Restoring LXC's from Proxmox 6.x -> 8.x Network Issue

[JakeC]

Hi All,
I was using Proxmox 6.x for a long while, and finally had some time to start fresh and install 8.x.
The install went fine, and new LXC containers work fine.
I tried restoring a centos 7.2 container from backup from the proxmox v6 setup, and the restore seemed to go ok.
I was able to start the container, but not connect with the console.
I noticed the warning when starting:
WARN: old systemd (< v232) detected, container won't run in a pure cgroupv2 environment! Please see documentation -> container -> cgroup version.

After some googling it looks like PVE 8 will be the last to support the hybrid cgroup versions, and the way to enable that is to modify this line in /etc/default/grub

GRUB_CMDLINE_LINUX_DEFAULT="quiet systemd.unified_cgroup_hierarchy=0"

Then run

update-grub

, and reboot.

So I tried that, and I still get the same warning as before when starting the LXC.
First question, should the warning go away after the modification to the grub file?

From here I am able to pct enter <LXC ID> and get into the container.
The files seem to be there which is good news.

However I am not able to get the network connectivity going.
Running

ip a

returns:

1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host
       valid_lft forever preferred_lft forever
2: eth0@if12: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN qlen 1000
    link/ether 8a:c4:f0:35:e7:94 brd ff:ff:ff:ff:ff:ff link-netnsid 0

The ifcfg-eth0 file looks like:

[root@HostTest ~]# cat /etc/sysconfig/network-scripts/ifcfg-eth0
DEVICE=eth0
ONBOOT=yes
UUID=47ffd31e-697d-4ebc-9405-42c312746d56
BOOTPROTO=none
IPADDR=192.168.1.203
NETMASK=255.255.0.0
GATEWAY=192.168.0.1
DNS1=192.168.0.1
DOMAIN=example.com

I thought it was weir that the link state was DOWN, so I ran

ip link set eth0 up

Interestingly enough my eero notified me of a new device joining the network.

Running ip a now shows:

1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host
       valid_lft forever preferred_lft forever
2: eth0@if12: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP qlen 1000
    link/ether 8a:c4:f0:35:e7:94 brd ff:ff:ff:ff:ff:ff link-netnsid 0
    inet6 fd08:72bf:6019:d8bf:88c4:f0ff:fe35:e794/64 scope global mngtmpaddr dynamic
       valid_lft 1741sec preferred_lft 1741sec
    inet6 fdf3:cf86:dc67:1:88c4:f0ff:fe35:e794/64 scope global mngtmpaddr dynamic
       valid_lft 2591941sec preferred_lft 604741sec
    inet6 fe80::88c4:f0ff:fe35:e794/64 scope link
       valid_lft forever preferred_lft forever

No ipv4 address.

pinging google.com returns "ping: unknown host google.com"


Second Question: Are my networking issues related to the cgroup warning? any suggestions for getting the network working again on my older restored containers?


Thank you very much for any insight or suggestions!

 

[JakeC]

I feel like I'm missing something obvious here.

What behavior should I see when adding:
GRUB_CMDLINE_LINUX_DEFAULT="quiet systemd.unified_cgroup_hierarchy=0"

Should I still get the warning when starting a CentOS 7 container?

They started ok (still no networking) even without the unified_cgroup_hierarchy setting in the grub settings.
Should that even be the case?

Thanks!