My first install of Proxmox 8.2 on an Intel NUC - no success

[jolene]

I want to share my experience so far with Proxmox 8.2-1. I bough an intel NUC to install it. Setup the BIOS and used latest version.

Downloaded the ISO, verified checksum, flashed 3 USB sticks with the command from the documentation:
`dd bs=1M conv=fdatasync if=./proxmox-ve_8.2-1.iso of=/dev/sdX`

1. I am trying to use the Graphical install since I am a beginner. WIth a chance of 40% it will fail to load X11 with `Bus error`, `X Server Temporarily Unavailable ` I just rebooted and it works.

Then I pass the whole wizard, with network configuration, root password, asf, confirm. Everytime I get different errors:

2. First a libgfrpg0_10.3+5_amd64.deb fails with `lzma error: compressed data corrupt`. I try another new flash drive with the image flashed with dd again. This time I get:

3. Command `unsquashfs -f -dest /target -i /cdrom/pve-base.squashfs` failed with exit code 1 at /usr/share/perl5/Proxmox/install.pm line 908`. Output:

umount: /target/run: not mounted.
umount: /target/tmp/pkg: no mount point specified.
umount: /target/tmp: not mounted.
umount: /target/proc: not mounted.
umount: /target/sys/firmware/efi/efivars: no mount point specified.
umount: /target/sys: not mounted
(II) AIGLX: Suspending AIGLX clients for VT switch

I am trying to understand what is going on but with no success. Does someone has an idea?

 

[jolene]

It tried Balena Etcher on different USB sticks, same it failed in some other .deb (librados2_17.2.7-pve3_amd64.deb).
I also run full suite of Memtest86 and 0 errors on my memory modules.

Getting even more hopeless.

 

[Kingneutron]

Try re-downloading the iso as torrent

The latest 8.2 is only ~1.3GB, can you try actually burning it to a DVD and boot from that?

Honestly I don't trust random usb sticks anymore, either I Ventoy boot from a known-good Kanguru USB3 with a physical write-protect switch or just use an sd card with an adapter.

 

[jolene]

@Kingneutron thanks for your reply, I redownloaded and verified the sha256sum. I have a brand new Samsung USB stick I also tried and the error continues. I cannot make Proxmox VE 8.2-1 complete the installation without those corrupt DEB errors no matter what I do. I think I will try another version.

 

[Kingneutron]

The only thing I can think of atm is if you have a Win10 desktop (or an Intel-based Mac) install Virtualbox on that and try booting your proxmox install ISO in a VM to see if it completes. If it does, it's not a problem with the iso

 

[Porencephaly13]

Did you solve this? I'm having the same issue with trying to install PBS.

 

[jolene]

No, I actually downloaded both ISOs 8.2 and 8.1 in torrents.. an interesting correlation, when I flash with Rufus/Balena etcher, I get:

On 8.2 error:

'librados2_17.2.7-pve3_amd64.deb' (size=3412020) member 'data.tar': lzma error: compressed data is corrupt
dpkg-deb error: <decompress> subprocess returned error exit status 2
dpkg: error processing archive /tmp/pkg/librados2_17.2.7-pve3_amd64.deb (--unpack):
  cannot copy extracted data for './usr/lib/ceph/libceph-common.so.2' to '/usr/lib/ceph/libceph-common.so.2.dpkg-new': unexpected end of file or stream
Errors were encountered while processing:
  librados2_17.2.7-pve2_amd64.deb
installation of package librados2_17.2.7-pve2_amd64.deb failed
umount: /target/sys/firmware/efi/efivars: no mount point specified.
(II) AIGLX: Suspending AIGLX clients for VT switch

On 8.1 error:

'librados2_17.2.7-pve2_amd64.deb' (size=3363056) member 'data.tar': lzma error: compressed data is corrupt
dpkg-deb error: <decompress> subprocess returned error exit status 2
dpkg: error processing archive /tmp/pkg/librados2_17.2.7-pve2_amd64.deb (--unpack):
  cannot copy extracted data for './usr/lib/ceph/libceph-common.so.2' to '/usr/lib/ceph/libceph-common.so.2.dpkg-new': unexpected end of file or stream
Errors were encountered while processing:
  librados2_17.2.7-pve2_amd64.deb
installation of package librados2_17.2.7-pve2_amd64.deb failed
umount: /target/sys/firmware/efi/efivars: no mount point specified.
(II) AIGLX: Suspending AIGLX clients for VT switch

I also found a post for 2020 of a guy having similar problem with the exact same Intel NUC model (NUC7i3BNH).

See a pattern? But anyway do not know what to do with that.

 

[Kingneutron]

It looks like this model of NUC is not compatible with Proxmox - I would recommend you return it and go with something different if you tested the ISO install in a VM and it succeeded (as recommended earlier)

Some manufacturers only test their equipment with Windows, odd that it's still a thing in 2024 where Windows as a desktop is dying

 

[alexskysilk]

The errors you are experiencing appear to be either frame buffer failure or decompression errors. the commonality between these functions on a unified memory hardware (eg, you nuc) is memory faults. run a full memtest before continuing troubleshooting.

 

[jolene]

@alexskysilk the only problem is that I already did a full test using memtest86, 4 passes, all tests, 0 errors. do you have a suggestion if I should do something different?

 

[alexskysilk]

well, you have two options-
1. use the text installer
2. install debian bookworm

if the text installer failed also, but debian installed you can then proceed to add the pve repos and install proxmox the rest of the way as described here: https://pve.proxmox.com/wiki/Install_Proxmox_VE_on_Debian_12_Bookworm

 

[jolene]

@alexskysilk I installed Debian 12, 0 problems. Then Proxmox on top of it, following the docs. It all worked, disabled enterprise, created interface. But after all that, the first time I restarted now I see:

[FAILED] Failed to start pve-cluster.service - The Proxmox VE cluster filesystem.
[FAILED] Failed to start pve-cluster.service - The Proxmox VE cluster filesystem.
[FAILED] Failed to start pve-cluster.service - The Proxmox VE cluster filesystem.
[FAILED] Failed to start pve-cluster.service - The Proxmox VE cluster filesystem.
[FAILED] Failed to start pve-cluster.service - The Proxmox VE cluster filesystem.
[FAILED] Failed to start pvestatd.service - The PVE status daemon.
[FAILED] Failed to start pve-cluster.service - The Proxmox VE cluster filesystem.
[FAILED] Failed to start pve-firewall.service - Proxmox VE firewall.

 

[alexskysilk]

you need to pay close attention to the instructions about hostname, hosts file, etc. this is critical.

 

[jolene]

you need to pay close attention to the instructions about hostname, hosts file, etc. this is critical.

yes, /etc/hosts was missing a name. thanks a lot.

 

[jolene]

@alexskysilk I have just 1 ssd on my NUC for the OS and VMs. It comes with just a local disk, I believe on previous versions it came with local and local-lvm. Should I create local-lvm to create the VMs?

 

[alexskysilk]

not necessarily. how is your disk organized? (lsblk, vgs, lvs)

 

[jolene]

not necessarily. how is your disk organized? (lsblk, vgs, lvs)

lsblk
NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINTS
sda 8:0 0 232.9G 0 disk
├─sda1 8:1 0 512M 0 part /boot/efi
├─sda2 8:2 0 231.4G 0 part /
└─sda3 8:3 0 977M 0 part [SWAP]

Currently vgs and lvs show as empty.

 

[alexskysilk]

pve should have created a "local" entry at /var/lib/vz. continue to use that for everything.

 

[jolene]

It's it slower than if I reinstall Debian and everything in a full LVM?

 

[justinclift]

if=./proxmox-ve_8.2-1.iso

As a data point, there seem to be several people having problems with the Proxmox 8.2 ISO installer... but when they go back to the Proxmox 8.1 ISO installer things work ok.

Not sure if that's useful information now though, as you've taken a different path for installing things...