Virtual interface not visible from container

[alex3137]

Hello

I updated my Proxmox host today to 4.2-18/158720b9

Since then, I have containers on which the network interface does not show up (ifconfig from the container returns empty result) and therefore I have no network connectivity. These containers used to work fine in the past.

Can you help me to troubleshoot ?

Thanks

 

[manu]

ifconfig will give an empty answer if the network interfaces are unconfigured ( but present )

what is the ouput of ip link in your container ?
can you also post the config file of the container ?

 

[alex3137]

Thanks Manu for your reply

Container config file:

arch: amd64
cpulimit: 1
cpuunits: 1024
hostname: deluge.besnard.me
memory: 2048
net0: name=eth0,bridge=vmbr1,gw=10.10.10.254,hwaddr=12:AD:9D:53:195,ip=10.10.10.110/24,type=veth
onboot: 1
ostype: debian
rootfs: local:110/vm-110-disk-1.subvol,size=0T
swap: 2048
lxc.aa_profile: lxc-container-default-with-nfs

ip link:

root@deluge:~# ip link
1: lo: <LOOPBACK> mtu 65536 qdisc noop state DOWN mode DEFAULT group default qlen 1
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
33: eth0@if34: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 1000
link/ether 12:ad:9d:53:19:d5 brd ff:ff:ff:ff:ff:ff

And actually I have noted a very low memory usage and I don't think the boot process completed:

root@deluge:~# ps aux
USER PID %CPU %MEM VSZ RSS TTY STAT START TIME COMMAND
root 1 0.0 0.0 27080 2036 ? Ss 04:17 0:00 /sbin/init
root 32 0.0 0.1 21876 3344 ? Ss 12:05 0:00 /bin/bash
root 34 0.0 0.1 19072 2132 ? R+ 12:05 0:00 ps aux

This is supposed to be a Deluge server but the process wasn't started at boot (it is supposed to and worked fine so far).

I have the same problem with another container but all the others are fine. Newly deployed containers work fine too.

 

[fabian]

quoting wbumiller's reply in https://forum.proxmox.com/threads/no-network-when-using-custom-apparmor-profile.28879/

If you're on lxc>=2.0.3 make sure you include the cgns parts in your profile (from the
`lxc-container-default-cgns` profile instead of `lxc-container-default`.)
Other than that... try debugging the container from the inside, maybe start with `# lxc-start -F -n $vmid` to see its sysinit output.

 

[alex3137]

Hello

So I modified my lxc-container-default-with-nfs profile to add this line: "mount fstype=cgroup -> /sys/fs/cgroup/**,"

which now looks like:

# Do not load this file. Rather, load /etc/apparmor.d/lxc-containers, which
# will source all profiles under /etc/apparmor.d/lxc

profile lxc-container-default-with-nfs flags=(attach_disconnected,mediate_deleted) {
#include <abstractions/lxc/container-base>

# allow NFS (nfs/nfs4) mounts.
mount fstype=nfs*,
mount fstype=cgroup -> /sys/fs/cgroup/**,
}

And it works But I have no idea what I am doing...