My Proxmox host boots to BIOS after upgrading from PVE 8 to 9.

I was able to use Rescue Boot from a Proxmox 9 install USB but it says /boot not found. In Rescue Boot, everything else seems to be working fine though.

I've only been using Proxmox for about 2 months now so I'm not familiar with all the troubleshooting options, but I've looked at other forums posts and tried some of the solutions posted but none have worked for me so far.

Here is my output for lsblk and there is no /boot or /boot/efi.

root@pve:~# lsblk
NAME                          MAJ:MIN RM   SIZE RO TYPE MOUNTPOINTS
sda                             8:0    0  14.6T  0 disk
└─sda1                          8:1    0  14.6T  0 part
sdb                             8:16   0  14.6T  0 disk
└─sdb1                          8:17   0  14.6T  0 part
sdc                             8:32   0   3.6T  0 disk
└─sdc1                          8:33   0   3.6T  0 part /mnt/backupdrive
sdd                             8:48   0 238.5G  0 disk
├─sdd1                          8:49   0  1007K  0 part
├─sdd2                          8:50   0     1G  0 part
└─sdd3                          8:51   0 237.5G  0 part
  ├─pve-swap                  252:0    0     8G  0 lvm  [SWAP]
  ├─pve-root                  252:1    0  69.4G  0 lvm  /
  ├─pve-data_tmeta            252:2    0   1.4G  0 lvm 
  │ └─pve-data-tpool          252:4    0 141.2G  0 lvm 
  │   ├─pve-data              252:5    0 141.2G  1 lvm 
  │   ├─pve-vm--1002--disk--0 252:6    0    32G  0 lvm 
  │   ├─pve-vm--1001--disk--0 252:7    0    64G  0 lvm 
  │   └─pve-vm--1000--disk--0 252:8    0    32G  0 lvm 
  └─pve-data_tdata            252:3    0 141.2G  0 lvm 
    └─pve-data-tpool          252:4    0 141.2G  0 lvm 
      ├─pve-data              252:5    0 141.2G  1 lvm 
      ├─pve-vm--1002--disk--0 252:6    0    32G  0 lvm 
      ├─pve-vm--1001--disk--0 252:7    0    64G  0 lvm 
      └─pve-vm--1000--disk--0 252:8    0    32G  0 lvm 
sde                             8:64   0  14.6T  0 disk
└─sde1                          8:65   0  14.6T  0 part
sdf                             8:80   0  14.6T  0 disk
└─sdf1                          8:81   0  14.6T  0 part
sdg                             8:96   1  59.8G  0 disk
├─sdg1                          8:97   1   282K  0 part
├─sdg2                          8:98   1     8M  0 part
├─sdg3                          8:99   1   1.5G  0 part
└─sdg4                          8:100  1   300K  0 part

I ran fdisk -l and there is an EFI System that is 8M but apparently it is supposed to be around 1G?

Device       Start     End Sectors  Size Type
/dev/sdg1       64     627     564  282K Microsoft basic data
/dev/sdg2      628   17011   16384    8M EFI System
/dev/sdg3    17012 3205631 3188620  1.5G Apple HFS/HFS+
/dev/sdg4  3205632 3206231     600  300K Microsoft basic data

And here are various commands I tried running based on other posts I have seen with no luck (proxmox-boot-tool status proxmox-boot-tool refresh apt install grub-efi-amd64)

root@pve:~# proxmox-boot-tool status
Re-executing '/usr/sbin/proxmox-boot-tool' in new private mount namespace..
E: /etc/kernel/proxmox-boot-uuids does not exist.

root@pve:~# proxmox-boot-tool refresh
Running hook script 'proxmox-auto-removal'..
Running hook script 'zz-proxmox-boot'..
Re-executing '/etc/kernel/postinst.d/zz-proxmox-boot' in new private mount namespace..
No /etc/kernel/proxmox-boot-uuids found, skipping ESP sync.

root@pve:~# apt install grub-efi-amd64
grub-efi-amd64 is already the newest version (2.12-9+pmx2).
The following packages were automatically installed and are no longer required:
  libva-wayland2  libwayland-client0  proxmox-kernel-6.14.5-1-bpo12-pve-signed  proxmox-kernel-6.8.12-11-pve-signed  proxmox-kernel-6.8.12-13-pve-signed
  libva-x11-2     libxfixes3          proxmox-kernel-6.14.8-1-bpo12-pve-signed  proxmox-kernel-6.8.12-12-pve-signed
Use 'apt autoremove' to remove them.

I'm honestly really lost on how I am supposed to fix this. Is anyone able to help or maybe point me in the right direction?

Thank you.

Since I didn't get any responses here, I asked chatgpt and eventually got it worked out. I want to post my fix in case anyone else comes across the same problem.

So my sdd had a 1G partition that used to be /boot/efi (sdd2) but was somehow removed during the pve8to9 update.

I ran blkid to get the UUID of sdd2, then did nano /etc/fstab to input this new line:

UUID=XXXX-XXXX /boot/efi vfat defaults 0 1

(with the UUID in place of the X's)

I ran mount -a and systemctl daemon-reload then did another lsblk to verify /boot/efi showed up on sdd2.

sdd                             8:48   0 238.5G  0 disk
├─sdd1                          8:49   0  1007K  0 part
├─sdd2                          8:50   0     1G  0 part /boot/efi
└─sdd3                          8:51   0 237.5G  0 part
  ├─pve-swap                  252:0    0     8G  0 lvm  [SWAP]
  ├─pve-root                  252:1    0  69.4G  0 lvm  /
  ├─pve-data_tmeta            252:2    0   1.4G  0 lvm

Then I did apt reinstall grub-efi-amd64 and grub-install. After that, I rebooted and proxmox booted up fine.

I don't know how this problem happened in the first place but I'm glad it's fixed now, and hopefully this can help someone else too.

Thanks for this. I had exactly the same thing and solved it the same way. FYI, you can remove the /boot/efi line from fstab afterwards.

Thanks for your post and solution. Same problem here and this fixed it.

Hi @Antics04 - Forgive my ignorance, but where are you running these commands from? Are you running then from the recovery mode option?

When I try to run the recovery option from the Proxmox 9 ISO, I do get the /boot warning, but, then it says there is no kernal loaded and goes back to the menu.

I have tried to run the commands from the Debug option, but Im thinking I need to somehow mount my PVE instance first before I can edit FSTAB.

Has anyone got any ideas? I have exactly this same problem, 1TB SSD, 2nd partition is 1GB and is/was the /boot.

Thanks for any info.

@Aussiebob66 I had the same issue as you. Spent 6 hours trying to figure it out. This is what work for me in the end.

Issues: Could not load the Proxmox Rescue, it would say some thing like "/boot/pve/v..." not found and would go back to the screen to choose what to start.

Solution:

Step 1: Boot from a rescue environment

• Use a live USB: Alternatively, boot from a live USB stick of another Linux distribution like Ubuntu (I used Ubuntu)or SystemRescue, as long as it's compatible with your filesystem (e.g., not ZFS for standard rescue CDs).

Step 2: Mount your root filesystem

• Create a mount point: mkdir /mnt/rescue.
• Mount your root partition to this new directory. You can find your partition using lsblk or fdisk -l. For example, if Proxmox is on /dev/sdX (mine was /nvme0n1p2), use: mount /dev/sdX /mnt/rescue

https://www.youtube.com/watch?v=ZhxBJ3yh2OY&t=460
Step 3: Mount necessary filesystems and chroot

• Bind mount the necessary virtual filesystems:
  • mount --bind /dev /mnt/rescue/dev
  • mount --bind /proc /mnt/rescue/proc
  • mount --bind /sys /mnt/rescue/sys
  • mount --bind /dev /mnt/rescue/dev
• Mount your EFI partition if you have one:
  • mount /dev/sdXY /mnt/rescue/boot/efi (replace sdXY with your EFI partition, e.g., sda1).
• Chroot into your system: chroot /mnt/rescue.

Step 4: Reinstall GRUB

• Install GRUB: Run grub-install to reinstall the GRUB bootloader.
  • If you are on an MBR/BIOS system, you may need grub-install /dev/sdX (where /dev/sdX is the disk, not a partition).
  • If you are on an EFI system, you may need to run grub-install --target=x86_64-efi --efi-directory=/boot/efi --bootloader-id=proxmox.
• Update GRUB configuration: Run update-grub to regenerate the GRUB configuration file.

Step 5: Exit and reboot

• Exit the chroot environment: exit.
• Unmount the filesystems: umount -R /mnt/rescue.
• Reboot the server, removing the rescue media.

Let me know if this works

Thanks @xxcite81 ! Instructions worked for me, though I couldn't mount the root partition directly: I had to use the root path and not the device (/devnvme0n1p3 for me). I used SystemRescue, as the Rescue Boot wasn't working at all for me.

I solved this problem by deleting the line . I had another server, and when I looked at its loading, I found that there was no such line at all. After deleting it, the 9th version loaded)))

Thank you so much for this, it fixed me right up. My partition was there and fstab was correct but grub installed wrong or something.

I used supergrub2 on my pen drive to just boot the proxmox server up natively right into proxmox itself. Bypass the broken grub and boot with supergrub.

Then I just logged in and ran

apt reinstall grub-efi-amd64 and grub-install

And rebooted....done!

https://www.supergrubdisk.org/super-grub2-disk/

Just ran into this exact issue after attempting an upgrade from 8.3.0 to the latest version. After hours of searching online for a fix, found your post. Noticed I had "EFI System" listed on one of my partitions when running fdisk -l but wasn't shown as mounted anywhere when running lsblk, just like OP had. Followed your exact steps and gave my system a reboot and bam! That did the trick.

Thanks for saving my night!

I'm another person saved by this. I did have trouble finding the partition, it is the only one of type vfat with just a block size of 512 and UUIDs. I did notice on my upgrade that it said something about /boot/efi when it was updating grub but it scrolled by too fast to see what it said. It might be related to it asking if i wanted to replace my grub file which had been modified and I said no, because I think I changed it to get my video card passed through.

Thanks @Antics04 that worked for me.

I booted from a Proxmox 9.1 iso usb and into the Recovery boot

Man can I just say a massive thank you for posting this. Same issue and finally found your article which resolved my issue.

Just wanted to bump this cuz I guess it's still an issue with the 8 to 9 upgrade.

Issue still exists on 6.20.26, I have 5 nodes, all worked swimmingly except my one node.

Basically stuck in a reboot loop after 8.4 to 9.2 upgrade. I had to do a boot override and select a different partition (not uefi proxmox), and was let in.

My grub boot cmd line is heavily customised, fortunately I have it saved in Joplin.

Since I’ve upgraded I’ve added a new node, and this one that’s been stuck in my bios can’t get quorum, so I probably have to rejoin cluster.

Curious why this node was affected, did same. Prep and upgrade steps on other nodes in cluster. This machine does have UEFI and legacy set. I think CSM too, that’s not the culprit, right?