I'm running Debian 12 (Bookworm) and trying to set up a GPT partition table on an old SSD that previously had an MBR with one partition from its time in a Windows machine. I've been using `parted` to do this, and it seems to go well, warning me that the current data will be unrecoverable. However, after I'm done and exit, the drive still shows the old MBR table and partition. Running `partprobe` right after doesn't help either. I've checked the drive with SMART tests, and everything passes. I can even mount and access the NTFS partition without problems, but I'm at a loss as to why this issue is happening. Any ideas?
3 Answers
It sounds like maybe your SSD might be stuck in some kind of 'read-only' state. I found that when I unmounted the drive after a change, the change still stayed as long as I didn’t remount it. But when I did, the changes disappeared. Also, it gave me a warning about the NTFS being unclean. So annoying! If it's really giving you that much trouble, maybe using a different drive altogether could save you some frustration.
Have you tried using GParted instead? It's a graphical way to manage partitions, which you might find easier. You can install it via `sudo apt install gparted`. Just a suggestion!
I don't have X11 set up for a GUI, so I can't really use GParted. Plus, I think it’s just the graphical version of `parted`, which I've been using.
Does the old partition table stay the same even after you reboot the system? Sometimes that can happen if there's a cache issue, so restarting might reset things. It's just a thought!
Yeah, I tried that. The old partition is still there. It works fine with my other drives, so I’m really confused.
You might want to run `wipefs --all /dev/sda` (make sure to replace sda with your actual device!) to clear out the old partition table. Then try creating a new GPT with gparted and see if that does the trick. If it still doesn't work, it might be a sign that your SSD is failing and has switched to a permanent read-only state.