Disk "SMART Data" no more accessible since installation of kernel 6.10.3

By “next kernel update” would that be 6.10.4-200 ?
That can be had from testing with:

sudo dnf upgrade --enablerepo=updates-testing kernel\*

I’m not sure that’s fixing it because I have 6.10.4-200 going right now and see this in coredumpctl list
Wed 2024-08-14 14:50:37 EDT 2301 0 0 SIGABRT inaccessible /usr/sbin/smartctl -
but that might be something else.

UPDATE - I’ve just installed gnome-disk-utility and ran ‘gnome-disks’ and my “Smart Data & Self-Tests…” menu item is greyed out, so 6.10.4-200 does not seem to fix that

It’s targeted for 6.11.

Okay, I’d written a different bug report on smart here:

https://bugzilla.redhat.com/show_bug.cgi?id=2277537

that’s still “NEW” mode. 2277537 is just about core dumping at boot time.

The interesting thing about this bug is that the kernel change addressed a pre-existing formatting bug in libata. But all the tools that read data from libata like smartmon are used to getting the old ‘malformed’ data and have been working with that for years, so they are now tripping over the fixed version of the data.

So now the kernel developers have decided to just reintroduce the old bug, because they expect that otherwise many of the tools reading the data will not get fixed anytime soon, and that would lead to a lot of systems with non working hdd status tools. It’s quite an interesting situation. :wink:

The kernel has a “bug compatibility” rule: if a userspace program depends on a bug, then the bug is effectively API and cannot be fixed.

I would argue that this is not a good rule, but it is how it is.

1 Like

FYI 2303813 – With 6.10 kernels before 6.10.5 and 6.11 kernels before rc4, gnome-disks is partly nonfunctional

“SMART data” back after yesterday’s last updates:

Current kernel version is now : 6.10.5-200.fc40.x86_64

1 Like