Possible broken SSD? BtrFS Read-only business.

English Hardware
1

Hello good people!

Some manner of witchcraft is going on here. I had a dual-boot system with Windows 10 and Leap 15.2. Both had their own SSD. Windows kept crashing giving me no error code. Just freezing as it were. OR it gave me a BSOD to short so it couldn’t be read but sometimes I got the error “CLOCK_WATCHDOG_TIMEOUT”. I assumed this was because my CPU was overclocked. So I unclocked it. Windows kept crashing. BUT OpenSUSE is stable and was even before the un-clocking. So I decided to run OpenSUSE only. I formated the former Windows 10 drive to BtrFS. It mounts fine but it is forced into read-only and I get the error “Caught signal #127: “/sbin/btrfs filesystem label ‘/dev/nvme1n1p4’ ‘Steam’”” when I try to re-lable it to “Steam” as you can see.

Could it be the my PCs problem is a faulty SSD rather then the CPU? Or can I somehow get the drive out of this “read-only” business?

It should also be mentioned that I’m not a master of computers.

Best regards
Hoff

2

Yes, that’s possible.

3

Ok! Thank you! A new SSD is cheaper then a new CPU. Is there a way to be certain?

4

Try to use XFS or ext4 with this SSD.
Check contacts and cooling.

5

Thank you for your reply!

I got this error message when I tried to format it as an XFS:

command '/usr/sbin/xfs_admin -L 'Windows 10' '/dev/nvme1n1p4'' failed:

stderr:
Usage: xfs_admin -efjlpuV] -c 0|1] -L label] -U uuid] device

exit code:
2

I continued despite the error and it is now XFS. But I’m guessing this is more evidence that it is broken? As to cooling it is dangerously close to my GFX card. It’s an M.2 drive mounted on the motherboard directly above the GFX card.

6

Ok somethings not right. Every partition I make on any SSD is “read-only”. How can I fix this please?

7

When there are problems during mount (but it can still be done). a fall-back to read-only is not unknown.

Thus what errors during mount? umount and mount from the terminal so that you see messages. Also

journalctl -f

as root from one terminal and then mounting in another will reveal logging.

8

Thank you for your reply! How do I mount/unmount them through the terminal?

EDIT: I could just google it :stuck_out_tongue: Sorry

9

What if I mounted them from root and edited privileges from there?

10

This I do not understand.

But to your earlier question: how do I mount from the terminal.
I hadn’t read all the posts here, but answered on the: hey it is mounted read-only, how comes.

My question is thus, how did you mount it (when not already from the terminal)?

In any case, you can also start the terminal (as root) with the journalctl -f and then mount it in the you did it all the time.

11

Hi!
I meant I’d log in as root, mount and change the privliges for the drive so that Other can also write on it.

Mounted thru Dolphin

Thank you!

12

When you mean with “log in as root” that that is on a terminal, that is OK (Easiest to use from the KDE menu > System > Terminal - Super user mode).
And then mounting it from there with the mount statement is what I mean. But there is no need whatsoever to change ownership of anything, we are trying to find out if it can be mounted, if we can, then we can go further.

13

I’m sorry that I’m so unfamiliar with Linux. But I think this has the information you sought? I have not changed ownership of any drive. On one drive I just added “write” privilege to “Other” on one drive. Root remains the owner.

-- Logs begin at Sun 2021-03-07 16:39:40 CET. --
Mar 07 20:20:56 50-3E-AA-E3-41-E7 polkit-kde-authentication-agent-1[1787]: Completed:  true
Mar 07 20:20:56 50-3E-AA-E3-41-E7 polkit-kde-authentication-agent-1[1787]: Finishing obtaining privileges
Mar 07 20:20:56 50-3E-AA-E3-41-E7 polkit-kde-authentication-agent-1[1787]: Listener adapter polkit_qt_listener_initiate_authentication_finish
Mar 07 20:20:56 50-3E-AA-E3-41-E7 polkit-kde-authentication-agent-1[1787]: polkit_qt_listener_initiate_authentication_finish callback for  0x55e3aef816a0
Mar 07 20:20:56 50-3E-AA-E3-41-E7 polkit-kde-authentication-agent-1[1787]: Finish obtain authorization: true
Mar 07 20:20:56 50-3E-AA-E3-41-E7 polkitd[1052]: Operator of unix-session:1 successfully authenticated as unix-user:root to gain TEMPORARY authorization for action org.freedesktop.udisks2.filesystem-mount-system for system-bus-name::1.152 [/usr/bin/dolphin] (owned by unix-user:hoff)
Mar 07 20:20:56 50-3E-AA-E3-41-E7 kernel: BTRFS info (device nvme1n1p1): disk space caching is enabled
Mar 07 20:20:56 50-3E-AA-E3-41-E7 kernel: BTRFS info (device nvme1n1p1): has skinny extents
Mar 07 20:20:56 50-3E-AA-E3-41-E7 udisksd[1791]: Mounted /dev/nvme1n1p1 at /run/media/hoff/55e26897-c1d6-42cd-8874-cf06a82475d0 on behalf of uid 1000
Mar 07 20:20:56 50-3E-AA-E3-41-E7 kernel: BTRFS info (device nvme1n1p1): enabling ssd optimizations
14

This is getting a bit difficult.

I saw your post about a file system that was mounted ro where you did not expect that (means you did not mount it ro on purpose). I then gave you some information on how to catch the messages/logging during the mount, so you may see what the problems are. It looks that I was too eager to help you and now I started reading from the beginning of the thread. My idea is that we have to start from the begin.

You have a mass-storage device that may be troublesome. To check that you created a new file system upon it (but you did not say on what, the device, a partition? and you also did not say how) and choose Btrfs for that. This is a bit strange, why Btrfs for what probably will be only a storage for data, not for a system.

People then advised you to try XFS or Ext4, good suggestion.
You say you tried XFS. but show an error message, but again you fail to tell what you did. The error message is apparently about a statement to attach a label to an XFS file system, but that is all we can guess.

The result of all this is that we have very little factual information. Let us try to gather that first. You show, as root and of course with the device connected (if it is removable):

fdisk -l

lsblk -f

Then you explain what you try to use to partition the device (if at all) and what to create that XFS file system on it.

As an afterburner, please stop talking about “changeing the owner of the device”. You can not change the owner (except when you sell it).

15

Thank you! Yes this is getting difficult.

Disk /dev/nvme1n1: 931.5 GiB, 1000204886016 bytes, 1953525168 sectors
Disk model: KINGSTON SKC2000M81000G                 
Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disklabel type: gpt
Disk identifier: 10CC1146-28FE-4E58-904A-647ED5CACB1F

Device         Start        End    Sectors   Size Type
/dev/nvme1n1p1  2048 1953525134 1953523087 931.5G Linux filesystem


Disk /dev/nvme0n1: 931.5 GiB, 1000204886016 bytes, 1953525168 sectors
Disk model: KINGSTON SA2000M81000G                  
Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disklabel type: dos
Disk identifier: 0xc9797a4b

Device         Boot      Start        End    Sectors   Size Id Type
/dev/nvme0n1p1            2048 1871599615 1871597568 892.5G 83 Linux
/dev/nvme0n1p2      1871599616 1872623615    1024000   500M ef EFI (FAT-12/16/32)
/dev/nvme0n1p3      1872623616 1949329407   76705792  36.6G 83 Linux
/dev/nvme0n1p4      1949329408 1953525167    4195760     2G 82 Linux swap / Solaris


Disk /dev/nvme1n1: 931.5 GiB, 1000204886016 bytes, 1953525168 sectors
Disk model: KINGSTON SKC2000M81000G                 
Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disklabel type: gpt
Disk identifier: 10CC1146-28FE-4E58-904A-647ED5CACB1F

Device         Start        End    Sectors   Size Type
/dev/nvme1n1p1  2048 1953525134 1953523087 931.5G Linux filesystem


Disk /dev/nvme0n1: 931.5 GiB, 1000204886016 bytes, 1953525168 sectors
Disk model: KINGSTON SA2000M81000G                  
Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disklabel type: dos
Disk identifier: 0xc9797a4b

Device         Boot      Start        End    Sectors   Size Id Type
/dev/nvme0n1p1            2048 1871599615 1871597568 892.5G 83 Linux
/dev/nvme0n1p2      1871599616 1872623615    1024000   500M ef EFI (FAT-12/16/32)
/dev/nvme0n1p3      1872623616 1949329407   76705792  36.6G 83 Linux
/dev/nvme0n1p4      1949329408 1953525167    4195760     2G 82 Linux swap / Solaris

I use Partitioner (out of YaST) to handle my drives.

To be fair I’m pretty sure that malarkey came out of us both to a degree.

16

Thank you! Yes this is getting difficult.

Disk /dev/nvme1n1: 931.5 GiB, 1000204886016 bytes, 1953525168 sectors
Disk model: KINGSTON SKC2000M81000G                 
Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disklabel type: gpt
Disk identifier: 10CC1146-28FE-4E58-904A-647ED5CACB1F

Device         Start        End    Sectors   Size Type
/dev/nvme1n1p1  2048 1953525134 1953523087 931.5G Linux filesystem


Disk /dev/nvme0n1: 931.5 GiB, 1000204886016 bytes, 1953525168 sectors
Disk model: KINGSTON SA2000M81000G                  
Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disklabel type: dos
Disk identifier: 0xc9797a4b

Device         Boot      Start        End    Sectors   Size Id Type
/dev/nvme0n1p1            2048 1871599615 1871597568 892.5G 83 Linux
/dev/nvme0n1p2      1871599616 1872623615    1024000   500M ef EFI (FAT-12/16/32)
/dev/nvme0n1p3      1872623616 1949329407   76705792  36.6G 83 Linux
/dev/nvme0n1p4      1949329408 1953525167    4195760     2G 82 Linux swap / Solaris


NAME        FSTYPE LABEL UUID                                 FSAVAIL FSUSE% MOUNTPOINT
nvme1n1                                                                      
└─nvme1n1p1 xfs          bf1e64bb-2026-4ee4-8b15-8785931eae27  930.1G     0% /run/media/
nvme0n1                                                                      
├─nvme0n1p1 btrfs        6540b7dd-d474-43f6-aadc-fc1bf307fc10  837.5G     6% /run/media/
├─nvme0n1p2 vfat         FBBD-96D6                               492M     2% /boot/efi
├─nvme0n1p3 btrfs        950591a9-d8c5-403a-9996-7a56074237b3    8.2G    77% /
└─nvme0n1p4 swap         acdaf638-7b60-40e7-9e9a-182b94e0b3ef                [SWAP]
50-3E-AA-E3-41-E7:~ #

I use Partitioner (out of YaST) to handle my drives.

To be fair seems that malarkey (ownership) came out of both ends here.

17

One important thing about posting CODE. Please, please, please include the command. Now we have only output, but can not check what command you used. Please one more line at the top end the bottom. Example:

henk@boven:~> uname -r
5.3.18-lp152.63-default
henk@boven:~>

Ok, you used YaST > System > Partitioner. Did you tell it to use a mount point? When not, why not? When yes, why /run/media?
And is it this Partitioner that gave you the error message?

Now we are going to check how it is mounted (because it apparently is already):

mount | grep nvme1n1
18

Thank you! This is exactly the information I want!

I chose /run/media because everything else was apparently “shadowed” error message follows:
"ERROR

The Btrfs subvolume mounted at /home is shadowed."

This goes for every standard mounting point I try

50-3E-AA-E3-41-E7:~ # mount | grep nvme1n1
/dev/nvme1n1p1 on /run/media/hoff/55e26897-c1d6-42cd-8874-cf06a82475d0 type xfs (rw,relatime,attr2,inode64,logbufs=8,logbsize=32k,noquota)
50-3E-AA-E3-41-E7:~ #

I hope I did it right.

19

I forgot. Yes it is this Partitioner that gives me the ERRORs

20

Try to reformat it with simple FS, such as FAT32 or exFat.

When SSD became read-only by itself it means that SSD’s firmware decided that it is severely broken, and user can save data from SSD with r/o, without possibility to break it with writes.
Check SMART indicators.
Possibly you can change it with your reseller for new one if you claim a warranty case.