SSD Health on Linux – How Often Do You Check Yours?

[Unison]

SSD Health on Linux – How Often Do You Check Yours?​

I've been using Linux more recently and it got me thinking about something that's easy to forget about until a drive starts causing problems — SSD health.


Linux can read a surprising amount of information about a drive using SMART.


For example, you can identify your drives with:



lsblk


and, with smartmontools installed, check a drive using:




sudo smartctl -a /dev/sda


For an NVMe SSD it might instead be:




sudo smartctl -a /dev/nvme0


This can show things such as drive temperature, power-on hours, errors, percentage used and other health information.


Obviously SMART can't guarantee that a drive won't suddenly fail, so it's no replacement for having a backup.


But I'm curious how many people actually keep an eye on it.


Do you periodically check the health of your SSDs and hard drives, or only investigate when something starts behaving strangely?


Also, has SMART ever warned you about a failing drive early enough to save your data?


Be interested to hear what tools everyone uses on Linux, Windows or macOS for keeping an eye on drive health.

 

[CyborJaX]

I will have to admit I don't do checks very often, when I first loaded apps I tried the GSmartControl but it didn't work on all my drives.
The only other is sudo smartctl -a /dev/nvme0 that I have used once.

So my daily main apps is Timeshift which is a no brainer for anybody using Linux, this is saved to a separate drive.
Plus also the backup script in Startup Application which copies the home DIR to a separate Drive.
ie:- rsync -a -delete --exclude=".*" ~/ /media/cyborjax/BackUp/HBackup

But for my peace of mind I have a full clone of my main PC's

 

[Unison]

I will have to admit I don't do checks very often, when I first loaded apps I tried the GSmartControl but it didn't work on all my drives.
The only other is sudo smartctl -a /dev/nvme0 that I have used once.

So my daily main apps is Timeshift which is a no brainer for anybody using Linux, this is saved to a separate drive.
Plus also the backup script in Startup Application which copies the home DIR to a separate Drive.
ie:- rsync -a -delete --exclude=".*" ~/ /media/cyborjax/BackUp/HBackup

But for my peace of mind I have a full clone of my main PC's

yes that confirms it for me
To see the SMART information for a drive:

sudo smartctl -a /dev/sdc

Obviously, change /dev/sdc to match your own drive.

You can also run a short SMART self-test:

sudo smartctl -t short /dev/sdc
Then, after waiting for the test to finish, check the result with:

sudo smartctl -l selftest /dev/sdc
On my Manjaro system mine came back with:

Short offline Completed without error
So that's a nice quick way of checking for obvious drive problems.

For backups, Timeshift is pretty much a no-brainer for me on Linux. It's useful for taking system snapshots so you have something to fall back on if an update or configuration change causes problems.

For the Home directory, rsync is another useful option. For example:

rsync -a --delete ~/ /path/to/backup/HBackup/
This mirrors the Home directory to the backup location.

One thing worth mentioning is --delete: if a file has been removed from the source, rsync will also remove it from the destination. So it's important to make absolutely sure the destination path is correct before running it.

You can exclude files and directories as well, although excluding all hidden files with:

--exclude=".*"
also excludes things such as .config and .local, which contain a lot of your Linux application settings.

Between Timeshift for the system, rsync for Home, and an occasional SMART check of the drive, you've got three simple tools that can save a lot of trouble later.