Strane UUID problem

[Chris Murphy]

On Fri, Mar 20, 2015 at 11:59 PM, Heinz Diehl <htd+ml at fritha.org> wrote:
> On 20.03.2015, Chris Murphy wrote:
>
>> If /dev/sda uses MBR, it doesn't really have a UUID, it might have a
>> serial number.
>
> It's MBR, and it didn't have a UUID before.
>
>> I think think needs more troubleshooting, rather
>> than bringing out the hammer before the problem is identified.
>> Feel free to post the results from fstab, and blkid.
>
> I'm nearly done with debugging, and are 99% sure it's a systemd problem.
> What happens is that when /home is nilfs2 formatted, systemd sees the identical
> UUID assigned to /dev/sda, and the systemd service fails to mount /home because
> it tries to mount /dev/sda rather than the correct partition /dev/sdaX.
>
> Reformatting with either ext4 or xfs solves the problem immediately.
> Btw: lsblk -f shows the faulty UUID assigned to /dev/sda, while blkid does not.

I don't think systemd has any concept of filesystems (volume formats).
It gets all of this from libblkid, udev, and the kernel. So I'd say
one of those three things is confused, and then confuses everything
else.

The fact you get a difference in output between lsblk and blkid is
itself a bug. I'd file it against libblkid and then include the bug
URL here.


-- 
Chris Murphy