant.
     
  8. nevcairiel

    nevcairiel Master Guru

    Messages:
    696
    Likes Received:
    235
    GPU:
    MSI 1080 Gaming X
    And then they fix it, months in advance of the bug even occuring? Sounds illogical.
     
  9. Aura89

    Aura89 Ancient Guru

    Messages:
    7,876
    Likes Received:
    1,043
    GPU:
    -
    Unless you have proof, there's no point throwing around accusations.

    That's what a conspiracy theorist does, and there's no point in conspiracy theories anywhere. You either have proof, or you don't. And if you want to get proof, then get the proof, then post it on the internet or wherever it is needed.

    Logically, this is a bug. Illogically, it is planned obsolescence and purposeful.

    "Oh hey, lets create a planned failure point in our product, that way when it dies, people will think "I loved that product that just died on me! Lets get another one from the same manufacturer so it can die too!" but wait, lets fix it as well so our plan to get more sales doesn't happen!"

    [​IMG]

    ^ This is not necessary or needed. Only makes you look like a fool, and derail topics/conversations.
     
  10. TieSKey

    TieSKey Member Guru

    Messages:
    163
    Likes Received:
    52
    GPU:
    Gtx870m 3Gb
    So, where is your proof? (No, personal "logic" is not a proof)

    People calling every yet unproven theory a "conspiracy" and making fun of others for lack of hard evidence, while at the same time failing to provide any hard evidence that debunks the theory (like... space photos showing the earth is not flat) are so funny... they even think they are somewhat superior to those they make fun of.


    U now, even in the legal/justice system, attorneys and detectives must first form a theory (a conspiracy?) and then work to get proof, they won't magically stumble unto proof and only after, formulate the theories (might happen but is unlikely).
    Now, attorneys don't go into forums posting their yet unproven theories, right, but people here are not justice officials, we just share our thoughts, some even try to ask or find information that can prove or debunk their theories. Even if they don't, making fun of other people "gut feeling" without providing any irrefutable proof is quite silly in itself.
     
    DG21 likes this.

  11. Crazy Serb

    Crazy Serb Member Guru

    Messages:
    152
    Likes Received:
    17
    GPU:
    270X Hawk 1200-1302
    It is well known fact that printers never stopped printing after certain amount of printed pages and required new cartridges even if "bugged" ones were not half-empty... It is also known that HP never made any printers...

    I have no idea why would people think that HP of all companies wants their SSDs to stop working after specified amount of power-on time. Nothing similar ever happened before.
     
  12. Ne1l

    Ne1l Member

    Messages:
    24
    Likes Received:
    9
    GPU:
    465 Sli
    I worked for one of the mentioned companies in manufacturing for more years than I care to remember, a dodgie firmware with planned obsolescence is nothing in the grand scheme of things..

    The fact it effects both DELL and HPE could mean they are innocent or both of them have no morals and sandisk was just doing what was asked.

    Firmware parameters are strictly controlled and supplied by the customer, that's why a "15K 800GB DP SAS" drive can be made by any number of HDD suppliers but still have the same PN. They all need to perform the same so they can be mixed in a raid array inside a server and support certain RAID 'smart' features.

    Corporations have money to burn and OEM's are more than happy to burn them to get at the money..
     
    Last edited: Mar 26, 2020
    Neo Cyrus and cryohellinc like this.
  13. HeavyHemi

    HeavyHemi Ancient Guru

    Messages:
    6,963
    Likes Received:
    959
    GPU:
    GTX1080Ti

    I worked in the industry for decades, was the ATE manager for on of the top 5 semi firms on the planet. This claim "a dodgie firmware with planned obsolescence is nothing...." is nonsensical. Never heard of it from any credible semi company. The closest I can think of was the logic board issues on iMac and Nvidia laptop GPU'. Design flaws, however unintentional, they tried to hide them.
    The most important fact here is: nobody would program a firmware would do this intentionally to profit in this manner so obviously. As I said, it would be like playing Russian Roulette without an empty cylinder. Your chances of discovery are 100% because they ALL FAIL FOR THE SAME THING AT THE SAME TIME.
     
    Aura89 likes this.
  14. Aura89

    Aura89 Ancient Guru

    Messages:
    7,876
    Likes Received:
    1,043
    GPU:
    -
    I'm not the one claiming anything, so the fact you think someone needs proof to not claim something is downright sad.
     
  15. cryohellinc

    cryohellinc Ancient Guru

    Messages:
    3,236
    Likes Received:
    2,480
    GPU:
    RX 5700 XT/GTX 1060
    Hmmm, don't have proof - my argument is wrong.
    You don't have proof either - your argument is right, and you don't need to prove anything.
    That sounds foolish to me.
    Counter
    [​IMG]

    Sure, 100% is too obvious, however, there are countless examples from various industries. Concept of "designed to fail" where after a certain amount of cycles equipment would break is spread across our world and isn't something new.
    Manufacturers would always try to implement it. It's a risk that is hard to prove (depending on a product of course) and many are willing to take it.
     
    DG21 likes this.

  16. Ne1l

    Ne1l Member

    Messages:
    24
    Likes Received:
    9
    GPU:
    465 Sli
    If it's nonsensical how come this isn't the first time we've seen drives dying from a pre-specified number of hours? SSD's 'usually' have 'over provisioning' chips which gets used as issues arise and bad sectors are listed in the 'defective-sector table', once the spare capacity is used up the drives 'should' fail into read only mode, not die totally! There is absolutely no reason to set a die date by hours? That's nonsensical! << unless it can help generate a future sale :)

    40,000 hours is just short of the usual 5 year extended life support contracts, extending support afterwards costs soo much, buying new hardware is more cost & tax efficient. Most corporations decommission their hardware before 5 years, as there is no 'depreciation of assets' tax benefits left and the hardware ends up on ebay and therefore someone else's issue.

    OEM's should be forced to show how many refunds and discounts (against future orders obviously) they give to customers as hush money for failing to deliver the quality the customer paid over the odds for.
     
  17. Reardan

    Reardan Master Guru

    Messages:
    355
    Likes Received:
    42
    GPU:
    GTX 2080
    Yeah they probably should give refunds/discounts, regardless of first time or not. This is a LOT of trouble if you have a lot of these servers. And the downside is potentially catastrophic
     
  18. Ne1l

    Ne1l Member

    Messages:
    24
    Likes Received:
    9
    GPU:
    465 Sli
    They get out of giving refunds on minor issues (to them); if you have a support contract then as part of that agreement you must install *** CRITICAL *** updates, you snooze you lose.

    Example of a *** critical *** update: https://support.hpe.com/hpsc/swd/public/detail?swItemId=MTX_cc50fd630ff3498a843c2ea723#tab4 - OEM's are loving that fact they can blame Intel for customers experiencing reduced performance on old hardware (spectre meltdown microcode changes) so they can sell them shiny new hardware.
     
  19. 0blivious

    0blivious Ancient Guru

    Messages:
    2,897
    Likes Received:
    430
    GPU:
    5700XT/ 1070/ RX580
    40,000 hours = 1,666 days and 16 hours
    (or)
    1,666.667 days

    Fascinating.
     
  20. Aura89

    Aura89 Ancient Guru

    Messages:
    7,876
    Likes Received:
    1,043
    GPU:
    -
    Not sure how people don't understand you need proof to validify a claim, you don't need proof to not.

    It's about the oddest arguement i've ever heard.

    I'm not making an arguement. I'm not making a claim. I have no proof because there's zero reason to have proof for a lack of a claim. Really can't get more simple then that.

    You on the other hand decided to make a claim with zero proof. That's called a conspiracy theory.

    Here let me put it in a way even the simplest person could understand.

    If you bring someone to court claiming they stole from you, it's up to YOU to prove it. If you don't have proof, your court claim is lost and you'll likely have to pay for their court fees. What do they have to prove? Absolutely nothing. They only have to prove their innocence if you HAVE proof. And if during this court case you stated this long story about how they stole from you, but then the lawyer (me in this instance) asked you for proof, you don't get to say "well, prove it didn't happen!". No, that's not how anything works, you don't prove something didn't happen, you prove that it did, and the people claiming it are the ones to do it.

    The world may be going crazy right now, but left is still left and right is still right, and having to prove a lack of a claim still makes zero sense.

    ....Unless you want me to claim you're biased @cryohellinc because you and i have worked in the past for a competitor to Dell and HP and the only reason you're making these statements is because you still work for them and are trying to drive business over to the company you work for and that it completely lines up with your personality that i know you have since again, i know you and have worked for you....

    Oh, want me to prove the statement i just said? No, you prove that it's not true.

    Lets all go back to the salem witch trials since that apparently is where you wish to be.
     
    Last edited: Mar 27, 2020

Share This Page