• [gentoo-user] biosdevname or equivalent for disks: did the order change

    From Larry the Plumber@21:1/5 to All on Wed Sep 11 15:20:01 2024
    Hi!
    A bit of history first: my Fedora, then (after reinstall) my openSUSE
    then (after reinstall) my Gentoo failed during boot. I thought it was
    something related to dracut so I've been adding `hostonly = false` to dracut.conf on all my home machines.

    In summary, yesterday it failed on my Gentoo so I had to investigate.
    * I had enable `unstable` use flags for the kernel only.
    * I hadn't double-check the boot entries created.
    * Power went off (no UPS yet) so I booted the new kernel.
    * It didn't boot. I fixed two things: changed the partition id of
    LUKSed swap from swap id to LVM id, to prevent systemd attempting to
    mount it at boot (and failing). Secondly, I checked the fstab and
    noticed all dev nodes were wrong (I have two SSD).
    * After fixing the dev nodes in fstab the system booted properly.

    This probably explains the Gentoo failure, since other distros (and
    me, starting some day) use UUID in fstab.

    So, my question is:
    Has anyone any input on mysterious failures across several distros,
    including Gentoo, around kernel 6.9.x <> 6.10.x?

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)