Announcement

Collapse
No announcement yet.

HDDs shutting down at 2/3 of the surface when writing on them?

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

    HDDs shutting down at 2/3 of the surface when writing on them?

    I have been lately experiencing this issue in way too many cases which makes me think this is some kind of new failure mode. Basically especially large-size drives (2+ TB, but also seen that for 250GB 2.5" drive too) read the whole surface fine, but when writing to it, at aprox. 2/3 the surface, likely after reaching a particular sector, they just shut down. After power cycling them they appear to work fine, but the same happens if I write to them again.

    Cuting the rotten space off by partition and using only the first 2/3rds of the total capacity appears to work.

    So what the heck is this?
    Less jewellery, more gold into electrotech industry! Half of the computer problems is caused by bad contacts

    Exclusive caps, meters and more!
    Hardware Insights - power supply reviews and more!

    #2
    Re: HDDs shutting down at 2/3 of the surface when writing on them?

    Sounds like a bad board or a bad head connection. Maybe a bad head amp.

    And HDDs usually don't power down by themselves, in the case of giving up on a sector, likely a click-of-death that can only be heard with an ear up real close...

    I had a Maxtor "slimeline" model give a quiet click-of-death before, probably when one of those little bastards failed at SpinRite, IIRC.
    Last edited by RJARRRPCGP; 01-25-2019, 11:52 AM.
    ASRock B550 PG Velocita

    Ryzen 9 "Vermeer" 5900X

    16 GB AData XPG Spectrix D41

    Sapphire Nitro+ Radeon RX 6750 XT

    eVGA Supernova G3 750W

    Western Digital Black SN850 1TB NVMe SSD

    Alienware AW3423DWF OLED




    "¡Me encanta "Me Encanta o Enlistarlo con Hilary Farr!" -Mí mismo

    "There's nothing more unattractive than a chick smoking a cigarette" -Topcat

    "Today's lesson in pissivity comes in the form of a ziplock baggie full of GPU extension brackets & hardware that for the last ~3 years have been on my bench, always in my way, getting moved around constantly....and yesterday I found myself in need of them....and the bastards are now nowhere to be found! Motherfracker!!" -Topcat

    "did I see a chair fly? I think I did! Time for popcorn!" -ratdude747

    Comment


      #3
      Re: HDDs shutting down at 2/3 of the surface when writing on them?

      sounds like whatever supplies current to the head is overheating or has a bad connection either directly or upstream.

      a test,
      write to 25% of the surface,
      wait 5 mins
      write to the next 25%
      repeat until it's complete - or resets.

      btw, what drives are they??

      Comment


        #4
        Re: HDDs shutting down at 2/3 of the surface when writing on them?

        Originally posted by stj View Post
        sounds like whatever supplies current to the head is overheating or has a bad connection either directly or upstream.

        a test,
        write to 25% of the surface,
        wait 5 mins
        write to the next 25%
        repeat until it's complete - or resets.
        Simpler test: write to the LAST block, first, then move towards the START of the volume. If it still fails at "2/3 progress", then you know you're at a different LBA than when writing front to back. (this aims to disprove any surface defects causing the problem)

        If the only tool you have is something like dd(1), then make a note of what block number it fails at, each time. Then, try increasing the block size (not to change the "block number" but, rather, to reduce the overhead, per transfer, in the hope of getting further into the drive in a given amount of time.

        dd if=/dev/zero of=/dev/<whateveryourOSnamesyourdrive> bs=1024K

        vs.

        dd if=/dev/zero of=/dev/<whateveryourOSnamesyourdrive>

        You can also try creating TWO partitions on the drive and see if you can write each completely -- though one at a time.

        Comment


          #5
          Re: HDDs shutting down at 2/3 of the surface when writing on them?

          does the controller/drivers support +2TB drives? Something in my memory is telling me there is an issue with certain certain older stuff and +2TB drives
          Cap Datasheet Depot: http://www.paullinebarger.net/DS/
          ^If you have datasheets not listed PM me

          Comment


            #6
            Re: HDDs shutting down at 2/3 of the surface when writing on them?

            If the BIOS didn't support the size of HDD, then you simply would get a sector-not-found error in DOS or its ilk. (until you shrunk the LBA, which can be done by setting the HPA)

            Or an error of a bad block in the event log, in NT-based Windows. Or simply an error being reported while attempting to write or read past a certain point in Windows.

            Or an OS crash.
            Last edited by RJARRRPCGP; 01-25-2019, 10:00 PM.
            ASRock B550 PG Velocita

            Ryzen 9 "Vermeer" 5900X

            16 GB AData XPG Spectrix D41

            Sapphire Nitro+ Radeon RX 6750 XT

            eVGA Supernova G3 750W

            Western Digital Black SN850 1TB NVMe SSD

            Alienware AW3423DWF OLED




            "¡Me encanta "Me Encanta o Enlistarlo con Hilary Farr!" -Mí mismo

            "There's nothing more unattractive than a chick smoking a cigarette" -Topcat

            "Today's lesson in pissivity comes in the form of a ziplock baggie full of GPU extension brackets & hardware that for the last ~3 years have been on my bench, always in my way, getting moved around constantly....and yesterday I found myself in need of them....and the bastards are now nowhere to be found! Motherfracker!!" -Topcat

            "did I see a chair fly? I think I did! Time for popcorn!" -ratdude747

            Comment


              #7
              Re: HDDs shutting down at 2/3 of the surface when writing on them?

              Been using HD Tune, with one of the drives I actually thought it is good and put it in my IBM server (NAS) I just started writing ton of movies onto it.

              Most of the bad drives were some 3TB Barracudas tested under Vista on SB710 southbridge. Since they have already been ditched by corporate as bad, the problem is definitelly in the drives. Tha last one, WD2500BEVT, also suggests it is nothing with capacity.

              With what can I start writing from the end? Everything I know always starts from the beginning.
              Less jewellery, more gold into electrotech industry! Half of the computer problems is caused by bad contacts

              Exclusive caps, meters and more!
              Hardware Insights - power supply reviews and more!

              Comment


                #8
                Re: HDDs shutting down at 2/3 of the surface when writing on them?

                get a Linux live disk and read up on the DD command.
                or type "man dd" for the onscreen manual

                and remove the pcb's and check the pads for tarnish.

                Comment


                  #9
                  Re: HDDs shutting down at 2/3 of the surface when writing on them?

                  Originally posted by Behemot View Post
                  Most of the bad drives were some 3TB Barracudas tested under Vista on SB710 southbridge. Since they have already been ditched by corporate as bad, the problem is definitelly in the drives. Tha last one, WD2500BEVT, also suggests it is nothing with capacity.
                  Most likely pulled from a shelf where they have been spinning, nonstop, for 3 or more years. Look at the SMART data before even attempting to use! Note that many companies just retire systems after a certain number of calendar months without actually considering how much "useful life" they may have remaining. caveat emptor

                  With what can I start writing from the end? Everything I know always starts from the beginning.
                  YOU probably can't. I write little programs to seek to specific spots on the "raw" device, write, seek to next, etc.

                  OTOH, you could split the drive into two (or more) partitions and verifying that the entire surface can be accessed -- by scrubbing each partition, separately.

                  Comment


                    #10
                    Re: HDDs shutting down at 2/3 of the surface when writing on them?

                    Originally posted by stj View Post
                    get a Linux live disk and read up on the DD command.
                    or type "man dd" for the onscreen manual

                    and remove the pcb's and check the pads for tarnish.
                    Tried scratching it a bit when moving PCBs around. It seems that the drives are HW calibrated for some time now and require re-doing it, or transfering the EEPROM with the data to the other board.

                    Originally posted by Curious.George View Post
                    Most likely pulled from a shelf where they have been spinning, nonstop, for 3 or more years. Look at the SMART data before even attempting to use! Note that many companies just retire systems after a certain number of calendar months without actually considering how much "useful life" they may have remaining. caveat emptor



                    YOU probably can't. I write little programs to seek to specific spots on the "raw" device, write, seek to next, etc.

                    OTOH, you could split the drive into two (or more) partitions and verifying that the entire surface can be accessed -- by scrubbing each partition, separately.
                    Must have some different corporates there, lol. These have been in RAID array, all of them have quite high runtime. I could try testing it the partition way I guess, especially the small WD one which I still have here, and also it is small enough so it won't take half a day
                    Last edited by Behemot; 01-27-2019, 02:36 AM.
                    Less jewellery, more gold into electrotech industry! Half of the computer problems is caused by bad contacts

                    Exclusive caps, meters and more!
                    Hardware Insights - power supply reviews and more!

                    Comment


                      #11
                      Re: HDDs shutting down at 2/3 of the surface when writing on them?

                      [retiring shelfs after a few years of uptime]

                      Originally posted by Behemot View Post
                      Must have some different corporates there, lol. These have been in RAID array, all of them have quite high runtime. I could try testing it the partition way I guess, especially the small WD one which I still have here, and also it is small enough so it won't take half a day
                      I've often encountered boxes full of identical drives and discovered they all had PoHrs of 43502, 43513, 43508, 43507, etc. Or, 34788, 34792, 34777, etc. So, you know they likely came out of the same device (or set of devices).

                      If you assume they are left spinning 24/7/365 (disk farm), then that's ~5 years (or 4 years for the 30K's) of continuous uptime.

                      I take a peak at the number of remapped sectors and, if not 0, consider the drive as scrap.

                      Comment


                        #12
                        Re: HDDs shutting down at 2/3 of the surface when writing on them?

                        No joy, shuts down in the very same manner. It also couldn't be anything like that mysterious supermagnetism as this bloody 250gig 2.5" drive is way too old with too low capacity per square unit (unlike the 3TB drives). I think that they just started screwing up big time, after only 30 years of making hard drives. What other explanation there could be? What could be so bad on the surface that it makes the electronics go crazy and shut down instead of just marking the area bad and reallocating the sectors?
                        Less jewellery, more gold into electrotech industry! Half of the computer problems is caused by bad contacts

                        Exclusive caps, meters and more!
                        Hardware Insights - power supply reviews and more!

                        Comment


                          #13
                          Re: HDDs shutting down at 2/3 of the surface when writing on them?

                          I don't think that a "bad surface" makes the HDD spin down. That one would be the first doing that. In extreme cases the platter looked like one had sanded it and fine metal particles inside the HDD while opening. So the HDD had actually seized because of the fine metal dust.
                          What you are describing is different. I don't think it is anything mechanical either, because it takes quite a bit of force to make the HDD stop spinning. Never the less, I wouldn't trust that HDD and throw it out.

                          Comment


                            #14
                            Re: HDDs shutting down at 2/3 of the surface when writing on them?

                            Originally posted by CapLeaker View Post
                            I don't think that a "bad surface" makes the HDD spin down.
                            No, but the head assembly may be encountering some problem trying to extend across the media to that point.

                            Regardless, its trivial to TEST if this is indeed the case -- and not a consequence of run-time, etc. -- just by deliberately positioning at that place (vicinity) at the START of your test (instead of waiting a long time to get to that place "eventually")

                            Comment


                              #15
                              Re: HDDs shutting down at 2/3 of the surface when writing on them?

                              Originally posted by Curious.George View Post
                              No, but the head assembly may be encountering some problem trying to extend across the media to that point.)
                              Even if that is the case I don't believe the HDD will spin down for that either. It just wouldn't read and the heads go back and forth.

                              Comment


                                #16
                                Re: HDDs shutting down at 2/3 of the surface when writing on them?

                                Originally posted by CapLeaker View Post
                                Even if that is the case I don't believe the HDD will spin down for that either. It just wouldn't read and the heads go back and forth.
                                Do you really think the controller (in the disk) will sit there tweaking the head assembly (in and out) FOREVER without taking some sort of action?

                                (I may drag out a discardable disk to test this -- pull the covers off and cram a Q-tip in the head assembly to prevent it from moving beyond a certain point just to see what it does...)

                                Comment


                                  #17
                                  Re: HDDs shutting down at 2/3 of the surface when writing on them?

                                  clicky-click - yes, most drives will just seek forever!!
                                  only drives intended for raids will quit and send a fail message instantly.

                                  Comment


                                    #18
                                    Re: HDDs shutting down at 2/3 of the surface when writing on them?

                                    Originally posted by Curious.George View Post
                                    Regardless, its trivial to TEST if this is indeed the case -- and not a consequence of run-time, etc. -- just by deliberately positioning at that place (vicinity) at the START of your test (instead of waiting a long time to get to that place "eventually")
                                    Did you read the thread before posting this, at all?
                                    Less jewellery, more gold into electrotech industry! Half of the computer problems is caused by bad contacts

                                    Exclusive caps, meters and more!
                                    Hardware Insights - power supply reviews and more!

                                    Comment


                                      #19
                                      Re: HDDs shutting down at 2/3 of the surface when writing on them?

                                      Originally posted by Behemot View Post
                                      Did you read the thread before posting this, at all?
                                      Yes.

                                      when writing to it, at aprox. 2/3 the surface, likely after reaching a particular sector, they just shut down
                                      Cuting the rotten space off by partition and using only the first 2/3rds of the total capacity appears to work.
                                      Note that "reaching a particular sector" doesn't have to mean that there is a surface defect. Rather, it can mean that the software is processing a certain LBA number (and has a bug). Or, the head assembly is being tasked with displacing a certain amount INTO the surface. Or, ...?

                                      If you're always "getting there" from the start of the volume, then a certain amount of elapsed time (at high duty cycle) has passed.

                                      [I had a DSL modem that would drop connections after a fixed amount of traffic -- regardless of the nature of the traffic. An internal counter would overflow and it would drop the connection. Try again. And again. And again. Nope!]

                                      To remove the apparent physical configuration of the drive at the time of failure (that "spot") from the equation, you need a way to seek to that point WITHOUT having to "start from 0" each time (to rule out a duty-cycle related problem).

                                      If, instead, you started your test in that general vicinity, then a problem with that "spot" on the media (whether it is surface, head assembly or whatever) should manifest VERY QUICKLY! If, instead, the process continues to a point that is obviously BEYOND the original hangup location, then you can rule out that part of the surface and the other related electronic and mechanical aspects of accessing that "spot" -- cuz you were able to do so by starting from a different location/offset!

                                      E.g., assume there are 100 sectors on the volume and that it's crapping out at sector #66. If you cut the drive into two partitions -- [0,49] and [50,99] -- then you would expect to be able to wipe the first partition non-stop, all day long (i.e., not thermal related, either!)... IF the problem was related to "66" in some way (electrically, mechanically, software, etc.)

                                      OTOH, if you tried to wipe the second partition, you would expect the failure to manifest VERY QUICKLY... after just "~16" sectors. But, if the process continues to completion for the entire second partition -- [50,99] -- then it's hard to argue that there is something physically/electrically/mechanically wrong with "66"!

                                      I see nothing in the thread to suggest that you've tried "skipping the first half of the volume" (in my example). Rather, you've assumed that skipping the last THIRD of the volume "fixes" the problem:

                                      Cuting the rotten space off by partition and using only the first 2/3rds of the total capacity appears to work.
                                      This doesn't eliminate "66" from the list of possible causes. OTOH, if you repartition the drive with the values I suggested (by way of example), then you should be able to endlessly write the first (non-66) partition OR the last partition without any problem IF it is unrelated to "66".

                                      Comment


                                        #20
                                        Re: HDDs shutting down at 2/3 of the surface when writing on them?

                                        Guess you must have ACCIDENTALLY missed posts #10 and #12.
                                        Less jewellery, more gold into electrotech industry! Half of the computer problems is caused by bad contacts

                                        Exclusive caps, meters and more!
                                        Hardware Insights - power supply reviews and more!

                                        Comment

                                        Working...
                                        X