I have been setting up a fresh new LXD instance on another computer of mine and skipping all parts of the LXC introduction. After several reboots, the usual test-ubuntu container does not have a usable IP address:
>>> lxc info ubuntu
Name: ubuntu
Status: Running
Init: 3615
Ips:
lo: IPV4 127.0.0.1
lo: IPV6 ::1
What am I missing?
srkunze
All 24 comments
what does "ifconfig lxcbr0" show you?
stgraber
on 11 Nov 2015
>>> ifconfig lxcbr0
lxcbr0 Link encap:Ethernet HWaddr fe:91:e5:d9:55:49
inet addr:10.0.3.1 Bcast:0.0.0.0 Mask:255.255.255.0
inet6 addr: fe80::945e:afff:fe82:5312/64 Scope:Link
UP BROADCAST MULTICAST MTU:1500 Metric:1
RX packets:0 errors:0 dropped:0 overruns:0 frame:0
TX packets:9 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:0 (0.0 B) TX bytes:1371 (1.3 KB)
Host OS: Fedora 23
LXD version: 0.21
>>> ifconfig lxcbr0
lxcbr0: flags=4163<UP,BROADCAST,RUNNING,MULTICAST> mtu 1500
inet 10.0.3.1 netmask 255.255.255.0 broadcast 0.0.0.0
inet6 fe80::74ba:c2ff:fe22:9423 prefixlen 64 scopeid 0x20<link>
ether fe:01:38:68:4f:5c txqueuelen 0 (Ethernet)
RX packets 463 bytes 40913 (39.9 KiB)
RX errors 0 dropped 0 overruns 0 frame 0
TX packets 431 bytes 51670 (50.4 KiB)
TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0
I'm also facing similar issue while testing for all images (amd64 only) as only for certain distros can obtain the ip address. Most recent distros can't seems to obtain ip while older distros can, see result below.
>>> lxc list
+---------------------+---------+------------+------+-----------+-----------+
| NAME | STATE | IPV4 | IPV6 | EPHEMERAL | SNAPSHOTS |
+---------------------+---------+------------+------+-----------+-----------+
| test-busybox | RUNNING | | | NO | 0 |
| test-centos-6 | RUNNING | 10.0.3.161 | | NO | 0 |
| test-centos-7 | RUNNING | | | NO | 0 |
| test-debian-jessie | RUNNING | | | NO | 0 |
| test-debian-sid | RUNNING | | | NO | 0 |
| test-debian-squeeze | RUNNING | 10.0.3.129 | | NO | 0 |
| test-debian-wheezy | RUNNING | 10.0.3.70 | | NO | 0 |
| test-gentoo-current | RUNNING | | | NO | 0 |
| test-oracle-65 | RUNNING | 10.0.3.237 | | NO | 0 |
| test-plamo-5x | RUNNING | 10.0.3.72 | | NO | 0 |
| test-ubuntu-cloud | RUNNING | 10.0.3.143 | | NO | 0 |
| test-ubuntu-precise | RUNNING | 10.0.3.42 | | NO | 0 |
| test-ubuntu-trusty | RUNNING | 10.0.3.111 | | NO | 0 |
| test-ubuntu-vivid | RUNNING | | | NO | 0 |
| test-ubuntu-wily | RUNNING | | | NO | 0 |
| test-ubuntu-xenial | RUNNING | | | NO | 0 |
+---------------------+---------+------------+------+-----------+-----------+
Rebuilding the containers yields similar result.
| NAME | STATE | IPV4 | IPV6 | EPHEMERAL | SNAPSHOTS |
+---------------------+---------+------------+------+-----------+-----------+
| test-busybox | RUNNING | | | NO | 0 |
| test-centos-6 | RUNNING | 10.0.3.153 | | NO | 0 |
| test-centos-7 | RUNNING | | | NO | 0 |
| test-debian-jessie | RUNNING | | | NO | 0 |
| test-debian-sid | RUNNING | | | NO | 0 |
| test-debian-squeeze | RUNNING | 10.0.3.158 | | NO | 0 |
| test-debian-wheezy | RUNNING | 10.0.3.236 | | NO | 0 |
| test-gentoo-current | RUNNING | | | NO | 0 |
| test-oracle-65 | RUNNING | 10.0.3.51 | | NO | 0 |
| test-plamo-5x | RUNNING | 10.0.3.5 | | NO | 0 |
| test-ubuntu-cloud | RUNNING | 10.0.3.232 | | NO | 0 |
| test-ubuntu-precise | RUNNING | 10.0.3.244 | | NO | 0 |
| test-ubuntu-trusty | RUNNING | 10.0.3.214 | | NO | 0 |
| test-ubuntu-vivid | RUNNING | | | NO | 0 |
| test-ubuntu-wily | RUNNING | | | NO | 0 |
| test-ubuntu-xenial | RUNNING | | | NO | 0 |
+---------------------+---------+------------+------+-----------+-----------+
kianmeng
on 12 Nov 2015
Is dnsmasq running on the lxcbr0 tot provide dhcp to the containers?
jsimonetti
on 12 Nov 2015
@jsimonetti How can I look this up?
srkunze
on 13 Nov 2015
You can see if dnsmasq is running with ps ax | grep dnsmasq
This should return the dnsmasq process.
Which distribution are you using?
jsimonetti
on 13 Nov 2015
Ubuntu 14.04
>>> ps ax | grep dnsmasq
1248 ? S 0:00 dnsmasq -u lxc-dnsmasq --strict-order --bind-interfaces --pid-file=/run/lxc/dnsmasq.pid --listen-address 10.0.3.1 --dhcp-range 10.0.3.2,10.0.3.254 --dhcp-lease-max=253 --dhcp-no-override --except-interface=lo --interface=lxcbr0 --dhcp-leasefile=/var/lib/misc/dnsmasq.lxcbr0.leases --dhcp-authoritative
2227 ? S 0:00 /usr/sbin/dnsmasq --conf-file=/var/lib/libvirt/dnsmasq/default.conf
2302 ? S 0:00 /usr/sbin/dnsmasq --no-resolv --keep-in-foreground --no-hosts --bind-interfaces --pid-file=/run/sendsigs.omit.d/network-manager.dnsmasq.pid --listen-address=127.0.1.1 --conf-file=/var/run/NetworkManager/dnsmasq.conf --cache-size=0 --proxy-dnssec --enable-dbus=org.freedesktop.NetworkManager.dnsmasq --conf-dir=/etc/NetworkManager/dnsmasq.d
24295 pts/15 S+ 0:00 grep --color=auto dnsmasq
I've managed to obtain the IP address of some of the container by entering the container the requesting an IP address manually. I suspect certain container template was not created correctly (certain services were not started)? But I may be wrong.
Note that I can't get the container to shut down properly (another issue I'm still troubleshooting), so I can't test whether the changes will be permanent on next container reboot.
The container without assigned IP.
>>> lxc info test-ubuntu-wily
Name: test-ubuntu-wily
Status: Running
Init: 18545
Ips:
lo: IPV4 127.0.0.1
lo: IPV6 ::1
Enter our container through Bash shell.
>>> lxc exec test-ubuntu-wily /bin/bash
root@test-ubuntu-wily:~#
Get our eth0 interface up.
root@test-ubuntu-wily:~# ifconfig eth0 up
root@test-ubuntu-wily:~# ifconfig
eth0 Link encap:Ethernet HWaddr 00:16:3e:af:0e:d6
inet6 addr: fe80::216:3eff:feaf:ed6/64 Scope:Link
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:6 errors:0 dropped:0 overruns:0 frame:0
TX packets:6 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:508 (508.0 B) TX bytes:508 (508.0 B)
lo Link encap:Local Loopback
inet addr:127.0.0.1 Mask:255.0.0.0
inet6 addr: ::1/128 Scope:Host
UP LOOPBACK RUNNING MTU:65536 Metric:1
RX packets:0 errors:0 dropped:0 overruns:0 frame:0
TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:0 (0.0 B) TX bytes:0 (0.0 B)
Update your name server details.
root@test-ubuntu-wily:~# resolvconf -u
resolvconf: Error: /run/resolvconf/interface either does not exist or is not a directory
Create the missing folder and try again.
root@test-ubuntu-wily:~# mkdir -p /run/resolvconf/interface
root@test-ubuntu-wily:~# resolvconf -u
Release and renew the IP address.
root@test-ubuntu-wily:~# dhclient -r
Killed old client process
root@test-ubuntu-wily:~# dhclient
root@test-ubuntu-wily:~#
Check your eth0 interface again.
root@test-ubuntu-wily:~# ifconfig eth0
eth0 Link encap:Ethernet HWaddr 00:16:3e:af:0e:d6
inet addr:10.0.3.217 Bcast:10.0.3.255 Mask:255.255.255.0
inet6 addr: fe80::216:3eff:feaf:ed6/64 Scope:Link
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:30 errors:0 dropped:0 overruns:0 frame:0
TX packets:25 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:4652 (4.6 KB) TX bytes:5262 (5.2 KB)
However, you still can't connect to the Internet from within the container. You'll need to create the /etc/resolv.conf file. You will need to reconfigure resolvconf to symbolic link from /etc/resolv.conf to /run/resolvconf/resolv.conf.
root@test-ubuntu-wily:~#
root@test-ubuntu-wily:~# dpkg-reconfigure resolvconf
root@test-ubuntu-wily:~# ll /etc/resolv.conf
lrwxrwxrwx. 1 root root 29 Nov 13 16:59 /etc/resolv.conf -> ../run/resolvconf/resolv.conf
Test your Internet connection.
root@test-ubuntu-wily:~# ping linuxcontainers.org
PING linuxcontainers.org (192.99.34.219) 56(84) bytes of data.
64 bytes from vorash.stgraber.org (192.99.34.219): icmp_seq=1 ttl=55 time=274 ms
64 bytes from vorash.stgraber.org (192.99.34.219): icmp_seq=2 ttl=55 time=282 ms
......
Exit and check the container info again.
root@test-ubuntu-wily:~# exit
exit
>>> lxc info test-ubuntu-wily
Name: test-ubuntu-wily
Status: Running
Init: 18545
Ips:
eth0: IPV4 10.0.3.217
lo: IPV4 127.0.0.1
lo: IPV6 ::1
@kianmeng Interesting.
Is there something which might have caused this behavior?
srkunze
on 13 Nov 2015
@srkunze Most likely the network interface (eth0) within the container failed to boot properly. Without eth0, surely the container cannot get an IP through DHCP. Why? No idea.
[root@test-centos-7 ~]# systemctl status network
Failed to get D-Bus connection: Failed to connect to socket /run/systemd/private:
Connection refused
Somewhat strange happened: I put the host into a proxy-free network. (cf. https://github.com/lxc/lxd/issues/1309#issuecomment-156171447)
That situation not only gave apt the ability to verify the lxd/lxc-stable repository but also gave LXD the ability to assign an IP address to the container.
@kianmeng @stgraber Does this make any sense?
srkunze
on 17 Nov 2015
@srkunze
I'm not sure about your environment and setup, but try troubleshoot your issue using Wireshark.
- Install and start Wireshark.
- Listen to the bridge interface lxcbr0 and start any container.
- Check the DHCP network traffic in Wireshark between the container and lxcbr0.
- Repeat this for for proxy and proxy-free network.
- Compare the results.
Maybe this may give you some hints.
kianmeng
on 18 Nov 2015
Couldn't reproduce for now. Closing.
srkunze
on 11 Dec 2015
@kianmeng What would be the procedure for centos 6/7 container, if it doesn't getting ip address. Some commands are not found in centos such as dhclient... ?
redhat57
on 3 Mar 2017
@redhat57
Unfortunately I can't reproduce your issue. Best if you open a new ticket and provides us with more details on how to reproduce your errors.
Host details.
$ lsb_release -a
No LSB modules are available.
Distributor ID: Ubuntu
Description: Ubuntu 16.10
Release: 16.10
Codename: yakkety
$ lxc version
2.9.1
For me, CentOS 6 images seems to find IP address and dhclient.
$ lxc launch images:centos/6/amd64 my-centos
$ lxc exec my-centos bash
[root@my-centos ~]# ip addr show eth0 | grep -w inet
inet 10.20.35.204/24 brd 10.20.35.255 scope global eth0
[root@my-centos ~]# dhclient --version
isc-dhclient-4.1.1-P1
i have the exact same issue on
driver: lxc
driver_version: 2.0.7
kernel: Linux
kernel_architecture: x86_64
kernel_version: 4.8.5-gentoo
server: lxd
server_pid: 5173
server_version: "2.11"
storage: dir
storage_version: "1"
problem occured after recompiling lxc without cgmanager support. it worked with cgmanager.
older clients work, newer debian or ubuntu dont get an ip. seems like it never triggered dhclient. can get it working by manually entering dhclient eth0. still cant shutdown machine. only occurs on newer images, older images like debian wheezy work well.
EDIT:
Reason: Host is not compatible with systemd, but client depends on it.
Solution:
Compile Kernel >4.5 with systemd compatibility
General setup --->
[*] open by fhandle syscalls
[*] Control Group support --->
[ ] Enable deprecated sysfs features to support old userspace tools
[*] Configure standard kernel features (expert users) --->
[*] Enable eventpoll support
[*] Enable signalfd() system call
[*] Enable timerfd() system call
[*] Networking support --->
Device Drivers --->
Generic Driver Options --->
[*] Maintain a devtmpfs filesystem to mount at /dev
File systems --->
[*] Inotify support for userspace
Pseudo filesystems --->
[*] /proc file system support
[*] sysfs file system support
then add the following entry into /etc/fstab:
cgroup /sys/fs/cgroup/systemd cgroup none,name=systemd 0 0
wmchris
on 22 May 2017
I had to add it there manually for it to work in Ubuntu 16.04 LTS. This is how my config looks like
# Common configuration
lxc.include = /usr/share/lxc/config/ubuntu.common.conf
# Container specific configuration
lxc.rootfs = /var/lib/lxc/c1/rootfs
lxc.mount = /var/lib/lxc/c1/fstab
lxc.utsname = c1
lxc.arch = amd64
# Network configuration
lxc.network.type = veth
lxc.network.flags = up
lxc.network.link = lxcbr0
lxc.network.hwaddr = 00:16:3e:3a:36:2f
#my configs
lxc.network.ipv4 = 10.0.3.100/24 10.0.3.255
lxc.network.ipv4.gateway = auto
lxc.start.auto = 1
lxc.start.delay = 15
lxc.start.order = 10
oquidave
on 20 Jun 2017
Make sure the container's network cloud-init inet setting is set to dhcp, not manual,
especially if container was created before initial bridge setup (sudo lxd init),
hopefully this will save someone a few hours troubleshooting e.g.:
lxc exec ubuntu -- grep eth0 /etc/network/interfaces.d/50-cloud-init.cfg
auto eth0
iface eth0 inet dhcp
If updating:
sudo systemctl restart lxd-bridge
lxc restart ubuntu
For more information:
man interfaces
Container Not Getting IP Address #1298
raymov
on 11 Aug 2017
Had the same issue, @raymov you made my day!
aknobloch
on 4 Feb 2018
solution based on @raymov 's comment:
run
lxc profile edit default
and change user.network_mode from link-local to dhcp. maybe you have to run systemctl restart lxd-bridge afterwards, can't remember.
thx @raymov !!!
sgeisbacher
on 9 Mar 2018
Throwing my 2 cents in here, before getting any of this, I highly recommend moving to the feature branch of lxd/lxc (version 3.0+)
This gives you access to: lxc network commands.
Reason for this, I installed it, and accidentally placed the dhcp pool outside of my subset that I assigned to the brdige (idiot me, check). Now lxc never complained about any of this, but it took me half an hour to figure out this dumb mistake.
Nowadays with version 2.3+ you can simply run: lxc network show lxdbr0
And verify you have configured it right.
mathieudevos
on 22 Aug 2018
I did lxc network show lxdbr0 and got:
config: {}
description: Default LXD profile
devices:
root:
path: /
pool: default
type: disk
name: default
used_by:
- /1.0/containers/ubuntu
Then I updated the config to:
config: {}
description: Default LXD profile
devices:
eth0:
name: eth0
nictype: bridged
parent: lxdbr0
type: nic
root:
path: /
pool: default
type: disk
name: default
used_by:
- /1.0/containers/ubuntu
Now my container has ipv4 ip
amadu80
on 17 Jan 2019
Coming from macOS, especially if you have just connected Apple Ethernet dongle on Ubuntu 16.04 (LXC 2.x for the sake of practice). Even fixing and restart the network couldn't get the IPv4 appeared in the lxc list, what I did to restart my MacBook and login to Ubuntu, re-run lxc list and finally shows 192.168.x.x successfully.
Other info, you can use ens# (Apple Ethernet dongle shown) instead of eth0, which is exactly the same network.
proyb6
on 30 Mar 2019
I've managed to obtain the IP address of some of the container by entering the container the requesting an IP address manually. I suspect certain container template was not created correctly (certain services were not started)? But I may be wrong.
Note that I can't get the container to shut down properly (another issue I'm still troubleshooting), so I can't test whether the changes will be permanent on next container reboot.
The container without assigned IP.
>>> lxc info test-ubuntu-wily Name: test-ubuntu-wily Status: Running Init: 18545 Ips: lo: IPV4 127.0.0.1 lo: IPV6 ::1Enter our container through Bash shell.
>>> lxc exec test-ubuntu-wily /bin/bash root@test-ubuntu-wily:~#Get our eth0 interface up.
root@test-ubuntu-wily:~# ifconfig eth0 up root@test-ubuntu-wily:~# ifconfig eth0 Link encap:Ethernet HWaddr 00:16:3e:af:0e:d6 inet6 addr: fe80::216:3eff:feaf:ed6/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:6 errors:0 dropped:0 overruns:0 frame:0 TX packets:6 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:508 (508.0 B) TX bytes:508 (508.0 B) lo Link encap:Local Loopback inet addr:127.0.0.1 Mask:255.0.0.0 inet6 addr: ::1/128 Scope:Host UP LOOPBACK RUNNING MTU:65536 Metric:1 RX packets:0 errors:0 dropped:0 overruns:0 frame:0 TX packets:0 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:0 (0.0 B) TX bytes:0 (0.0 B)Update your name server details.
root@test-ubuntu-wily:~# resolvconf -u resolvconf: Error: /run/resolvconf/interface either does not exist or is not a directoryCreate the missing folder and try again.
root@test-ubuntu-wily:~# mkdir -p /run/resolvconf/interface root@test-ubuntu-wily:~# resolvconf -uRelease and renew the IP address.
root@test-ubuntu-wily:~# dhclient -r Killed old client process root@test-ubuntu-wily:~# dhclient root@test-ubuntu-wily:~#Check your eth0 interface again.
root@test-ubuntu-wily:~# ifconfig eth0 eth0 Link encap:Ethernet HWaddr 00:16:3e:af:0e:d6 inet addr:10.0.3.217 Bcast:10.0.3.255 Mask:255.255.255.0 inet6 addr: fe80::216:3eff:feaf:ed6/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:30 errors:0 dropped:0 overruns:0 frame:0 TX packets:25 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:4652 (4.6 KB) TX bytes:5262 (5.2 KB)However, you still can't connect to the Internet from within the container. You'll need to create the /etc/resolv.conf file. You will need to reconfigure resolvconf to symbolic link from /etc/resolv.conf to /run/resolvconf/resolv.conf.
root@test-ubuntu-wily:~# root@test-ubuntu-wily:~# dpkg-reconfigure resolvconf root@test-ubuntu-wily:~# ll /etc/resolv.conf lrwxrwxrwx. 1 root root 29 Nov 13 16:59 /etc/resolv.conf -> ../run/resolvconf/resolv.confTest your Internet connection.
root@test-ubuntu-wily:~# ping linuxcontainers.org PING linuxcontainers.org (192.99.34.219) 56(84) bytes of data. 64 bytes from vorash.stgraber.org (192.99.34.219): icmp_seq=1 ttl=55 time=274 ms 64 bytes from vorash.stgraber.org (192.99.34.219): icmp_seq=2 ttl=55 time=282 ms ......Exit and check the container info again.
root@test-ubuntu-wily:~# exit exit >>> lxc info test-ubuntu-wily Name: test-ubuntu-wily Status: Running Init: 18545 Ips: eth0: IPV4 10.0.3.217 lo: IPV4 127.0.0.1 lo: IPV6 ::1
every i turn of lxc container, restat container i have type command to have internet.
hoatienii
on 16 Aug 2019
Related issues
mlaradji
路
4Comments
spacekookie
路
3Comments
rrva
路
5Comments
tebanep
路
5Comments
simos
路
3Comments
Most helpful comment
Make sure the container's network cloud-init inet setting is set to dhcp, not manual,
especially if container was created before initial bridge setup (sudo lxd init),
hopefully this will save someone a few hours troubleshooting e.g.:
If updating:
For more information:
man interfacesContainer Not Getting IP Address #1298