[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: how to test disk for bad sector



On 29.08.20 14:20, Long Wind wrote:
it pass write test!  i really don't know where installation go wrong.

After the disk itself seems to not cause the problem, at least not because of bad blocks, confirm that the data highways on your motherboard and any involved data caches are all still operating fine.

(1) Read at highest possible speed from the device from which during installation is read from by directing the read data to the device "/dev/zero" ("/dev/zero" is a device which is not doing anything with received input, all received data is immediately after reception simply ignored = forgotten = thrown away):
  sudo dd if=/dev/DeviceReadingFrom of=/dev/zero bs=8M status=progress

(2) Write at highest possible speed a huge amount of data to the disk in order to check if the path to the disk also works free of errors under high load, and likewise also check that the cache inside the disk still performs free of error, by writing data received from the device "/dev/zero" (you will destroy all data on the disk, this begins at the beginning of the disk and will thus first erase a present partition table; the disk will after this test have to be partitioned and filesystems have to be newly created; no old data can be expected to be recoverable from the disk) (in this case, from the device "/dev/zero" there will be received values "0" as long as requested and thus the disks becomes overwritten with zeros):
  sudo dd if=/dev/zero of=/dev/WritingToThisDisk bs=8M status=progress
If you do not cancel this process by STRG+C before, then it will end with an error message when the disk is full and you can see from the last reported status how much data has been written to the disk.

(3) Do the same as above but involving the full data highway from your device you are reading from to the device you are writing to: sudo dd if=/dev/DeviceReadingFrom of=/dev/WritingToThisDisk bs=8M status=progress

If all these tests run through without errors (with test (2) running until the disk is full), then I would say that it is safe to assume that it is not a hardware problem of disks, disk controllers or motherboard. My only left assumptions would then be, that maybe in the BIOS some security feature is activated which interferes with the installation process, or some other peace of hardware lets the system crash when becoming addressed during the installation process, like some incompatible graphics board or some other incompatible peripheral component.

Good Luck!
Marco


Reply to: