0

This is an old flash disk that I have not used in a while. It was a small drive I was planning to use for TinyCore Linux. I plug it in, and it does not show up in KDE Partition Manager.

I query lsblk, lsusb and usb-devices in bash (Ubuntu 20.04.2) twice each , once with and once without it plugged in. I manage to find that it mounts at /dev/sdc , so i proceed to do a sudo fdisk /dev/sdc , but get fdisk: cannot open /dev/sdc: Read-only file system .

Alright, I do sudo hdparm -r0 /dev/sdc and get :

/dev/sdc:
 setting readonly to 0 (off)
 readonly      =  0 (off)

Which appears to state that I've been successful in changing the readonly variable to False ?

Also, KDE Partition Manager now sees it but says No valid partition table was found on this device. But any attempts at formatting result in immediate There were errors while applying operations. Aborted. with the dialog box below displaying :

Create a new partition table (type: msdos) on ‘/dev/sdc’ 
Job: Create new partition table on device ‘/dev/sdc’ 
Command: sfdisk /dev/sdc 
Create new partition table on device ‘/dev/sdc’: Error
Create a new partition table (type: msdos) on ‘/dev/sdc’: Error

Calling fdisk on /dev/sdc still gives the same read-only error message. Sigh.

I thus conclude something has gone wrong with the drive , perhaps it's days of being written to are over. Okay, so does it have any data on it , since it can be read ?

I run intdump on it (hexdump should yield similar results) and I get an endless stream of 0s until I decide to ^C and kill the program.

Any idea what could have happened and if there's anything else that can be tried to recover or revive it ? Or is it dead, as I presumed ?

EDIT :

On Ubuntu , just as on macOS and Windows 7 machines that I tried on, this 2 Gib drive seems to show up as a 7.5 Gib drive. Is this normal or does it indicate some low-level error that can be undone to revive this disk ?

EDIT 2:

I attempted reformatting/erasing on Windows 7 SP1 as well as macOS 11.1 's DiskUtility. Of course, it failed, with Windows also saying it was "read-only".

EDIT 3:

Sadly , @Tetsujin suggested and perhaps got this question closed because apparently they found it similar to 1125282 . Untrue. A sincere reading of the question in full is enough to indicate that the reality is quite to the contrary. Here is how :

I have tried the usual methods and the hypothesis presented in those answers and that does not check out here. "inserting the drive into another computer" does not work, as mentioned in edits 1 & 2. "This behavior is typical of flash drive controllers when they detect ... too many bad blocks" and "any data on the drive is still accessible" - well, all bytes are 0 , as scanning the device reveals, and "overriding this" does nothing. "Ensure ... "Write-Protect Switch" is not locked" There is no such thing. "low-level format tools may help" - nope.

An Ant
  • 77
  • 1
  • 12
  • 3
    The USB stick is dead. Once writes no longer work to it then the only thing you can do is to replace it. "read-only" does not necessarily mean that you see a write-protect bit you can modify, it can mean that any writes seem to succeed but upon inspection they apparently failed and hence are "read-only". If low level tools do not work then replace it. – Mokubai Feb 12 '21 at 13:32
  • @Mokubai Thanks , I was thinking the same too... any idea why it appears as a 7.5 Gib drive ? – An Ant Feb 12 '21 at 13:35
  • Could be that the controller defaults to that size when the flash vanishes, or some corruption in the identification of the flash chip, or anything. In either case unless you can physically replace the bad controller or flash chips there is not much you can do. – Mokubai Feb 12 '21 at 13:38
  • If it won't format, it's dead, no matter how you argue it. These things are just not worth fighting once they start to fail. I used to get through literally thousands of them for work. – Tetsujin Feb 12 '21 at 14:40
  • @Tetsujin Maybe I worded myself wrong. I'm not arguing anything except that my question is not a duplicate, which is objectively true. And thus, it shouldn't have been closed ! – An Ant Feb 12 '21 at 15:46
  • If it leads to the same conclusion/solution, it's better to link as a duplicate. It's not a punishment, it's a way that many more searchers can find an acceptable solution from differently-worded searches. Over time, the 'master' question may change, or the answers on any 'master' duplicate are more likely to be updated with newer 'fixes' if such become possible. Also on occasions, such as this one, the question was written to actually **become** the master. – Tetsujin Feb 12 '21 at 15:53
  • @Tetsujin I politely disagree. I don't think the question or problem was in any way similar enough to close for duplicate, and I do not agree that it is a similar issue or has similar solutions at all. We will have to agree to disagree. – An Ant Feb 12 '21 at 15:56
  • Test: put it in 3 different machines. Format (or attempt to write to it by any method, eject, insert, Does anything actually change? If no, it's a duplicate. Firmware has tripped the write lock. – Tetsujin Feb 12 '21 at 16:01
  • You have SLIGHTLY different symptoms, but the same problem. As Bones said, "he's dead, Jim". Your wishing it otherwise doesn't change it. – music2myear Feb 13 '21 at 04:42

0 Answers0