Cant reboot after kernel update : error parsing PCC subspaces from PCCT

[NicolasC]

Hello all,

after upgrading proxmox, including the kernel, today, our proxmox server can't boot anymore on the new kernel version.

The newly installed version is Linux 4.15.15.1-pve

While booting, the following hangs for a while :
error parsing PCC subspaces from PCCT
Couldn't get size: 0x80000000 (... a certain number of zero... ) 0e
megaraid_sas 0000:65:00.0: Init cmd return status SUCCESS for SCSI host 0
Reading all physical volumes. This may take a while...

Then, after several minutes, the OS complains on the logical volumes.


When booting on the previous kernel, Linux 4.13.13.6-pve, everything runs fine as far as I can tell.

Any idea on how to fix this issue,
or which manipulation I should carry out to provide more accurate details needed to investigate ?

Thanks very much,
Nicolas

 

[Hartwin]

Just to confirm: you are not the only one. I got the exact same behavior at two identical machines ... including the successful boot to the previous kernel.
These are Proliant ML350p G8 equipped with two E5-2680 v2 and the onboard Smart Array P420i (Firmware 8.32; 2G Cache installed; 6 SAS-drives installed).

After waiting for some time, there will show up some more output from systemd. It tolds, there are several startprocesses stalled for more than 120 seconds.

To have some hints for fixing this issue could help a lot.
Thanks a lot,
Hartwin.

 

[Cyrgia]

Hello everyone.
I have the same issue too, after upgrading an hour ago.
Any help will be appreciated.
Thanks.

Edit: i forgot to mention, i have exactly the same specs as Hartwin (Proliant ML350p G8 etc...)

 

[Hartwin]

https://forum.proxmox.com/threads/4-15-based-test-kernel-for-pve-5-x-available.42097/page-4

Someone seems to running mad ... They choose to set the new default kernel to 4.15.15.1
They say "having a kernel that never shows any issues on all possible hardware is an illusion" ... because there are some now unsupported bits of hardware.
Oops ... most of the servers out there running SAS-Controllers of the same type. Doubleoops - this is one of the now unsupported bits of hardware.
The linked thead currently explodes. Failures reported about nearly every major server brand.

 

[Hartwin]

https://forum.proxmox.com/threads/4...or-pve-5-x-available.42097/page-4#post-208348
provided the information, it's a bug in kernel, fixed in 4.15.0-19.20 ... so maybe there will be an other kernel-update soon.

 

[Cyrgia]

https://forum.proxmox.com/threads/4-15-based-test-kernel-for-pve-5-x-available.42097/page-4
it work for me.

 

[NicolasC]

Upgrading today to 4.15.17 solved my issue.