Proxmox 9.2 installer hangs: waiting for /dev to be fully populated

S

slavik-f

Member
May 22, 2024
17
10
8
USA
System:
- Dell Precision T7960
- Intel Xeon W5-3425
- Nvidia RTX 4090D
- 2 of NIC Intel cards

Starting Proxmox installer from USB ISO.
Tried Graphic mode, Text mode, used nomodeset. Same result:

log out put prints few lines. Last line:
waiting for /dev to be fully populated

And then it hangs.

If I use Proxmox 9.1 ISO - it loads with no problem.

I had Proxmox 9.1 system up and running. After I ran `pveupgrade` - I can't boot it anymore.
 

Attachments

  • tui1.jpeg
    tui1.jpeg
    193.7 KB · Views: 26
Last edited:
  • Like
Reactions: viktis
I also tried DEBUG mode.

Below, the output from debug boot log:

```
I'm booting Proxmox installer on Dell Precision workstaiton,
I'm using nomodeset parameter.
It starts and hangs.

Here is logs:

```
Proxmox VE 9.2 (iso release 1) - https://www.proxmox.com/

Loading Proxmox TUI Installer (debug) ...
Loading initial ramdisk ...

Welcome to the Proxmox VE 9.2 (ISO 1) installer!
UEFI boot mode detected, mounting efivars filesystem
modprobe: FATAL: Module shpchp not found in directory /lib/modules/7.0.2-6-pve
nvidiafb: Device ID: 10de2685
nvidiafb: unknown NV_ARCH

searching for block device containing the ISO proxmox-ve-9.2-1
with ISO ID '019e4a4f-460a-76ae-9662-b996097dce93'
found Proxmox VE ISO

preparing installer mount points and working environment

GPT: Primary header thinks Alt. header is not at the end of the disk.
GPT:3332379 != 488570879
GPT: Alternate GPT header not at the end of the disk.
GPT: Use GNU Parted to correct GPT errors.
sdc: sdc1 sdc2 sdc3 sdc4

[10.338541] nouveau: Unknown symbol drm_atomic_get_old_mst_topology_state (err -2)
[10.338714] nouveau: Unknown symbol drm_dp_mst_put_port_malloc (err -2)
[10.338879] nouveau: Unknown symbol drm_dp_atomic_release_time_slots (err -2)
[10.339037] nouveau: Unknown symbol drm_gpuvm_bo_evict (err -2)
[10.339193] nouveau: Unknown symbol drm_gpuvm_sm_unmap_ops_create (err -2)
[10.339347] nouveau: Unknown symbol drm_gpuva_link (err -2)
[10.339503] nouveau: Unknown symbol ttm_a?sp_unbind (err -2)
[10.339646] nouveau: Unknown symbol drm_dp_lttpr_init (err -2)
[10.339793] nouveau: Unknown symbol drm_gem_ttm_vmap (err -2)
[10.339953] nouveau: Unknown symbol drm_dp_cec_register_connector (err -2)
[10.340107] nouveau: Unknown symbol drm_exec_init (err -2)
[10.340279] nouveau: Unknown symbol drm_dp_mst_detect_port (err -2)
[10.340450] nouveau: Unknown symbol drm_gem_ttm_mmap (err -2)
[10.340603] nouveau: Unknown symbol drm_gpuva_remove (err -2)
[10.341422] nouveau: Unknown symbol drm_dp_read_dpcd_caps (err -2)
[10.341572] nouveau: Unknown symbol drm_atomic_get_mst_topology_state (err -2)
[10.341738] nouveau: Unknown symbol ttm_device_init (err -2)
[10.341896] nouveau: Unknown symbol ttm_device_fini (err -2)
[10.342032] nouveau: Unknown symbol drm_gpuvm_exec_lock (err -2)
[10.342190] nouveau: Unknown symbol drm_dp_check_act_status (err -2)
[10.342341] nouveau: Unknown symbol drm_gpuvm_sm_map_ops_create (err -2)
[10.342493] nouveau: Unknown symbol drm_gpuva_insert (err -2)
[10.342628] nouveau: Unknown symbol drm_edp_backlight_init (err -2)
[10.342774] nouveau: Unknown symbol drm_gem_ttm_dumb_map_offset (err -2)
[10.342923] nouveau: Unknown symbol drm_sched_job_add_dependency (err -2)
[10.343067] nouveau: Unknown symbol ttm_bo_unpin (err -2)
[10.343214] nouveau: Unknown symbol ttm_bo_vm_fault_reserved (err -2)
[10.343395] nouveau: Unknown symbol ttm_pool_free (err -2)
[10.343540] nouveau: Unknown symbol drm_dp_cec_set_edid (err -2)
[10.343688] nouveau: Unknown symbol drm_dp_get_vc_payload_bw (err -2)
[10.343826] nouveau: Unknown symbol drm_gem_ttm_vunmap (err -2)
[10.343988] nouveau: Unknown symbol drm_dp_read_sink_count_cap (err -2)
[10.344124] nouveau: Unknown symbol drm_dp_read_downstream_info (err -2)
[10.344278] nouveau: Unknown symbol drm_dp_cec_irq (err -2)
[10.344476] nouveau: Unknown symbol drm_edp_backlight_set_level (err -2)
[10.344622] nouveau: Unknown symbol drm_sched_fini (err -2)
[10.344822] nouveau: Unknown symbol ttm_bo_setup_export (err -2)
[10.344969] nouveau: Unknown symbol drm_sched_entity_fini (err -2)
[10.345132] nouveau: Unknown symbol ttm_bo_move_to_lru_tail (err -2)
[10.345331] nouveau: Unknown symbol drm_gpuvm_validate (err -2)
[10.345475] nouveau: Unknown symbol drm_gpuvm_bo_extobj_add (err -2)
[10.345658] nouveau: Unknown symbol drm_fbdev_ttm_driver_fbdev_probe (err -2)
[10.345809] nouveau: Unknown symbol drm_dp_read_lttpr_common_caps (err -2)
[10.345958] nouveau: Unknown symbol drm_exec_fini (err -2)
[10.346127] nouveau: Unknown symbol drm_gpuva_unmap (err -2)
[10.346279] nouveau: Unknown symbol ttm_resource_manager_evict_all (err -2)

[10.362333] NovaCore 0000:ac:00.0: NVIDIA (Chipset: AD102, Architecture: Ada, Revision: a.1)
[10.537808] atlantic 0000:02:00.0 enp2s0: renamed from eth0
[10.539157] Creating 1 MTD partitions on "0000:00:1f.5":
[10.539348] 0x000000000000-0x000001000000 : "BIOS"
[10.543348] spi-nor spi0.1: supply vcc not found, using dummy regulator
[10.543564] input: Dell Privacy Driver as /devices/platform/PNP0C14:02/wmi_bus/.../input/input10
[10.544144] input: Dell WMI hotkeys as /devices/platform/PNP0C14:02/wmi_bus/.../input/input11
[10.545196] Creating 1 MTD partitions on "0000:00:1f.5-chip1":
[10.545634] 0x000000000000-0x000002000000 : "BIOS1"

[10.560837] EDAC MC0: Giving out device to module i10nm_edac controller Intel_10nm Socket#0 IMC#0: DEV 0000:fe:0c.0 (INTERRUPT)
[10.561166] EDAC MC1: Giving out device to module i10nm_edac controller Intel_10nm Socket#0 IMC#1: DEV 0000:fe:0d.0 (INTERRUPT)
[10.561599] EDAC MC2: Giving out device to module i10nm_edac controller Intel_10nm Socket#0 IMC#2: DEV 0000:fe:0e.0 (INTERRUPT)
[10.561847] EDAC MC3: Giving out device to module i10nm_edac controller Intel_10nm Socket#0 IMC#3: DEV 0000:fe:0f.0 (INTERRUPT)
[10.562080] EDAC i10nm: v0.0.6
[10.566623] ixgbe: Intel(R) 10 Gigabit PCI Express Network Driver
[10.566953] ixgbe: Copyright (c) 1999-2016 Intel Corporation.
[10.868329] e1000e 0000:00:1f.6 0000:00:1f.6 (uninitialized): registered PHC clock
[10.955640] e1000e 0000:00:1f.6 eth0: (PCI Express:2.5GT/s:Width x1) 00:be:43:be:15:3e
[10.955794] e1000e 0000:00:1f.6 eth0: Intel(R) PRO/1000 Network Connection
[10.955989] e1000e 0000:00:1f.6 eth0: MAC: 15, PHY: 12, PBA No: FFFFFF-0FF
[10.957714] e1000e 0000:00:1f.6 enp0s31f6: renamed from eth0
```
 
  • Like
Reactions: viktis
And when I attempt to boot from Proxmox 9.2 ISO in the console mode:

KERNEL PANIC!
 

Attachments

  • console.jpeg
    console.jpeg
    61.2 KB · Views: 19
Installed Proxmox from 9.1 ISO:

Code: 
uname -a
Linux t7960 6.17.2-1-pve #1 SMP PREEMPT_DYNAMIC PMX 6.17.2-1 (2025-10-21T11:55Z) x86_64 GNU/Linux

Then switched to `no-subscription` repo and upgraded:

Code: 
apt update
pveupgrade

Computer rebooted, printed few boot lines and stopped the output (screen frozen).
However, now web-access and SSH-access is working:

Code: 
uname -a
Linux t7960 7.0.2-6-pve #1 SMP PREEMPT_DYNAMIC PMX 7.0.2-6 (2026-05-20T08:55Z) x86_64 GNU/Linux

But on-screen, I can't access terminal.

see the diff for `dmesg` output for 6.17.2 vs 7.0.2 kernel on that workstation:

https://gist.github.com/SlavikCA/99e1f922e511f04f2b60f38cc542ec15/revisions

Code: 
lspci -nnk | grep -A 2 -i nvidia

ac:00.0 VGA compatible controller [0300]: NVIDIA Corporation AD102 [GeForce RTX 4090 D] [10de:2685] (rev a1)
    Subsystem: NVIDIA Corporation Device [10de:1697]
    Kernel driver in use: NovaCore
    Kernel modules: nvidiafb, nouveau, nova_core
 
Last edited:
1. I think that because of NovaCore, the Proxmox 9.2 doesn't work with Nvidia RTX cards. `nomodeset` doesn't help.

2. In some cases, during upgrade, something is failing / misconfigured with the network stack, so even SSH access is not available on upgrade.
 
  • Like
Reactions: viktis
Ok the fix was fairly easy : Scroll down to the line starting with linux. Press the right arrow key to go to the very end of that line, and add nomodeset (ensure there is a space separating it from the previous parameter).
No other changes needed, system started normally
 
Seems like 9.2 is having serious issues for everyone running RTX cards. I wonder if it's happening only on systems that don't have an iGPU?
 
slavik-f said:
Installed Proxmox from 9.1 ISO:

Code: 
uname -a
Linux t7960 6.17.2-1-pve #1 SMP PREEMPT_DYNAMIC PMX 6.17.2-1 (2025-10-21T11:55Z) x86_64 GNU/Linux

Then switched to `no-subscription` repo and upgraded:

Code: 
apt update
pveupgrade

Computer rebooted, printed few boot lines and stopped the output (screen frozen).
However, now web-access and SSH-access is working:

Code: 
uname -a
Linux t7960 7.0.2-6-pve #1 SMP PREEMPT_DYNAMIC PMX 7.0.2-6 (2026-05-20T08:55Z) x86_64 GNU/Linux

But on-screen, I can't access terminal.

see the diff for `dmesg` output for 6.17.2 vs 7.0.2 kernel on that workstation:

https://gist.github.com/SlavikCA/99e1f922e511f04f2b60f38cc542ec15/revisions

Code: 
lspci -nnk | grep -A 2 -i nvidia

ac:00.0 VGA compatible controller [0300]: NVIDIA Corporation AD102 [GeForce RTX 4090 D] [10de:2685] (rev a1)
    Subsystem: NVIDIA Corporation Device [10de:1697]
    Kernel driver in use: NovaCore
    Kernel modules: nvidiafb, nouveau, nova_core
Click to expand...
After over 10 hours of troubleshooting and having to dig through a bunch of docs...

Still no clue why RTX cards are freezing up on the installer even with nomodeset on the 9.2 install. 9.1 install with nomodeset will complete the install and run successfully. Updating the following files:

/etc/apt/sources.list.d/ceph.sources
Types: deb
URIs: http://download.proxmox.com/debian/ceph-squid
Suites: trixie
Components: no-subscription
Signed-By: /usr/share/keyrings/proxmox-archive-keyring.gpg

/etc/apt/sources.list.d/pve-enterprise.sources
Types: deb
URIs: http://download.proxmox.com/debian/pve
Suites: trixie
Components: pve-no-subscription
Signed-By: /usr/share/keyrings/proxmox-archive-keyring.gpg

Then running your commands, I am able to run the latest version. The console does still hang but SSH/web works. No idea how to fix the console but at this moment I don't really care. Hopefully someone at the dev team can figure this out and issues a patch. I would expect a lot more complaints rolling through as this seems to be a fairly large problem.
 
These files have nothing to do with gpu issues though. Disabling the Enterprise Repos and enabling no-subscription repo should be possible from the web ui
 
Johannes S said:
These files have nothing to do with gpu issues though. Disabling the Enterprise Repos and enabling no-subscription repo should be possible from the web ui
Click to expand...
Never said they did.

The point is the 9.1 installer and build works and in order to upgrade to 9.2 you have to swap to the free repos.

9.2 installer and build is having issues directly. Something was changed in the installer that is causing issues but upgrading from 9.1 works fine.

The 9.2 installer needs to be looked at.
 
You must log in or register to reply here.
Top Bottom
Back