I've recently moved my linux system from hdd to ssd. I managed to get it booting and everything was fine until I deleted old boot partitions from hdd. Now when I try to boot I end up in dracut shell with /dev/disk/by-uuid/xxxxx doesn't exist error, where xxxxx is UUID of my former /boot or /boot/efi partition (I'm not exactly sure which one is it since I no longer have them).

Now I need to regenerate initramfs to use the new partitions I cloned from hdd to ssd. I tried running dracut --regenerate-all --force from rescue boot but it didn't help.

I'm using Fedora 21, so how can I do it? Or is there problem in something else?

5 Answers 5


Do you still have a F21 install media?

If so do the following:

boot to "troubleshooting" > Rescue a Fedora System

Allow for mounting (normally not r/o)

chroot /mnt/sysimage 

dracut --regenerate-all -f && grub2-mkconfig -o /boot/grub2/grub.cfg

If you are sure it is not throwing errors on the dracut --regenerate-all previous made (i.e. ls /boot shows them but won't boot) just rebuild grub with the second command.

~ Sent from a Fedora 22 box

  • That's basically what I did (although I ran dracut --regenerate-all -f from the rescue boot option that was installed), however it didn't help. I reinstalled now anyway as I realized that it's just not possible to ensure 100% compatibility when virtually anything could be relying on properties of the hdd (eg, UUIDs, partition names, partion sizes, ...) that's just not possible to reliably handle without deeper understaning.
    – user437209
    Apr 17, 2015 at 21:43

try the following link instructions:


scan for logical volumes first:

vgscan vgchange -ay

Find swap partitions:

blkid|grep swap

and an example:

/dev/mapper/fedora-swap: UUID="28d71a2c-1b34-4115-aa19-083373ec4d8a" TYPE="swap"

Edit /etc/fstab to use the swap partition, here's an example: nano /etc/fstab

/dev/mapper/fedora-swap swap swap defaults 0 0

To regenerate all initramfs, do:

dracut --regenerate-all --force


I had fixed everything in fstab and grub entries and all, but still I was getting into maintenance mode at boot time with dracut time out errors:

dracut --regenerate-all -f && grub2-mkconfig -o /boot/grub2/grub.cfg

This fixed my problem. (CentOS7 Dedicated Server)


After removing swap partition from my CentOS7 machine, I got dracut time out errors. Regenerating initramfs and grub2 didn't solve. In order to solve the issue I had to manually modify the grub2 configuration file:

## Look for swap in grub configuration file 

cat /boot/grub2/grub.cfg | grep swap

## Delete old swap
remove rd.lvm.lv=<lvm_name>/swap string from grub

Pretty much every time this happens, I boot to previous kernel and in Gnome's terminal remove and install kernel and related packages again. Note that the version string will be different, thus just look on currently installed kernel version via your distors preffered method. Below is example for Fedora 37 when it happened last time for me.

$ sudo dnf remove *6.2.9-200*
$ sudo dnf upgrade kernel --refresh

Few notes:

  • It is not mandatory, but just to be sure the metadata and package managers cache is fresh, the " --refresh" switch is added. Feel free to remove it if conditions below does not apply.
  • If you are fully sure and confident that you know that there is no newer kernel suddenly pushed or you are not using unstable/testing repo, add " -y" in the end of second command. It will accept changes for you; Reboot and you should be good.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .