ssd drive stops working after a while

[Yellow Pasque]

Unsafe Shutdowns:                   645

Did you turn off Fast Startup/Boot in Windows?

[ricardodevpt]

Looks like no smart tests have been performed. Run either a short or long test to capture some data, then re-run the reports.

Num   ErrCount  SQId   CmdId  Status  PELoc          LBA  NSID    VS
  0       2954     0  0x100a  0x4004      -            0     0     -

is a bit concerning, but since the brand and model of the SSD are being hidden, we can't look up what that means, if anything.

BTW. Windows is well-known for ignoring hardware errors. I've never seen a reason why MSFT has been doing this, but it is common for every other OS to complain about hardware issues that Windows ignores, until the blue screen. Other OSes log the errors much earlier. Is there anything in the system logs? I'd look for errors and warnings over the last few weeks.

That is in drive A, which is not being mounted in ubuntu. I'll take a look at the system logs tomorrow and see if something stands out.
The drive A is SAMSUNG MZVLB512HBJQ-000L2
Drive B is Silicon Motion based SSDs - model TS512GMTS430S
Thanks

[ricardodevpt]

Unsafe Shutdowns:                   645

Did you turn off Fast Startup/Boot in Windows?

No, it's still enabled.
Should I disable it?

[TheFu]

No, it's still enabled.
Should I disable it?

YES! If any file system is shared/accessed by both OSes. Disabling both of those is SOP - standard operating procedure and has been since MSFT decided that closing file systems was an unnecessary luxury, incorrectly.

Odd branded SSDs tend to wear out much faster than the reputable brands too. I had a few no-name SSDs that died after about 2 yrs of use. I was lucky, in that
a) I have excellent backups
b) before total failure, the file systems would drop into read-only mode first, so I had a few weeks of notification that the SSDs were failing. I understand that is atypical. They do fail, without any warning too.

[ricardodevpt]

YES! If any file system is shared/accessed by both OSes. Disabling both of those is SOP - standard operating procedure and has been since MSFT decided that closing file systems was an unnecessary luxury, incorrectly.

Odd branded SSDs tend to wear out much faster than the reputable brands too. I had a few no-name SSDs that died after about 2 yrs of use. I was lucky, in that
a) I have excellent backups
b) before total failure, the file systems would drop into read-only mode first, so I had a few weeks of notification that the SSDs were failing. I understand that is atypical. They do fail, without any warning too.

Ok thanks. I did that

[ricardodevpt]

I've resized my windows partition on drive A and created a partition to install ubuntu (it was previously installed in drive B and failing).

Now, with ubuntu in drive A, everything works well (I kept drive B unmounted).

I guess that the most likely cause of failure is some bug on the drivers for the ssd controller for this particular ssd disk. Does anyone know where can I file a bug for this? I guess it should be filed in the kernel repo?

Thank you all for the suggestions.

[TheFu]

I threw the model of SSD into google and if found this: https://bbs.archlinux.org/viewtopic.php?id=251762
In that thread, it shows some troubleshooting things and a few commands to help determine what the problem might be. The "vmd" module helped 2 people, so I'd give that a try.

[ricardodevpt]

I threw the model of SSD into google and if found this: https://bbs.archlinux.org/viewtopic.php?id=251762
In that thread, it shows some troubleshooting things and a few commands to help determine what the problem might be. The "vmd" module helped 2 people, so I'd give that a try.

ok, thanks. I'll take a look