pve7to8 checklist script warning - unified cgroup v2 layout

[Patrik385]

When running the pve7to8 checklist script I get this warning:
"WARN: Found at least one CT (157) which does not support running in a unified cgroup v2 layout. Consider upgrading the Containers distro or set systemd.unified_cgroup_hierarchy=0 in the Proxmox VE hosts' kernel cmdline! Skipping further CT compat checks."

CT 157 is the latest Rocky Linux 9.2 container distro.

Searching Proxmox help pages I found this:
https://pve.proxmox.com/wiki/Upgrade_from_6.x_to_7.0#Old_Container_and_CGroupv2

So, if the container is an old distribution like CentOS 7 or Ubuntu 16.10 this problem appears.

But this cannot be true for new distributions, especially for the latest.

I have installed Rocky Linux 9 container distro on a Proxmox 8 (new installation) an everything went fine.

Can someone explain why this warning appears when running the pve7to8 checklist script? Is this a problem with the script? I assume I would get the same warning with Alma Linux 9 and CentOS Stream 9 container distros.

 

[hlindqvist]

When running the pve7to8 checklist script I get this warning:
"WARN: Found at least one CT (157) which does not support running in a unified cgroup v2 layout. Consider upgrading the Containers distro or set systemd.unified_cgroup_hierarchy=0 in the Proxmox VE hosts' kernel cmdline! Skipping further CT compat checks."

CT 157 is the latest Rocky Linux 9.2 container distro.

Searching Proxmox help pages I found this:
https://pve.proxmox.com/wiki/Upgrade_from_6.x_to_7.0#Old_Container_and_CGroupv2

So, if the container is an old distribution like CentOS 7 or Ubuntu 16.10 this problem appears.

But this cannot be true for new distributions, especially for the latest.

I have installed Rocky Linux 9 container distro on a Proxmox 8 (new installation) an everything went fine.

Can someone explain why this warning appears when running the pve7to8 checklist script? Is this a problem with the script? I assume I would get the same warning with Alma Linux 9 and CentOS Stream 9 container distros.

I cannot actually explain it, but I did have the same thing happen on one host that I was upgrading (I think it was a Fedora 38 CT that triggered it in my case).

I noted down the warning and then proceeded with the upgrade, thinking that I might have to either rework something with that CT or maybe even add that kernel command line argument, but I was pleasantly surprised to see that it all just worked.

So my impression is that there may be some scenario where the pve7to8 script gives a false positive for that check.