Bonjour,

Je constate que depuis que je suis passé à Fedora 35, mon arrivée jusqu'à la session utilisateur est extrêmement longue : plus de 3 minutes dont 1min 6s en initrd et plus de 2min 25s dans l'userspace comme le montre la commande :
$ systemd-analyze time
Startup finished in 3.112s (kernel) + 1min 6.629s (initrd) + 2min 25.210s (userspace) = 3min 34.952s 
graphical.target reached after 2min 25.204s in userspace
Ces temps se répartissent ainsi :
$ systemd-analyze blame 
  2min 10ms systemd-udev-settle.service
1min 5.996s dracut-initqueue.service
    42.630s lvm2-pvscan@9:127.service
    18.402s NetworkManager-wait-online.service
    12.389s smartd.service
    11.983s udisks2.service
    10.581s lvm2-monitor.service
     4.802s plymouth-quit-wait.service
     1.563s akmods.service
      809ms packagekit.service
      677ms initrd-switch-root.service
      597ms colord.service
      510ms ModemManager.service
      466ms power-profiles-daemon.service
      464ms polkit.service
      461ms accounts-daemon.service
      449ms lvm2-pvscan@8:66.service
      400ms upower.service
      382ms avahi-daemon.service
      379ms firewalld.service
      373ms rtkit-daemon.service
      370ms switcheroo-control.service
      368ms systemd-homed.service
      368ms systemd-logind.service
      366ms systemd-machined.service
      365ms thermald.service
      364ms abrtd.service
      343ms systemd-resolved.service
      310ms sssd-kcm.service
      277ms dbus-broker.service
      273ms systemd-journal-flush.service
      203ms mnt-RED\x2dMMEDIA.mount
      198ms mnt-RED\x2dDATA.mount
      151ms rsyslog.service
      150ms systemd-udevd.service
      134ms user@1000.service
      133ms NetworkManager.service
      127ms initrd-parse-etc.service
      112ms nvidia-powerd.service
       97ms chronyd.service
       93ms gssproxy.service
       85ms cups.service
       71ms auditd.service
       62ms systemd-oomd.service
       60ms systemd-vconsole-setup.service
       56ms systemd-udev-trigger.service
       52ms systemd-tmpfiles-setup-dev.service
       44ms dev-zram0.swap
       44ms virtqemud.service
       42ms systemd-tmpfiles-setup.service
       42ms dracut-cmdline.service
       37ms var-lib-nfs-rpc_pipefs.mount
       37ms systemd-zram-setup@zram0.service
       36ms lvm2-pvscan@8:2.service
       30ms boot.mount
       25ms plymouth-switch-root.service
       25ms systemd-userdbd.service
       24ms systemd-fsck-root.service
       24ms dracut-pre-pivot.service
       23ms systemd-fsck@dev-disk-by\x2duuid-d9c71f34\x2d3973\x2d486e\x2d895e\x2d062fd7eb1bcd.service
       22ms modprobe@drm.service
       19ms systemd-random-seed.service
       17ms livesys.service
       15ms dev-mapper-fedora_localhost\x2d\x2dlive\x2dswap.swap
       15ms dev-hugepages.mount
       15ms import-state.service
       15ms dev-mqueue.mount
       15ms systemd-journald.service
       14ms sys-kernel-debug.mount
       13ms sys-kernel-tracing.mount
       13ms plymouth-read-write.service
       13ms kmod-static-nodes.service
       12ms gdm.service
       12ms uresourced.service
       12ms plymouth-start.service
       12ms rpc-statd-notify.service
       12ms systemd-modules-load.service
       11ms dracut-pre-udev.service
       11ms livesys-late.service
       11ms nfs-convert.service
       11ms initrd-cleanup.service
        9ms wpa_supplicant.service
        9ms systemd-sysctl.service
        8ms systemd-update-utmp.service
        6ms systemd-remount-fs.service
        5ms user-runtime-dir@1000.service
        4ms systemd-backlight@leds:g15::kbd_backlight.service
        4ms systemd-user-sessions.service
        4ms systemd-update-utmp-runlevel.service
        4ms sys-fs-fuse-connections.mount
        4ms initrd-udevadm-cleanup-db.service
        3ms sys-kernel-config.mount
        2ms dracut-shutdown.service
        1ms tmp.mount
        1ms modprobe@configfs.service
        1ms modprobe@fuse.service
       22us systemd-homed-activate.service
Voici les logs du dernier boot
$ journalctl --since "2022-04-18 19:27:00" -p 0..3
Apr 18 19:27:02 fedora kernel: x86/cpu: SGX disabled by BIOS.
Apr 18 19:27:02 fedora kernel: ACPI BIOS Error (bug): Could not resolve symbol [\_SB.PR00._CPC], AE_NOT_FOUND (20210930/psargs-330)
Apr 18 19:27:02 fedora kernel: ACPI Error: Aborting method \_SB.PR01._CPC due to previous error (AE_NOT_FOUND) (20210930/psparse-529)
Apr 18 19:27:02 fedora kernel: ACPI BIOS Error (bug): Could not resolve symbol [\_SB.PR00._CPC], AE_NOT_FOUND (20210930/psargs-330)
Apr 18 19:27:02 fedora kernel: ACPI Error: Aborting method \_SB.PR02._CPC due to previous error (AE_NOT_FOUND) (20210930/psparse-529)
Apr 18 19:27:02 fedora kernel: ACPI BIOS Error (bug): Could not resolve symbol [\_SB.PR00._CPC], AE_NOT_FOUND (20210930/psargs-330)
Apr 18 19:27:02 fedora kernel: ACPI Error: Aborting method \_SB.PR03._CPC due to previous error (AE_NOT_FOUND) (20210930/psparse-529)
Apr 18 19:27:02 fedora kernel: ACPI BIOS Error (bug): Could not resolve symbol [\_SB.PR00._CPC], AE_NOT_FOUND (20210930/psargs-330)
Apr 18 19:27:02 fedora kernel: ACPI Error: Aborting method \_SB.PR04._CPC due to previous error (AE_NOT_FOUND) (20210930/psparse-529)
Apr 18 19:27:02 fedora kernel: ACPI BIOS Error (bug): Could not resolve symbol [\_SB.PR00._CPC], AE_NOT_FOUND (20210930/psargs-330)
Apr 18 19:27:02 fedora kernel: ACPI Error: Aborting method \_SB.PR05._CPC due to previous error (AE_NOT_FOUND) (20210930/psparse-529)
Apr 18 19:27:02 fedora kernel: ACPI BIOS Error (bug): Could not resolve symbol [\_SB.PR00._CPC], AE_NOT_FOUND (20210930/psargs-330)
Apr 18 19:27:02 fedora kernel: ACPI Error: Aborting method \_SB.PR06._CPC due to previous error (AE_NOT_FOUND) (20210930/psparse-529)
Apr 18 19:27:02 fedora kernel: ACPI BIOS Error (bug): Could not resolve symbol [\_SB.PR00._CPC], AE_NOT_FOUND (20210930/psargs-330)
Apr 18 19:27:02 fedora kernel: ACPI Error: Aborting method \_SB.PR07._CPC due to previous error (AE_NOT_FOUND) (20210930/psparse-529)
Apr 18 19:28:05 fedora kernel: ata7.00: exception Emask 0x100 SAct 0x800 SErr 0x0 action 0x6 frozen
Apr 18 19:28:05 fedora kernel: ata7.00: failed command: READ FPDMA QUEUED
Apr 18 19:28:05 fedora kernel: ata7.00: cmd 60/08:58:00:0a:00/00:00:00:00:00/40 tag 11 ncq dma 4096 in
                                        res 40/00:00:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout)
Apr 18 19:28:05 fedora kernel: ata7.00: status: { DRDY }
Apr 18 19:28:08 fedora kernel: I/O error, dev sdf, sector 2560 op 0x0:(READ) flags 0x80700 phys_seg 1 prio class 0
Apr 18 19:28:10 fedora kernel: 
Apr 18 19:29:12 fedora kernel: ata7.00: exception Emask 0x100 SAct 0x200000 SErr 0x0 action 0x6 frozen
Apr 18 19:29:12 fedora kernel: ata7.00: failed command: READ FPDMA QUEUED
Apr 18 19:29:12 fedora kernel: ata7.00: cmd 60/08:a8:00:10:00/00:00:00:00:00/40 tag 21 ncq dma 4096 in
                                        res 40/00:00:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout)
Apr 18 19:29:12 fedora kernel: ata7.00: status: { DRDY }
Apr 18 19:29:15 fedora kernel: I/O error, dev sdf, sector 4096 op 0x0:(READ) flags 0x80700 phys_seg 1 prio class 0
Apr 18 19:29:45 fedora kernel: ata7.00: exception Emask 0x100 SAct 0x0 SErr 0x0 action 0x6 frozen
Apr 18 19:29:45 fedora kernel: ata7.00: failed command: READ DMA
Apr 18 19:29:45 fedora kernel: ata7.00: cmd c8/00:02:00:10:00/00:00:00:00:00/e0 tag 7 dma 1024 in
                                        res 40/00:00:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout)
Apr 18 19:29:45 fedora kernel: ata7.00: status: { DRDY }
Apr 18 19:30:09 fedora systemd[1]: Failed to start Wait for udev To Complete Device Initialization.
Apr 18 19:30:09 fedora /usr/sbin/irqbalance[978]: libcap-ng used by "/usr/sbin/irqbalance" failed dropping bounding set due to not having CAP_SETPCAP in capng_apply
Apr 18 19:30:10 fedora /usr/bin/nvidia-powerd[984]: No matching GPU found
Apr 18 19:30:10 fedora /usr/bin/nvidia-powerd[984]: Failed to initialize RM Client
Apr 18 19:30:10 fedora systemd[1]: Failed to start nvidia-powerd service.
Apr 18 19:30:18 fedora kernel: ata7.00: exception Emask 0x100 SAct 0x0 SErr 0x0 action 0x6 frozen
Apr 18 19:30:18 fedora kernel: ata7.00: failed command: READ DMA
Apr 18 19:30:18 fedora kernel: ata7.00: cmd c8/00:02:02:10:00/00:00:00:00:00/e0 tag 22 dma 1024 in
                                        res 40/00:00:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout)
Apr 18 19:30:18 fedora kernel: ata7.00: status: { DRDY }
Apr 18 19:30:22 fedora smartd[994]: Device: /dev/sdd [SAT], 616 Currently unreadable (pending) sectors
Apr 18 19:30:22 fedora smartd[994]: Device: /dev/sde [SAT], 51 Currently unreadable (pending) sectors
Apr 18 19:30:22 fedora smartd[994]: Device: /dev/sde [SAT], 19 Offline uncorrectable sectors
Apr 18 19:30:31 fedora dbus-broker-launch[1415]: Ignoring duplicate name 'org.freedesktop.FileManager1' in service file '/usr/share//dbus-1/services/org.kde.dolphin.FileManager1.service'
Apr 18 19:30:31 fedora gnome-session-binary[1420]: GLib-GIO-CRITICAL: g_bus_get_sync: assertion 'error == NULL || *error == NULL' failed
Apr 18 19:30:31 fedora gnome-session-binary[1420]: GLib-GIO-CRITICAL: g_bus_get_sync: assertion 'error == NULL || *error == NULL' failed
Apr 18 19:30:32 00-D8-61-52-75-8C pipewire-pulse[1538]: pw.conf: execvp error 'pactl': No such file or directory
Apr 18 19:30:41 00-D8-61-52-75-8C gdm-password][1851]: gkr-pam: unable to locate daemon control file
Apr 18 19:30:43 00-D8-61-52-75-8C dbus-broker-launch[1969]: Ignoring duplicate name 'org.freedesktop.FileManager1' in service file '/usr/share//dbus-1/services/org.kde.dolphin.FileManager1.service'
Apr 18 19:30:50 00-D8-61-52-75-8C pipewire-pulse[2878]: pw.conf: execvp error 'pactl': No such file or directory
Apr 18 19:30:54 00-D8-61-52-75-8C gdm-launch-environment][1345]: GLib-GObject: g_object_unref: assertion 'G_IS_OBJECT (object)' failed
Apr 18 19:35:26 00-D8-61-52-75-8C kernel: EXT4-fs error (device dm-3): ext4_lookup:1785: inode #30146561: comm kioslave5: iget: checksum invalid
Apr 18 20:00:23 00-D8-61-52-75-8C smartd[994]: Device: /dev/sdd [SAT], 616 Currently unreadable (pending) sectors
Apr 18 20:00:23 00-D8-61-52-75-8C smartd[994]: Device: /dev/sde [SAT], 51 Currently unreadable (pending) sectors
Apr 18 20:00:23 00-D8-61-52-75-8C smartd[994]: Device: /dev/sde [SAT], 19 Offline uncorrectable sectors
J'ai une grappe de 3 disques montés en RAID 5 (/dev/sdb, sdc, et sdd). Le mapper se présente ainsi :
$ ls -l /dev/mapper/
total 0
crw-------. 1 root root 10, 236 Apr 18 19:28 control
lrwxrwxrwx. 1 root root       7 Apr 18 19:28 fedora_localhost--live-root -> ../dm-0
lrwxrwxrwx. 1 root root       7 Apr 18 19:28 fedora_localhost--live-swap -> ../dm-1
lrwxrwxrwx. 1 root root       7 Apr 18 19:28 fedora_tron-root -> ../dm-2
lrwxrwxrwx. 1 root root       7 Apr 18 19:28 raiddata-mmedia -> ../dm-4
lrwxrwxrwx. 1 root root       7 Apr 18 19:28 raiddata-reddata -> ../dm-3
Est-ce que je dois comprendre que la ligne "kernel: EXT4-fs error (device dm-3): ext4_lookup:1785: inode #30146561: comm kioslave5: iget: checksum invalid" m'indique un problème avec mon RAID ?

Comment puis-je réduire le temps nécessaire au démarrage jusqu'à ma session ?
On peut remarquer beaucoup d'erreurs sur tes disques. Il va falloir songer à s'assurer d'avoir des sauvegardes des choses importantes.

Pour rappel un RAID n'est pas une sauvegarde.
Oui, effectivement, je n'utilise pas le RAID pour garantir mes données. J'ai sauvegardé tout ce qui est important, et même un peu plus, sur un disque externe.
Ca me surprend de voir qu'il y a déjà pas mal d'erreurs en si peu de temps d'utilisation. J'ai dû monter mon RAID il y a 2 ans je crois et l'usage de mon système n'est quand même pas intensif.
Les secteurs illisibles semblent concerner les disques /dev/sdd et /dev/sde. Seul le sdd fait partie du RAID. Je n'ai donc que celui-ci à remplacer pour faire perdurer mon RAID ? (après reconstruction du disque bien sûr).
Que retourne la commande du paquet smartmontools :
sudo smartctl -aA /dev/sdX
En remplaçant X par la lettre de ton disque?

Ta tenté de remplacer tes nappes SATA pourvoir si cela ne venez pas de là ? La qualité de celle ci n'est pas toujours garantie.
En effet, tu peux remplacer uniquement celui défaillant après avoir contrôlé le tout.

Comme le précise VINDICATORs, à voir en remplaçant les câbles SATA aussi car ça peut venir de là.