Intel acknowledges complaints about 320 Series bug

As much as the folks at Intel like to talk up the reliability of their solid-state drives, firmware bugs still crop up every once in a while. This time, Intel has acknowledged some user reports about problems with the freshly released 320 Series SSDs. Complaints have piled up in Intel’s Support Community message board in a thread entitled, "Is 320 firmware buggy?"

Simply put, the users allege that 320 Series SSDs are wiping themselves clean in the event of a system power loss. One user complaint dated July 1 reads:

Had the 8MB issue happen to me as well with my 160 GB 320. On Monday, I cloned the drive using the Intel Data Migration Software and installed the SSD in my new laptop. Monday night there was a software problem and I had to hold down the power button to cause the system to reboot. On reboot, drive was not recognized. Went though various procedures to try to get it to work. Nothing. Restarted the computer about 30-40 times. Eventually, the computer booted from the SSD. It ran fine for a while, and then I hibernated the computer. Took computer out of hibernation, drive not recognized.
After that I put in the original drive (a HDD) that laptop shipped with, machine booted right up. Attached the SSD to the laptop over USB 3.0 with the adapter the drive came with (Retail Box). Drive didn’t show in "Computer". Right clicked on Computer –> Manage –> Disk Management. Drive shows as unformatted, no partition table, with an 8MB capacity. The original vendor is replacing the SSD, because I bought it less than 30 days ago.

Intel quietly responded earlier this week with a Support Community message board thread of its own. Here’s what the company’s saying:

SSD power loss report updates
Intel is aware of the customer sightings on Intel SSD 320 Series. If you experience any issue with your Intel SSD, please contact your Intel representative or Intel customer support (via web: www.intel.com or phone: www.intel.com/p/en_US/support/contact/phone) . We will provide an update when we have more information.

Alan

Intel’s NVM Solutions Group

This isn’t the first Intel SSD users have had to contend with data-wiping firmware issues. Two years ago, Intel had to halt shipments of second-generation X25-M SSDs because of a similar bug. That bug caused data loss in systems whose BIOS passwords were either changed or disabled.

Comments closed
    • cynan
    • 8 years ago

    So SSD Gurus, in light of this what brand/model of SSD is recommended for the best balance of bleeding edge performance and reliability these days?

      • NeelyCam
      • 8 years ago

      Intel 510?

    • Spotpuff
    • 8 years ago

    This is a pretty atrocious bug. Hibernate/have to hold down power button = formatted drive? This combined with the previous bug makes me hesitant to upgrade to an SSD.

    • ronch
    • 8 years ago

    Although it’s obviously not good when bad things slip past quality control, things like this happen and I applaud companies who have the balls to admit the problem and take the necessary steps to help people affected by the problem. It’s been a while since Intel tried to shove the Pentium FDIV bug under the rug, and the recent Sandy Bridge chipset bug (B3) demonstrated just how Intel has evolved. I hope AMD gets to this level someday, given how they still sell Phenoms with the TLB bug even after it’s been publicized just to make every penny they possibly could.

      • poulpy
      • 8 years ago

      [quote<]given how they still sell Phenoms with the TLB bug even after it's been publicized just to make every penny they possibly could.[/quote<] Smells like biased to me: Exactly which Phenoms on the market today still have the TLB bug?! Unless I missed something the bug got fixed with the stepping B3 of the Phenom I in 2008. Also AFAIK no regular user ever hit the TLB bug, as you needed to be in a very particular setup and workload (e.g. lots of VMs concurrently seriously crunching numbers) for it to occur. Therefore I wouldn't really classify this at the same level as mom-an-pop having their HDD wiping itself clean after a power loss.

        • ronch
        • 8 years ago

        You sound like a fanboi to me, given how you seem to have forgotten history.

        AMD still SOLD B2 Phenoms even after they were publicized as having the TLB bug. Perhaps it no longer applies today, but I remember still seeing those chips being sold at TigerDirect long after the disclosure.

        PS – I may sound biased against AMD, but I’ve used [u<]way more[/u<] AMD processors than Intel processors in my entire PC computing life. I root for AMD, but I'll gladly point out the areas where they are weak and need improvement, and commend Intel for doing something great even if I have no love for them, because I'm not blindly biased to a brand, which you seem to be. PSS - Oh, I found one for you. Buy your copy before stocks run out. Trust me, you won't run into the bug. [url<]http://www.tigerdirect.com/applications/SearchTools/item-details.asp?EdpNo=7329858&CatId=3637[/url<]

          • Chrispy_
          • 8 years ago

          The TLB bug was fixed for existing silicon in software. No harm can come to anyone as a result of using an old B2 Phenom, unless they somehow manage to avoid all driver and BIOS updates AND bought their B2 Phenom before the fix was issued.

          In instances where the software workaround kicks in, performance will be lower than the same model on a B3 revision, but that’s still not an issue that should prevent sales. This is a very different issue to willingly selling a product that will knowingly cause data loss to a percentage of users, no matter how small that percentage, and that is why Intel behaved as they did.

          From AT:
          “AMD gave us the two confirmed situations where the TLB erratum would rear its ugly head in real world usage:

          1) Windows Vista 64-bit running SPEC CPU 2006
          2) Xen Hypervisor running Windows XP and an unknown configuration of applications

          AMD insisted that the TLB erratum was a highly random event that would not occur during normal desktop usage and we’ve never encountered it during our testing of Phenom.”

    • DrCR
    • 8 years ago

    I wonder if only the partition table or the MBR is being blown. I’ve had USB Thumbdrivess lose their MBR/partition table after being yanked out — kind of like the sudden power outage discussed in the article.

    I wonder if [url=http://www.cgsecurity.org/wiki/TestDisk<]testdisk[/url<] would recover the partition table for these users. Perhaps I'm nerdy, but I always run a [b<]sfdisk -d /dev/sdx > partitiontabledump[/b<] and a [b<] dd if=/dev/sda of=mbr_backup.bin bs=512 count=1 [/b<] for all my new installs. Better safe than sorry -- It's saved me a lot of time more than once. (color bbcode not working. using bolding instead)

      • cynan
      • 8 years ago

      A question from the ignorant:

      What does a partition table backup get you that a regular disk image does not?

      And what does [b<]dd if=/dev/sda of=mbr_backup.bin[/b<] do? Just a backup of the boot record? Why is this important? [i<] Edit:[/i<] OK, so that was more than 1 question.

        • helix
        • 8 years ago

        Diskdump input-file=/devices/yourharddrive output-file=where-to-dump-it bytes-at-a-time=512 how-many-chunks=1.
        It reads the first 1 x 512 bytes (the master boot record) from the harddrive and writes it to a file.

        The partition table backup is a short plain text file that can be read by sfdisk to write an identical partition table to this or another disk.

        Total size ~1kB, which is a lot less than what a disk image would require.

    • j1o2h3n4
    • 8 years ago

    Nobody knows about this. I mean, intel brand has always given me the feel of assurance, confidence that although it might not be the fastest/best in this case, SSD, but i can stand firm on its quality. This not the case. If i were to lose all my whole years data, just coz a system reboot, i would go Mad, seriously. This is a very Huge Critical Error. Sadly, it will just pass by quietly.

      • aceuk
      • 8 years ago

      Restore from backup? This applies to HDDs too (because they also fail).

      • helix
      • 8 years ago

      An external harddrive to put backups on is likely cheaper than prescription neuroleptica.

    • fuzzhead
    • 8 years ago

    Buy a UPS!

    I have a 120g in a laptop and a 160g in the
    desktop w/12Gb of memory. No problems so far.
    Laptop has built in UPS(battery). Desktop on APC,
    you would crazy in Florida not to have a UPS. Brown
    outs, Black outs, Hurricanes.

    Fuzzhead

      • xeridea
      • 8 years ago

      Yep, the solution to having a buggy Intel SSD is buy a UPS… you could just buy a reliable SSD from a company that doesn’t have buggy hardware in every department.

      • syndicatedragon
      • 8 years ago

      Did you read the user complaint? He had to hard shutdown his laptop. PCs can be in that situation too. A UPS is going to absolutely useless for that situation.

      Of course you shouldn’t hard shutdown or reset a computer unless absolutely necessary, but stuff happens and you shouldn’t lose an entire drive’s worth of data when it does.

    • xeridea
    • 8 years ago

    Intel has been shipping a lot of buggy hardware. They had buggy SSDs in the past, and just barely got over the Sandy Bridge recall. With all the billions of dollars they are making on their high priced hardware, you would think they could afford to hire a QA team. Reminds me of Apple who claims to be virus free, but then tell their repair shops to not fix your viruses when you get them, and not even confirm the existence of a virus… kuz that would hurt their image, and cut into their AppleTax. And claiming to have better built computers when statistically they are less reliable than other major PC vendors.

    • 5150
    • 8 years ago

    So who are we down to with reliable drives? The C400 and 510?

      • xeridea
      • 8 years ago

      OCZ Vertex 2 is quite good. Don’t know bout the Vertex 3.

        • 5150
        • 8 years ago

        I stopped reading at OCZ.

    • derFunkenstein
    • 8 years ago

    Well that’s a bummer. Guess I’ll just keep backing up.

    [url<]http://communities.intel.com/thread/22227?start=60&tstart=0[/url<] If you can secure erase, you can at least restore a drive image after the fact, it seems. It's not permanently dead, just kind of temporarily comatose.

    • swaaye
    • 8 years ago

    I read about this kind of thing with other SSDs too. SSDs seem to need more time to get the reliability up for whatever reason.

      • [SDG]Mantis
      • 8 years ago

      I love reading messages like this one after buying an Intel 320, 120 GB, model and then having SMART failure messages and no booting 10 days into a new system. But since it was so close to when I bought the drive, I just went to NewEgg with this one for a replacement.

      • Ushio01
      • 8 years ago

      I am so glad I bought an enterprise SLC SSD back in july 08, problem free for three years seems really exceptional compared to MLC SSD’s.

      • axeman
      • 8 years ago

      yeah good old mechanical drives, those never fail.

        • Forge
        • 8 years ago

        He didn’t say they never fail, and I agree with him, because in many years, many mechanical hard disks, many failures, both for and without cause, I’ve never had one go submarine over a cold power off like these Intel ones do. This is a new and exciting type of failure, pretty much unique to SSDs.

        It makes sense that they’d have new types of failures, since others are no longer applicable. No moving parts means no head crashes, but there’s also no platters to remove and transplant.

        He’s not saying SSDs are BAD, but they are different.

          • axeman
          • 8 years ago

          Man, you just ruined my sarcastic trolling with facts and logic! Be gone with you!

Pin It on Pinterest

Share This