Opt-in Linux 6.8 Kernel for Proxmox VE 8 available on test & no-subscription

[jsterr]

I got a reply to one of my linkedin posts that recommend the following parameters, can someone with 6.8 and crashes try these out in /etc/kernel/cmdline and apply those pcie_port_pm=off libata.force=noncq and do a proxmox-boot-tool refresh and reboot?

There are also few new issues, mostly related with big jump in kernel version
- when using veeam agents installed on nodes, dkms fails with building and preparing these (silently) which can lead to broken initramfs rebuild - leaving you with unbootable kernel.
Resolve: apt purge veeam blksnapd
- some 'older' hardware seems to have problems with Power Management SSD/NVME drives (magical slowdowns and hangs after some I/O operations)
Resolve: add kernel parameters
pcie_port_pm=off libata.force=noncq

 

[Ivan Dimitrov]

Hi, about the issue with AMD and passthrough - is there a thread or bug report I can follow to know when it is fixed? I see reports in multiple threads but nothing specific like AMD generation that is impacting, etc.

About the AMD stability problems - are they related or independent to the passthrough issues? Again ticket or forum thread to follow?

 

[peteb]

I installed 6.8.5.3 yesterday and my server froze after a few hours. Had to hard reset it and it would freeze again after a few hours.
Kernel 6.8 is too unstable for me. I have now reverted to kernel 6.5.13-5.

 

[Inglebard]

Our fujitsu primergy rx100 s7p decide to crash and no longer boot on kernel 6.8. It works properly during 20 days on 6.8.
It works properly on 6.5.

I now require to set "intel_iommu=off" to boot with 6.8.

 

[SoftLost]

I had to back it down to 6.5 after the server started to slow down I/O operations over days. I have no further information to contribute except it is unusable slow after 2 days. 6.5 kernel runs fine.

 

[vmware]

I've just installed proxmox-kernel-6.8.4-3-pve-signed, and seeing lots of these audit errors in pve's dmesg:

[  +0.016235] audit: type=1400 audit(1717024560.796:117358): apparmor="DENIED" operation="mount" class="mount" info="failed flags match" error=-13 profile="lxc-323_</var/lib/lxc>" name="/run/systemd/unit-root/proc/" pid=3055651 comm="(d-logind)" fstype="proc" srcname="proc" flags="rw, nosuid, nodev, noexec"
[  +0.016115] audit: type=1400 audit(1717024560.812:117359): apparmor="DENIED" operation="mount" class="mount" info="failed flags match" error=-13 profile="lxc-323_</var/lib/lxc>" name="/run/systemd/unit-root/proc/" pid=3055655 comm="(d-logind)" fstype="proc" srcname="proc" flags="rw, nosuid, nodev, noexec"
[May29 23:16] audit: type=1400 audit(1717024575.376:117360): apparmor="DENIED" operation="mount" class="mount" info="failed flags match" error=-13 profile="lxc-322_</var/lib/lxc>" name="/run/systemd/unit-root/proc/" pid=3055750 comm="(d-logind)" fstype="proc" srcname="proc" flags="rw, nosuid, nodev, noexec"
[  +0.015229] audit: type=1400 audit(1717024575.392:117361): apparmor="DENIED" operation="mount" class="mount" info="failed flags match" error=-13 profile="lxc-322_</var/lib/lxc>" name="/run/systemd/unit-root/proc/" pid=3055754 comm="(d-logind)" fstype="proc" srcname="proc" flags="rw, nosuid, nodev, noexec"
[  +0.014810] audit: type=1400 audit(1717024575.406:117362): apparmor="DENIED" operation="mount" class="mount" info="failed flags match" error=-13 profile="lxc-322_</var/lib/lxc>" name="/run/systemd/unit-root/proc/" pid=3055758 comm="(d-logind)" fstype="proc" srcname="proc" flags="rw, nosuid, nodev, noexec"
[  +0.015541] audit: type=1400 audit(1717024575.422:117363): apparmor="DENIED" operation="mount" class="mount" info="failed flags match" error=-13 profile="lxc-322_</var/lib/lxc>" name="/run/systemd/unit-root/proc/" pid=3055762 comm="(d-logind)" fstype="proc" srcname="proc" flags="rw, nosuid, nodev, noexec"
[  +0.014464] audit: type=1400 audit(1717024575.436:117364): apparmor="DENIED" operation="mount" class="mount" info="failed flags match" error=-13 profile="lxc-322_</var/lib/lxc>" name="/run/systemd/unit-root/proc/" pid=3055766 comm="(d-logind)" fstype="proc" srcname="proc" flags="rw, nosuid, nodev, noexec"
[ +11.316393] audit: type=1400 audit(1717024586.753:117365): apparmor="DENIED" operation="mount" class="mount" info="failed flags match" error=-13 profile="lxc-323_</var/lib/lxc>" name="/run/systemd/unit-root/proc/" pid=3055849 comm="(d-logind)" fstype="proc" srcname="proc" flags="rw, nosuid, nodev, noexec"
[  +0.015839] audit: type=1400 audit(1717024586.769:117366): apparmor="DENIED" operation="mount" class="mount" info="failed flags match" error=-13 profile="lxc-323_</var/lib/lxc>" name="/run/systemd/unit-root/proc/" pid=3055853 comm="(d-logind)" fstype="proc" srcname="proc" flags="rw, nosuid, nodev, noexec"

Interestingly, the 2 CT's (322, 323) are relatively new - Debian Bookworm created in Proxmox 8.1. The old CTs created in Proxmox 7.x have no errors.

arch: amd64
cmode: tty
console: 1
cores: 2
cpulimit: 0
cpuunits: 1024
hostname: foo
memory: 4096
nameserver: 192.168.1.1
net0: name=eth0,bridge=vmbr30,hwaddr=foo,ip=dhcp,ip6=auto,type=veth
onboot: 1
ostype: debian
protection: 0
rootfs: local-lvm:vm-323-disk-0,size=64G
swap: 0
tty: 2
unprivileged: 1

Help appreciated thanks!

 

[gfngfn256]

Help appreciated thanks!

Try turning on nesting in LXC Features

 

[vmware]

Try turning on nesting in LXC Features

Nice thanks. nesting has fixed the issue. I guess because I used Terraform to provision the CT's, it messed up the default of nesting=1

 

[gfngfn256]

I used Terraform to provision the CT's

I suspected you used some provisioning method. Glad you got it working.