Kernel panic on boot after upgrading to 25.10.4 - Ryzen 7 5800X / X570S AORUS MASTER

Hi all,

After upgrading from TrueNAS 24.10.2.4 to any 25.x release (tested both 25.04 and 25.10.4), my system kernel panics on every boot. Rolling back to the 24.10.2.4 boot environment works every time. I’ve tried upgrading multiple times with the same result.

Error:

Kernel panic - not syncing: Attempted to kill init! exitcode=0x00000100

Hardware Bare Metal:

  • Motherboard: Gigabyte X570S AORUS MASTER, BIOS F4e 07/20/2022

  • CPU: AMD Ryzen 7 5800X (family 0x19, model 0x21, stepping 0x2)

  • RAM**:** 64GB DDR4 3200MT/s ECC (2x 32GB)

  • GPU: NVIDIA GeForce RTX 3090 (GA102)

  • HBA: Broadcom/LSI SAS2008

  • NIC: QLogic cLOM8214 dual 10GbE

  • NIC: Intel I225-V 2.5GbE

  • Boot drive: Lexar 128GB SATA SSD

  • NVMe: 2x Crucial P2 465GB

  • Data: 6x 14.55TB SATA drives

Working: TrueNAS 24.10.2.4 (kernel 6.6.44)
Failing: TrueNAS 25.04 and 25.10.4 (kernel 6.12.91)

Boot parameters (identical on both working and failing):

amd_iommu=on iommu=pt kvm_amd.npt=1 kvm_amd.avic=1 intel_iommu=on zfsforce=1 nvme_core.multipath=N

What I’ve checked:

  • Boot pool is healthy with no errors

  • Boot drive SMART is clean

  • No logs survive the panic — it happens too early for journald to write anything

  • Another machine with the same PCIe cards upgraded without any issues

Has anyone seen this or know what might be causing it?

Thanks any help would be appreciated!

I would check Gigabyte for BIOS / UEFI updates first. Is the HBA card up to date with the latest firmware, etc? You can check with sudo sas2flash -list Post the results back using (</>) Preformatted Text mode, if you are unsure of the HBA card status.

I’ll take a look for a BIOS update. Below is the result of the sas2flash command.
I checked my other box with the same card that did update and it has the same firmware version on it.

LSI Corporation SAS2 Flash Utility
Version 20.00.00.00 (2014.09.18) 
Copyright (c) 2008-2014 LSI Corporation. All rights reserved 

Adapter Selected is a LSI SAS: SAS2008(B2)   

    Controller Number              : 0
    Controller                     : SAS2008(B2)   
    PCI Address                    : 00:0d:00:00
    SAS Address                    : 56c92bf-0-0003-c444
    NVDATA Version (Default)       : 14.01.00.08
    NVDATA Version (Persistent)    : 14.01.00.08
    Firmware Product ID            : 0x2213 (IT)
    Firmware Version               : 20.00.07.00
    NVDATA Vendor                  : LSI
    NVDATA Product ID              : SAS9211-8i
    BIOS Version                   : 07.39.02.00
    UEFI BSD Version               : N/A
    FCODE Version                  : N/A
    Board Name                     : SAS9211-8i
    Board Assembly                 : N/A
    Board Tracer Number            : N/A

    Finished Processing Commands Successfully.
    Exiting SAS2Flash.

Updated to the latest BIOS and I’m still getting the same Kernel panic error as before.

At this point I would consider pulling all PCIe cards just to see if it manages to boot. I’m aware you said another machine works with the same types/models (?) of cards, but you did not say if that machine used the same MB.

Sorry no the other machine is using a different motherboard. They just share the same HBA and NIC pcie cards.

Right, which is why it would be of use to know if the problem occurs in this system, without the PCIe cards.

If it does, you know the issue isn’t likely to be related to the PCIe cards or the drives attached to it. If it doesn’t, you add them back one at a time to see when it reoccurs, starting with the HBA.

Thanks I will try that. Also posting screenshots when its booting and kernel panics.

What are your boot parameters used for? You have some error messages regarding those. SVM is disabled so I am guessing you don’t use virtualization on top of TrueNAS. Both in first screenshot.

The boot parameters are what ever is set by Truenas I haven’t overrid them and no I don’t use virtualization only docker.

Mine are very similar, here’s the complete line from dmesg from my 25.10.4 system:

[ 0.031048] Kernel command line: BOOT_IMAGE=/ROOT/25.10.4@/boot/vmlinuz-6.12.91-production+truenas root=ZFS=boot-pool/ROOT/25.10.4 ro console=tty1 console=ttyS0,115200 libata.allow_tpm=1 amd_iommu=on iommu=pt kvm_amd.npt=1 kvm_amd.avic=1 intel_iommu=on zfsforce=1 nvme_core.multipath=N

I pulled out both pcie cards and it still kernel panics

I’m going to test installing it on a fresh drive to see if that will work.

Do you think the “ACPI Warning” is something? Around 1.5 or 1.7 in screenshots.

Have you disabled the automatic overclocking in the BIOS? TrueNAS is known to be unhappy with any overclocking.

Also, your BIOS is very old. Current for that board is F9a.

Finally, you should implement these BIOS settings known for increasing stability for Ryzen processors.

https://forums.truenas.com/t/nas-random-hangs/38470/3

I do like to point those BIOS changes out when I see people posting about earlier Ryzen crashing, and I see no harm in doing them here as well, but I doubt they are the cause in this specific case. The crashes those changes seek to address happen when the system is idle, not during boot.

I had the overclocking setting off. I did do the update to the bios to F9a and no difference. I will check to see if the bios update re-enabled the overclocking setting.

I’ll also chekcout the link you provided.

Ok just did a test with a fresh install and it works fine. So it has to be something with my existing system that is causing the kernel panic to happen on upgrade.

I ended up just restoring my previous config on the fresh install and everything is working fine now.

Thank you @SmallBarky @neofusion @Samuel_Tai looking into this with me. I appreciate it.

1 Like

My guesses for what might be behind that would be some form of hardware fault. Perhaps an unstable (overclocked?) CPU, bad RAM (due to being faulty or perhaps just overclocked) or possibly a failing boot media.