No IP address with "default" profile

LXD
1

Hi,
I am trying to learn how to use LXD.

I am using a Debian “bullseye” host and I need to run a Ubuntu/Xenial container.

I created it with plain:

mcon@Lenovo:~$ sudo /snap/bin/lxc launch images:ubuntu/xenial macinino
Creating macinino
Starting macinino
mcon@Lenovo:~$ sudo /snap/bin/lxc list
+----------+---------+------+------+-----------+-----------+
|   NAME   |  STATE  | IPV4 | IPV6 |   TYPE    | SNAPSHOTS |
+----------+---------+------+------+-----------+-----------+
| macinino | RUNNING |      |      | CONTAINER | 0         |
+----------+---------+------+------+-----------+-----------+
mcon@Lenovo:~$ sudo /snap/bin/lxc info --show-log macinino
Name: macinino
Status: RUNNING
Type: container
Architecture: x86_64
PID: 10223
Created: 2021/12/14 00:04 CET
Last Used: 2021/12/14 00:04 CET

Resources:
  Processes: 1
  CPU usage:
    CPU usage (in seconds): 0
  Memory usage:
    Memory (current): 6.86MiB
  Network usage:
    lo:
      Type: loopback
      State: UP
      MTU: 65536
      Bytes received: 0B
      Bytes sent: 0B
      Packets received: 0
      Packets sent: 0
      IP addresses:
        inet:  127.0.0.1/8 (local)
        inet6: ::1/128 (local)
    eth0:
      Type: broadcast
      State: UP
      Host interface: veth92bede54
      MAC address: 00:16:3e:eb:c4:f8
      MTU: 1500
      Bytes received: 2.02kB
      Bytes sent: 726B
      Packets received: 44
      Packets sent: 9
      IP addresses:
        inet6: fe80::216:3eff:feeb:c4f8/64 (link)

Log:

lxc macinino 20211213230410.128 WARN     conf - conf.c:lxc_map_ids:3579 - newuidmap binary is missing
lxc macinino 20211213230410.128 WARN     conf - conf.c:lxc_map_ids:3585 - newgidmap binary is missing
lxc macinino 20211213230410.129 WARN     conf - conf.c:lxc_map_ids:3579 - newuidmap binary is missing
lxc macinino 20211213230410.129 WARN     conf - conf.c:lxc_map_ids:3585 - newgidmap binary is missing

mcon@Lenovo:~$

My problem is I have no IP address in my container, as you can see.
Most likely I’m missing some configuration, but I’m unable to divine what.
Can someone help, please?

2

You need to post your LXD network configuration in here:

lxc network list

Is it a bridge? Is it a FAN?

3

Usual suspects are something interfering with network on your system, most common firewalld (if running on Fedora) or having Docker installed on your system.

4

Pronto!

mcon@Lenovo:~/Downloads$ sudo /snap/bin/lxc network list
+--------+----------+---------+-----------------+------+-------------+---------+
|  NAME  |   TYPE   | MANAGED |      IPV4       | IPV6 | DESCRIPTION | USED BY |
+--------+----------+---------+-----------------+------+-------------+---------+
| enp9s0 | physical | NO      |                 |      |             | 0       |
+--------+----------+---------+-----------------+------+-------------+---------+
| lxcbr0 | bridge   | NO      |                 |      |             | 0       |
+--------+----------+---------+-----------------+------+-------------+---------+
| lxdbr0 | bridge   | YES     | 10.100.177.1/24 | none |             | 3       |
+--------+----------+---------+-----------------+------+-------------+---------+
| wlp8s0 | physical | NO      |                 |      |             | 0       |
+--------+----------+---------+-----------------+------+-------------+---------+
5

Thanks stgraber,
My host is a very recent and completely up-to-date Debian bullseye (not Fedora).
I do not have docker installed, but I have ZeroTier-One if it matters.

I do not think I have any strange firewalling active (I did not explicitly install any, that’ for sure).

My full net config is:

mcon@Lenovo:~/Downloads$ ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default 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: enp9s0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc pfifo_fast state DOWN group default qlen 1000
    link/ether f0:76:1c:57:93:b1 brd ff:ff:ff:ff:ff:ff
3: wlp8s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
    link/ether d0:7e:35:48:8c:21 brd ff:ff:ff:ff:ff:ff
    inet 10.73.246.55/21 brd 10.73.247.255 scope global dynamic noprefixroute wlp8s0
       valid_lft 1451sec preferred_lft 1451sec
    inet6 fe80::38aa:daea:f776:66d8/64 scope link noprefixroute 
       valid_lft forever preferred_lft forever
4: lxdbr0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
    link/ether 00:16:3e:b9:85:f0 brd ff:ff:ff:ff:ff:ff
    inet 10.100.177.1/24 scope global lxdbr0
       valid_lft forever preferred_lft forever
    inet6 fe80::216:3eff:feb9:85f0/64 scope link 
       valid_lft forever preferred_lft forever
7: lxcbr0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc noqueue state DOWN group default qlen 1000
    link/ether 00:16:3e:00:00:00 brd ff:ff:ff:ff:ff:ff
    inet 10.0.3.1/24 brd 10.0.3.255 scope global lxcbr0
       valid_lft forever preferred_lft forever
8: ztyqbt4opw: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 2800 qdisc pfifo_fast state UNKNOWN group default qlen 1000
    link/ether 22:bd:97:56:25:80 brd ff:ff:ff:ff:ff:ff
    inet 172.28.157.25/16 brd 172.28.255.255 scope global ztyqbt4opw
       valid_lft forever preferred_lft forever
    inet6 fc5d:f964:6fcb:bb80:6a92::1/40 scope global 
       valid_lft forever preferred_lft forever
    inet6 fe80::20bd:97ff:fe56:2580/64 scope link 
       valid_lft forever preferred_lft forever
12: veth92bede54@if11: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue master lxdbr0 state UP group default qlen 1000
    link/ether 72:aa:fe:02:26:97 brd ff:ff:ff:ff:ff:ff link-netnsid 0

Any hint about what to check/try?
TiA!

6

Just another data point: I have lxc installed from the standard Debian repos (not from snap).
Can that conflict? Should I purge it or is it somehow needed?

7

FYI: purging lxc via sudo apt purge apt did not change situation at all.

8

Are you using ufw firewall?

9

Sorry for the late comeback (I was on a business trip).
No, ufw is not even installed on laptop.

10

I had this same problem after running lxd recover

I finally did a system update and rebooted and it seemed to fix the problem.

11

I managed to “somehow” overcome the problem, but in a highly unsatisfactory way:

  • completely remove all lxd snap
  • compile from sources (including hand-installing a recent version of go)
  • incur in the same problem discussed here
  • Apply solution (boot with systemd.unified_cgroup_hierarchy=0)
  • add root:100000:65536 to /etc/subuid and /etc/subgid (apparently debian “standard install” adds that only for “normal users”)

This actually works (including networking), but (IMHO, of course) is an indication something is “not aligned” in CGroup2 handling.
I know too little about the whole matter to venture in a guess about root cause, but I strongly believe it should be investigated, perhaps on both sides.

I am willing to help in testing, if deemed useful.
Please advise.

12

Its likely the containers operating system not likely a pure cgroup2 environment and needing a hybrid environment.

13

Thank Thomas.
As stated laptop where I’m testing is a pure Debian 11 “bullseye” (stable), not exactly an “unheard of” distribution.
I would like to get to the bottom of this, if at all possible, before filing a bug report to debian, lxd or both.
I am aware they are actively working in the issue but I would like to have a “stable” install without waiting for them to finish their work.

How sh0uld I proceed?

14

booting with systemd.unified_cgroup_hierarchy=0 is the solution to running older container oses on newer systems that use pure cgroupv2 by default.

15

Thanks Thomas,
but I wasn’t clear enough, evidently.

I am not trying to run old containers.
I currently have no containers.
The errors I see are either on freshly made containers or at lxd daemon start (before lxd init).

I am unsure if transition to “pure CGroupv2” on Debian bullseye is really complete or if problem is somewhere else.

16

This is where I thought you were trying to run an old container os that doesn’t support cgroupv2.

17

My bad.
I actually have a build environment needing Ubuntu/xenial (don’t ask me why!).
I wanted to use a container to avoid tying a machine to that ancient piece of software while avoiding the inefficiencies of a full Virtual Machine (using all cores/threads in my hardware is useful).
There I started to get problems.
Thanks for your help.
I suppose converting host to a recent Ubuntu might be the solution, if I insist on using lxd.

18

LXD supports VMS using the ‘–vm’ instance creation flag if you decided to go that route.