[SOLVED] HP DL385 Gen10 kernel panic new install

[martijnk]

Hello all,

I'm getting this error on a new Proxmox installation using the latest ISO 6.1-2.

After that it just panics.

I did a reinstall without result. Any idea?

 

[oguz]

hi,

can you check the sum of the image and see if it matches with the website?
how did you burn the ISO?

i found this thread[0] which says:

VFS: unable to mount root fs on unknown-block(0 0) means the kernel was unable to mount the root filesystem. There are two common causes for this:

1. The kernel doesn't support the filesystem on the device. If you compiled your own kernel, this is usually because you specified the filesystem driver should be built as a module rather than a native part of the kernel; if you're using the distro's kernel, this is usually because you picked an exotic format for your root filesystem. In either case, don't do that.
2. The name of the root device passed to the kernel is wrong. This one can be tricky to fix: the best method I've found is to modify the kernel command line from the bootloader, making educated guesses about what the root= parameter should look like until I find something that works.

which filesystem did you use during setup?

maybe something went wrong in the setup... it should be possible to chroot into the system with a livecd (pve installer in debug mode can be used)[1]

[0]: https://unix.stackexchange.com/ques...s-unable-to-mount-root-fs-on-unknown-block0-0
[1]: https://pve.proxmox.com/wiki/Installation

 

[martijnk]

Hi I don't burn the iso I just mount it in ILO. I did a reinstall 3 times from scratch, same result. I'm just using standard ext4 filesystem during setup (server has hardware raid).

I used the exact same ISO on 4 other servers, Dell 730rx and Poweredge 610 and no problems there. So it must be something with the DL385.

 

[martijnk]

Ok update, so I download the older ISO, PV 6.0-1. That one installed and booted fine.
Then I did a apt update from the no-subscription repository to update to the latest version and boom, same error: couldn't find suitable memory target.

I read this error is caused by a too big initrd. Anyway my initramfs can't load hence the kernel panic.

So i will just stick with 6.0-1 for now. It's not ideal but it's doable. I'm pretty sure this is a bug though.

 

[Stoiko Ivanov]

* how much memory does the server have installed?
* how much memory is detected once the system is booted?

 

[martijnk]

Hi Stoiko,

The system has 256 GB memory. I can't answer your second question I don't get that far

I also tested the new kernel for Proxmox 6.2. But same result. I might have to take this to HP though, I read in other topics that a BIOS firmware update solves the issue, unfortunately I'm on the latest firmware already.

 

[Stoiko Ivanov]

I can't answer your second question I don't get that far

sorry - did not think this through completely ...
does any linux live-cd boot? (debian installer, archiso, ubuntu)? - if yes - free -h in a shell would tell as much

I might have to take this to HP though

probably a good idea - keep us posted if something comes up in that investigation

 

[martijnk]

Yes they all boot. CentOS 8, Debian 10 and I already had XenServer and ESXi on it without any issues.

It starts failing with anything newer than Proxmox 6.0-1 but it works on my Dell servers which is weird.

 

[martijnk]

Ok another update, I worked around the issue by switching bios from UEFI boot to Legacy BIOS mode. Apparently something in HP's UEFI implementation causes this issue.

I don't like it, I rather use UEFI but performance wise it shouldn't matter so I will leave it at this. I'm glad it's working now. I might try again over a year of new BIOS updates

 

[Stoiko Ivanov]

glad you found a workaround - should the issue ever be fixed (insofar as that booting and installing with UEFI works) - and you still remember - we'd be grateful for a comment here.

please mark the thread as 'SOLVED'